67936528b5
Model: fpadovani/eng-latn-10mb-after-ppt-shuff-dyck-disjoint-tok-ckpt500_seed3407 Source: Original Platform
9134 lines
249 KiB
JSON
9134 lines
249 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 4.865332612222823,
|
|
"eval_steps": 500,
|
|
"global_step": 4500,
|
|
"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
|
|
},
|
|
{
|
|
"entropy": 6.240487289428711,
|
|
"epoch": 1.6273661438615468,
|
|
"grad_norm": 0.7734375,
|
|
"learning_rate": 0.0004958588382973809,
|
|
"loss": 5.9793,
|
|
"mean_token_accuracy": 0.1554610624909401,
|
|
"num_tokens": 3382648.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 6.16507658958435,
|
|
"epoch": 1.6327744726879394,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004957765241182826,
|
|
"loss": 6.0378,
|
|
"mean_token_accuracy": 0.15304033160209657,
|
|
"num_tokens": 3393871.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 6.113471508026123,
|
|
"epoch": 1.638182801514332,
|
|
"grad_norm": 0.80078125,
|
|
"learning_rate": 0.0004956934076367854,
|
|
"loss": 5.9194,
|
|
"mean_token_accuracy": 0.15826143324375153,
|
|
"num_tokens": 3405148.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 6.1743358135223385,
|
|
"epoch": 1.6435911303407247,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.000495609489154935,
|
|
"loss": 5.9637,
|
|
"mean_token_accuracy": 0.16070945411920548,
|
|
"num_tokens": 3416647.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 6.237698650360107,
|
|
"epoch": 1.6489994591671173,
|
|
"grad_norm": 0.8203125,
|
|
"learning_rate": 0.0004955247689776911,
|
|
"loss": 6.0604,
|
|
"mean_token_accuracy": 0.1475292205810547,
|
|
"num_tokens": 3428331.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 6.06461238861084,
|
|
"epoch": 1.65440778799351,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.000495439247412927,
|
|
"loss": 5.9116,
|
|
"mean_token_accuracy": 0.15969590619206428,
|
|
"num_tokens": 3440683.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 6.163879871368408,
|
|
"epoch": 1.6598161168199026,
|
|
"grad_norm": 0.765625,
|
|
"learning_rate": 0.0004953529247714285,
|
|
"loss": 5.9067,
|
|
"mean_token_accuracy": 0.15767832696437836,
|
|
"num_tokens": 3452375.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 6.1659508228302,
|
|
"epoch": 1.6652244456462952,
|
|
"grad_norm": 0.78125,
|
|
"learning_rate": 0.0004952658013668922,
|
|
"loss": 6.0629,
|
|
"mean_token_accuracy": 0.14571874141693114,
|
|
"num_tokens": 3463255.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 6.228959417343139,
|
|
"epoch": 1.670632774472688,
|
|
"grad_norm": 0.8203125,
|
|
"learning_rate": 0.0004951778775159246,
|
|
"loss": 6.0125,
|
|
"mean_token_accuracy": 0.15071145743131636,
|
|
"num_tokens": 3473642.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 6.167908191680908,
|
|
"epoch": 1.6760411032990805,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004950891535380414,
|
|
"loss": 5.9723,
|
|
"mean_token_accuracy": 0.15263262540102004,
|
|
"num_tokens": 3484877.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 6.128119039535522,
|
|
"epoch": 1.6814494321254734,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004949996297556656,
|
|
"loss": 5.8839,
|
|
"mean_token_accuracy": 0.16038721650838852,
|
|
"num_tokens": 3495500.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 6.181710481643677,
|
|
"epoch": 1.6868577609518658,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004949093064941268,
|
|
"loss": 6.0093,
|
|
"mean_token_accuracy": 0.1580350860953331,
|
|
"num_tokens": 3508057.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 6.178483009338379,
|
|
"epoch": 1.6922660897782587,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.0004948181840816601,
|
|
"loss": 6.1681,
|
|
"mean_token_accuracy": 0.1513574853539467,
|
|
"num_tokens": 3521110.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 6.245854187011719,
|
|
"epoch": 1.697674418604651,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004947262628494045,
|
|
"loss": 6.0903,
|
|
"mean_token_accuracy": 0.1530960515141487,
|
|
"num_tokens": 3530692.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 6.209924554824829,
|
|
"epoch": 1.703082747431044,
|
|
"grad_norm": 0.80859375,
|
|
"learning_rate": 0.0004946335431314018,
|
|
"loss": 6.0352,
|
|
"mean_token_accuracy": 0.1460341766476631,
|
|
"num_tokens": 3542195.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 6.230369758605957,
|
|
"epoch": 1.7084910762574363,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004945400252645959,
|
|
"loss": 5.935,
|
|
"mean_token_accuracy": 0.15976211726665496,
|
|
"num_tokens": 3552676.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 6.089649963378906,
|
|
"epoch": 1.7138994050838292,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004944457095888308,
|
|
"loss": 5.8993,
|
|
"mean_token_accuracy": 0.155808487534523,
|
|
"num_tokens": 3563308.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 6.155344152450562,
|
|
"epoch": 1.7193077339102216,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00049435059644685,
|
|
"loss": 5.9277,
|
|
"mean_token_accuracy": 0.16010808050632477,
|
|
"num_tokens": 3574085.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 6.163233900070191,
|
|
"epoch": 1.7247160627366145,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004942546861842947,
|
|
"loss": 5.9989,
|
|
"mean_token_accuracy": 0.15085040628910065,
|
|
"num_tokens": 3585816.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 6.170480680465698,
|
|
"epoch": 1.730124391563007,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004941579791497032,
|
|
"loss": 6.0483,
|
|
"mean_token_accuracy": 0.15046859830617904,
|
|
"num_tokens": 3597776.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 6.253198623657227,
|
|
"epoch": 1.7355327203893998,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004940604756945091,
|
|
"loss": 6.0541,
|
|
"mean_token_accuracy": 0.1547902449965477,
|
|
"num_tokens": 3607991.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 6.138210487365723,
|
|
"epoch": 1.7409410492157922,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004939621761730399,
|
|
"loss": 5.9398,
|
|
"mean_token_accuracy": 0.16401894092559816,
|
|
"num_tokens": 3618204.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 6.10772967338562,
|
|
"epoch": 1.746349378042185,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004938630809425166,
|
|
"loss": 6.0458,
|
|
"mean_token_accuracy": 0.15313471108675003,
|
|
"num_tokens": 3629558.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 6.200343942642212,
|
|
"epoch": 1.7517577068685775,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 0.0004937631903630513,
|
|
"loss": 5.9972,
|
|
"mean_token_accuracy": 0.1569698140025139,
|
|
"num_tokens": 3640773.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 6.135195684432984,
|
|
"epoch": 1.7571660356949703,
|
|
"grad_norm": 0.70703125,
|
|
"learning_rate": 0.0004936625047976468,
|
|
"loss": 5.902,
|
|
"mean_token_accuracy": 0.1601807177066803,
|
|
"num_tokens": 3652186.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 6.117422199249267,
|
|
"epoch": 1.7625743645213627,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0004935610246121944,
|
|
"loss": 5.9641,
|
|
"mean_token_accuracy": 0.16303962767124175,
|
|
"num_tokens": 3663068.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 6.162671804428101,
|
|
"epoch": 1.7679826933477556,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004934587501754736,
|
|
"loss": 6.1034,
|
|
"mean_token_accuracy": 0.14989171624183656,
|
|
"num_tokens": 3673803.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 6.316601324081421,
|
|
"epoch": 1.7733910221741482,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004933556818591498,
|
|
"loss": 6.0672,
|
|
"mean_token_accuracy": 0.1504330776631832,
|
|
"num_tokens": 3685482.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 6.123309183120727,
|
|
"epoch": 1.7787993510005409,
|
|
"grad_norm": 0.76953125,
|
|
"learning_rate": 0.0004932518200377736,
|
|
"loss": 5.9284,
|
|
"mean_token_accuracy": 0.1638890653848648,
|
|
"num_tokens": 3697641.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 6.133664608001709,
|
|
"epoch": 1.7842076798269335,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004931471650887789,
|
|
"loss": 5.9427,
|
|
"mean_token_accuracy": 0.15842915028333665,
|
|
"num_tokens": 3709560.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 6.200141048431396,
|
|
"epoch": 1.7896160086533262,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004930417173924822,
|
|
"loss": 6.0434,
|
|
"mean_token_accuracy": 0.149042746424675,
|
|
"num_tokens": 3720422.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 6.0688094139099125,
|
|
"epoch": 1.7950243374797188,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.0004929354773320806,
|
|
"loss": 5.8772,
|
|
"mean_token_accuracy": 0.1585821643471718,
|
|
"num_tokens": 3731104.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 6.127728843688965,
|
|
"epoch": 1.8004326663061114,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0004928284452936506,
|
|
"loss": 6.0515,
|
|
"mean_token_accuracy": 0.14892184734344482,
|
|
"num_tokens": 3742766.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 6.21225938796997,
|
|
"epoch": 1.805840995132504,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.0004927206216661473,
|
|
"loss": 5.9295,
|
|
"mean_token_accuracy": 0.1595839112997055,
|
|
"num_tokens": 3753064.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 6.09962568283081,
|
|
"epoch": 1.8112493239588967,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004926120068414015,
|
|
"loss": 5.973,
|
|
"mean_token_accuracy": 0.1538739800453186,
|
|
"num_tokens": 3765183.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 6.164170360565185,
|
|
"epoch": 1.8166576527852893,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.0004925026012141201,
|
|
"loss": 5.9579,
|
|
"mean_token_accuracy": 0.1512836143374443,
|
|
"num_tokens": 3777902.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 6.080730342864991,
|
|
"epoch": 1.822065981611682,
|
|
"grad_norm": 0.8203125,
|
|
"learning_rate": 0.0004923924051818832,
|
|
"loss": 5.9259,
|
|
"mean_token_accuracy": 0.15897325575351715,
|
|
"num_tokens": 3788461.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 6.090081596374512,
|
|
"epoch": 1.8274743104380746,
|
|
"grad_norm": 0.80859375,
|
|
"learning_rate": 0.0004922814191451435,
|
|
"loss": 5.858,
|
|
"mean_token_accuracy": 0.1614278480410576,
|
|
"num_tokens": 3799574.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 6.144002056121826,
|
|
"epoch": 1.8328826392644673,
|
|
"grad_norm": 0.7890625,
|
|
"learning_rate": 0.0004921696435072245,
|
|
"loss": 6.0176,
|
|
"mean_token_accuracy": 0.15314162373542786,
|
|
"num_tokens": 3811889.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 6.159265851974487,
|
|
"epoch": 1.83829096809086,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.0004920570786743191,
|
|
"loss": 5.9568,
|
|
"mean_token_accuracy": 0.15708799809217452,
|
|
"num_tokens": 3822949.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 6.123097229003906,
|
|
"epoch": 1.8436992969172525,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004919437250554882,
|
|
"loss": 6.0601,
|
|
"mean_token_accuracy": 0.15241828560829163,
|
|
"num_tokens": 3834734.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 6.17847056388855,
|
|
"epoch": 1.8491076257436452,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004918295830626592,
|
|
"loss": 5.9247,
|
|
"mean_token_accuracy": 0.16275494247674943,
|
|
"num_tokens": 3845508.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 6.1110372066497805,
|
|
"epoch": 1.8545159545700378,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004917146531106242,
|
|
"loss": 5.9003,
|
|
"mean_token_accuracy": 0.1607997313141823,
|
|
"num_tokens": 3856500.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 6.182471418380738,
|
|
"epoch": 1.8599242833964305,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004915989356170389,
|
|
"loss": 5.9822,
|
|
"mean_token_accuracy": 0.1572587251663208,
|
|
"num_tokens": 3867877.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 6.160144996643067,
|
|
"epoch": 1.865332612222823,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004914824310024213,
|
|
"loss": 5.984,
|
|
"mean_token_accuracy": 0.16093191802501677,
|
|
"num_tokens": 3878613.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 6.139301109313965,
|
|
"epoch": 1.870740941049216,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004913651396901489,
|
|
"loss": 5.8982,
|
|
"mean_token_accuracy": 0.1604979693889618,
|
|
"num_tokens": 3888418.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 6.104005002975464,
|
|
"epoch": 1.8761492698756084,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.000491247062106459,
|
|
"loss": 5.9607,
|
|
"mean_token_accuracy": 0.15970670878887178,
|
|
"num_tokens": 3899359.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 6.161985731124878,
|
|
"epoch": 1.8815575987020012,
|
|
"grad_norm": 0.78515625,
|
|
"learning_rate": 0.0004911281986804455,
|
|
"loss": 6.0492,
|
|
"mean_token_accuracy": 0.155263289809227,
|
|
"num_tokens": 3911568.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 6.086746501922607,
|
|
"epoch": 1.8869659275283936,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004910085498440587,
|
|
"loss": 5.9885,
|
|
"mean_token_accuracy": 0.15208180546760558,
|
|
"num_tokens": 3921597.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 6.106437063217163,
|
|
"epoch": 1.8923742563547865,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004908881160321025,
|
|
"loss": 5.8945,
|
|
"mean_token_accuracy": 0.15527794659137725,
|
|
"num_tokens": 3933124.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 6.101495933532715,
|
|
"epoch": 1.897782585181179,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004907668976822338,
|
|
"loss": 5.9075,
|
|
"mean_token_accuracy": 0.1635837063193321,
|
|
"num_tokens": 3944388.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 6.124244594573975,
|
|
"epoch": 1.9031909140075718,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004906448952349603,
|
|
"loss": 5.8894,
|
|
"mean_token_accuracy": 0.1622192844748497,
|
|
"num_tokens": 3955024.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 5.979105043411255,
|
|
"epoch": 1.9085992428339642,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.0004905221091336392,
|
|
"loss": 5.8657,
|
|
"mean_token_accuracy": 0.16770282983779908,
|
|
"num_tokens": 3966029.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 6.045361518859863,
|
|
"epoch": 1.914007571660357,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004903985398244753,
|
|
"loss": 5.8343,
|
|
"mean_token_accuracy": 0.1611066535115242,
|
|
"num_tokens": 3976066.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 6.184390306472778,
|
|
"epoch": 1.9194159004867495,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004902741877565201,
|
|
"loss": 6.0466,
|
|
"mean_token_accuracy": 0.15331881493330002,
|
|
"num_tokens": 3986467.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 6.192496967315674,
|
|
"epoch": 1.9248242293131423,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.000490149053381669,
|
|
"loss": 6.0091,
|
|
"mean_token_accuracy": 0.1580958053469658,
|
|
"num_tokens": 3998208.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 6.161718368530273,
|
|
"epoch": 1.9302325581395348,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004900231371546608,
|
|
"loss": 5.9688,
|
|
"mean_token_accuracy": 0.15808582603931426,
|
|
"num_tokens": 4008202.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 6.006673288345337,
|
|
"epoch": 1.9356408869659276,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004898964395330753,
|
|
"loss": 5.7802,
|
|
"mean_token_accuracy": 0.16955413669347763,
|
|
"num_tokens": 4019256.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 6.0177160739898685,
|
|
"epoch": 1.94104921579232,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.0004897689609773318,
|
|
"loss": 5.8303,
|
|
"mean_token_accuracy": 0.1599978283047676,
|
|
"num_tokens": 4029738.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 6.115762710571289,
|
|
"epoch": 1.946457544618713,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0004896407019506879,
|
|
"loss": 6.0592,
|
|
"mean_token_accuracy": 0.14657395482063293,
|
|
"num_tokens": 4042277.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 6.15372142791748,
|
|
"epoch": 1.9518658734451053,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004895116629192371,
|
|
"loss": 5.9099,
|
|
"mean_token_accuracy": 0.16865991055965424,
|
|
"num_tokens": 4054019.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 6.047328758239746,
|
|
"epoch": 1.9572742022714982,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.0004893818443519074,
|
|
"loss": 5.9918,
|
|
"mean_token_accuracy": 0.15915630012750626,
|
|
"num_tokens": 4065562.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 6.187886524200439,
|
|
"epoch": 1.9626825310978906,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.00048925124672046,
|
|
"loss": 6.0174,
|
|
"mean_token_accuracy": 0.15806035250425338,
|
|
"num_tokens": 4076617.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 6.088625478744507,
|
|
"epoch": 1.9680908599242835,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0004891198704994868,
|
|
"loss": 5.8464,
|
|
"mean_token_accuracy": 0.16766615808010102,
|
|
"num_tokens": 4087191.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 6.122962379455567,
|
|
"epoch": 1.973499188750676,
|
|
"grad_norm": 0.79296875,
|
|
"learning_rate": 0.0004889877161664096,
|
|
"loss": 5.8828,
|
|
"mean_token_accuracy": 0.15968139767646788,
|
|
"num_tokens": 4099150.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 6.043911170959473,
|
|
"epoch": 1.9789075175770687,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004888547842014771,
|
|
"loss": 5.8843,
|
|
"mean_token_accuracy": 0.1640367776155472,
|
|
"num_tokens": 4110299.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 6.044490718841553,
|
|
"epoch": 1.9843158464034614,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004887210750877648,
|
|
"loss": 5.9045,
|
|
"mean_token_accuracy": 0.15955243557691573,
|
|
"num_tokens": 4121682.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 6.088275671005249,
|
|
"epoch": 1.989724175229854,
|
|
"grad_norm": 0.81640625,
|
|
"learning_rate": 0.0004885865893111716,
|
|
"loss": 5.8377,
|
|
"mean_token_accuracy": 0.16278486102819442,
|
|
"num_tokens": 4132914.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 6.083577823638916,
|
|
"epoch": 1.9951325040562466,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004884513273604195,
|
|
"loss": 5.9558,
|
|
"mean_token_accuracy": 0.15854081958532334,
|
|
"num_tokens": 4144213.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 6.0928910573323565,
|
|
"epoch": 2.0,
|
|
"grad_norm": 1.7734375,
|
|
"learning_rate": 0.0004883152897270502,
|
|
"loss": 5.9708,
|
|
"mean_token_accuracy": 0.16234943601820204,
|
|
"num_tokens": 4153778.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 6.038874340057373,
|
|
"epoch": 2.005408328826393,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0004881784769054253,
|
|
"loss": 5.6325,
|
|
"mean_token_accuracy": 0.16707420349121094,
|
|
"num_tokens": 4165536.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 6.132140398025513,
|
|
"epoch": 2.0108166576527853,
|
|
"grad_norm": 0.77734375,
|
|
"learning_rate": 0.0004880408893927226,
|
|
"loss": 5.699,
|
|
"mean_token_accuracy": 0.16293582916259766,
|
|
"num_tokens": 4176182.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 5.968934869766235,
|
|
"epoch": 2.016224986479178,
|
|
"grad_norm": 0.80078125,
|
|
"learning_rate": 0.0004879025276889355,
|
|
"loss": 5.6808,
|
|
"mean_token_accuracy": 0.1741676703095436,
|
|
"num_tokens": 4187592.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 5.9331096649169925,
|
|
"epoch": 2.0216333153055706,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.00048776339229687077,
|
|
"loss": 5.5039,
|
|
"mean_token_accuracy": 0.18047458976507186,
|
|
"num_tokens": 4198199.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 6.032744646072388,
|
|
"epoch": 2.0270416441319634,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.00048762348372214683,
|
|
"loss": 5.7308,
|
|
"mean_token_accuracy": 0.17150004357099533,
|
|
"num_tokens": 4208116.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 5.954532957077026,
|
|
"epoch": 2.032449972958356,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.00048748280247319165,
|
|
"loss": 5.51,
|
|
"mean_token_accuracy": 0.1745804950594902,
|
|
"num_tokens": 4218569.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 6.012410497665405,
|
|
"epoch": 2.0378583017847487,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0004873413490612412,
|
|
"loss": 5.6683,
|
|
"mean_token_accuracy": 0.17654334455728532,
|
|
"num_tokens": 4228785.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 5.967253875732422,
|
|
"epoch": 2.043266630611141,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00048719912400033757,
|
|
"loss": 5.6134,
|
|
"mean_token_accuracy": 0.17209813743829727,
|
|
"num_tokens": 4238814.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 5.937309980392456,
|
|
"epoch": 2.048674959437534,
|
|
"grad_norm": 0.78125,
|
|
"learning_rate": 0.000487056127807327,
|
|
"loss": 5.5989,
|
|
"mean_token_accuracy": 0.1668916642665863,
|
|
"num_tokens": 4250544.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 5.972121667861939,
|
|
"epoch": 2.0540832882639264,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00048691236100185795,
|
|
"loss": 5.7262,
|
|
"mean_token_accuracy": 0.1700291305780411,
|
|
"num_tokens": 4261806.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 6.081006002426148,
|
|
"epoch": 2.0594916170903192,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004867678241063792,
|
|
"loss": 5.6597,
|
|
"mean_token_accuracy": 0.17602354884147645,
|
|
"num_tokens": 4272472.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 5.96207184791565,
|
|
"epoch": 2.0648999459167117,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.00048662251764613833,
|
|
"loss": 5.6454,
|
|
"mean_token_accuracy": 0.17205152809619903,
|
|
"num_tokens": 4283758.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 5.960440158843994,
|
|
"epoch": 2.0703082747431045,
|
|
"grad_norm": 0.81640625,
|
|
"learning_rate": 0.0004864764421491793,
|
|
"loss": 5.7573,
|
|
"mean_token_accuracy": 0.17194358259439468,
|
|
"num_tokens": 4295930.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 6.125428771972656,
|
|
"epoch": 2.075716603569497,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.0004863295981463408,
|
|
"loss": 5.685,
|
|
"mean_token_accuracy": 0.1795160949230194,
|
|
"num_tokens": 4306711.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 5.9089724063873295,
|
|
"epoch": 2.08112493239589,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004861819861712543,
|
|
"loss": 5.6524,
|
|
"mean_token_accuracy": 0.17312369346618653,
|
|
"num_tokens": 4317694.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 6.112170314788818,
|
|
"epoch": 2.086533261222282,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.00048603360676034203,
|
|
"loss": 5.7425,
|
|
"mean_token_accuracy": 0.16565849781036376,
|
|
"num_tokens": 4329552.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 5.929319953918457,
|
|
"epoch": 2.091941590048675,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.00048588446045281505,
|
|
"loss": 5.6863,
|
|
"mean_token_accuracy": 0.16775599867105484,
|
|
"num_tokens": 4340479.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 5.901368093490601,
|
|
"epoch": 2.0973499188750675,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004857345477906714,
|
|
"loss": 5.5506,
|
|
"mean_token_accuracy": 0.17553742080926896,
|
|
"num_tokens": 4351147.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 5.842008829116821,
|
|
"epoch": 2.1027582477014604,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.00048558386931869417,
|
|
"loss": 5.558,
|
|
"mean_token_accuracy": 0.17481432408094405,
|
|
"num_tokens": 4362962.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 6.008848047256469,
|
|
"epoch": 2.1081665765278528,
|
|
"grad_norm": 0.796875,
|
|
"learning_rate": 0.0004854324255844492,
|
|
"loss": 5.6692,
|
|
"mean_token_accuracy": 0.17549348920583724,
|
|
"num_tokens": 4373723.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 5.859284687042236,
|
|
"epoch": 2.1135749053542456,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00048528021713828336,
|
|
"loss": 5.6279,
|
|
"mean_token_accuracy": 0.17020242065191268,
|
|
"num_tokens": 4383434.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 5.940160512924194,
|
|
"epoch": 2.118983234180638,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004851272445333225,
|
|
"loss": 5.5911,
|
|
"mean_token_accuracy": 0.17354973703622817,
|
|
"num_tokens": 4394847.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 5.966811418533325,
|
|
"epoch": 2.124391563007031,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.00048497350832546937,
|
|
"loss": 5.6989,
|
|
"mean_token_accuracy": 0.17423516064882277,
|
|
"num_tokens": 4406367.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 6.029182243347168,
|
|
"epoch": 2.1297998918334233,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004848190090734018,
|
|
"loss": 5.6875,
|
|
"mean_token_accuracy": 0.17333697378635407,
|
|
"num_tokens": 4417636.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 5.945331001281739,
|
|
"epoch": 2.135208220659816,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.00048466374733857035,
|
|
"loss": 5.702,
|
|
"mean_token_accuracy": 0.17600674629211427,
|
|
"num_tokens": 4428473.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 5.949156856536865,
|
|
"epoch": 2.1406165494862086,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004845077236851966,
|
|
"loss": 5.626,
|
|
"mean_token_accuracy": 0.17079080790281295,
|
|
"num_tokens": 4439780.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 5.931070041656494,
|
|
"epoch": 2.1460248783126015,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.00048435093868027076,
|
|
"loss": 5.5751,
|
|
"mean_token_accuracy": 0.1878390669822693,
|
|
"num_tokens": 4450929.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 5.899034023284912,
|
|
"epoch": 2.151433207138994,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.00048419339289354995,
|
|
"loss": 5.6144,
|
|
"mean_token_accuracy": 0.1744977578520775,
|
|
"num_tokens": 4462967.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 5.990045833587646,
|
|
"epoch": 2.1568415359653867,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.000484035086897556,
|
|
"loss": 5.7256,
|
|
"mean_token_accuracy": 0.17501339465379714,
|
|
"num_tokens": 4473970.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 5.987075233459473,
|
|
"epoch": 2.162249864791779,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004838760212675732,
|
|
"loss": 5.6984,
|
|
"mean_token_accuracy": 0.1695317879319191,
|
|
"num_tokens": 4485660.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 2.162249864791779,
|
|
"eval_entropy": 5.967337659272281,
|
|
"eval_loss": 6.1111321449279785,
|
|
"eval_mean_token_accuracy": 0.16192942243916067,
|
|
"eval_num_tokens": 4485660.0,
|
|
"eval_runtime": 3.531,
|
|
"eval_samples_per_second": 995.473,
|
|
"eval_steps_per_second": 124.611,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 6.009161901473999,
|
|
"epoch": 2.167658193618172,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004837161965816464,
|
|
"loss": 5.6804,
|
|
"mean_token_accuracy": 0.17444284856319428,
|
|
"num_tokens": 4497261.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 5.890836334228515,
|
|
"epoch": 2.1730665224445644,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.00048355561342057886,
|
|
"loss": 5.5702,
|
|
"mean_token_accuracy": 0.18166276961565017,
|
|
"num_tokens": 4508120.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 5.997749853134155,
|
|
"epoch": 2.1784748512709573,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.00048339427236793024,
|
|
"loss": 5.6643,
|
|
"mean_token_accuracy": 0.17516567558050156,
|
|
"num_tokens": 4519286.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 5.983465051651001,
|
|
"epoch": 2.1838831800973497,
|
|
"grad_norm": 0.75,
|
|
"learning_rate": 0.0004832321740100143,
|
|
"loss": 5.7145,
|
|
"mean_token_accuracy": 0.17017477452754975,
|
|
"num_tokens": 4531912.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 6.05001859664917,
|
|
"epoch": 2.1892915089237426,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.00048306931893589666,
|
|
"loss": 5.7422,
|
|
"mean_token_accuracy": 0.170116625726223,
|
|
"num_tokens": 4544370.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 5.916760873794556,
|
|
"epoch": 2.194699837750135,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.00048290570773739327,
|
|
"loss": 5.6841,
|
|
"mean_token_accuracy": 0.17166270464658737,
|
|
"num_tokens": 4556013.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 5.9200887203216555,
|
|
"epoch": 2.200108166576528,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.00048274134100906747,
|
|
"loss": 5.6354,
|
|
"mean_token_accuracy": 0.1718206897377968,
|
|
"num_tokens": 4566200.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 5.96041841506958,
|
|
"epoch": 2.2055164954029207,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.00048257621934822835,
|
|
"loss": 5.6847,
|
|
"mean_token_accuracy": 0.1735834926366806,
|
|
"num_tokens": 4576899.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 6.021680450439453,
|
|
"epoch": 2.210924824229313,
|
|
"grad_norm": 0.79296875,
|
|
"learning_rate": 0.0004824103433549285,
|
|
"loss": 5.668,
|
|
"mean_token_accuracy": 0.16992697566747667,
|
|
"num_tokens": 4588983.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 5.919964599609375,
|
|
"epoch": 2.2163331530557056,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004822437136319616,
|
|
"loss": 5.6343,
|
|
"mean_token_accuracy": 0.17985793501138686,
|
|
"num_tokens": 4599758.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 5.931059885025024,
|
|
"epoch": 2.2217414818820984,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004820763307848604,
|
|
"loss": 5.6744,
|
|
"mean_token_accuracy": 0.1735081821680069,
|
|
"num_tokens": 4610927.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 5.934611368179321,
|
|
"epoch": 2.2271498107084913,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.00048190819542189456,
|
|
"loss": 5.7026,
|
|
"mean_token_accuracy": 0.17243197113275527,
|
|
"num_tokens": 4623719.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 5.981340122222901,
|
|
"epoch": 2.2325581395348837,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.00048173930815406847,
|
|
"loss": 5.7642,
|
|
"mean_token_accuracy": 0.1689779832959175,
|
|
"num_tokens": 4635389.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 5.9712042808532715,
|
|
"epoch": 2.2379664683612766,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004815696695951187,
|
|
"loss": 5.7046,
|
|
"mean_token_accuracy": 0.1701254963874817,
|
|
"num_tokens": 4645093.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 5.921566295623779,
|
|
"epoch": 2.243374797187669,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004813992803615123,
|
|
"loss": 5.6952,
|
|
"mean_token_accuracy": 0.16522742807865143,
|
|
"num_tokens": 4656490.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 6.0387671947479244,
|
|
"epoch": 2.248783126014062,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00048122814107244414,
|
|
"loss": 5.7102,
|
|
"mean_token_accuracy": 0.17248750180006028,
|
|
"num_tokens": 4667603.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 5.879037141799927,
|
|
"epoch": 2.2541914548404542,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.00048105625234983466,
|
|
"loss": 5.6542,
|
|
"mean_token_accuracy": 0.16627398133277893,
|
|
"num_tokens": 4678669.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 5.898302793502808,
|
|
"epoch": 2.259599783666847,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00048088361481832807,
|
|
"loss": 5.4423,
|
|
"mean_token_accuracy": 0.18547236919403076,
|
|
"num_tokens": 4689757.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 5.818560361862183,
|
|
"epoch": 2.2650081124932395,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00048071022910528934,
|
|
"loss": 5.6272,
|
|
"mean_token_accuracy": 0.17067526429891586,
|
|
"num_tokens": 4699332.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 5.984354543685913,
|
|
"epoch": 2.2704164413196324,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00048053609584080274,
|
|
"loss": 5.7134,
|
|
"mean_token_accuracy": 0.1770985320210457,
|
|
"num_tokens": 4709485.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 5.869245529174805,
|
|
"epoch": 2.275824770146025,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.00048036121565766876,
|
|
"loss": 5.6335,
|
|
"mean_token_accuracy": 0.17194894403219224,
|
|
"num_tokens": 4720935.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 5.875689172744751,
|
|
"epoch": 2.2812330989724177,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0004801855891914026,
|
|
"loss": 5.5988,
|
|
"mean_token_accuracy": 0.17105952203273772,
|
|
"num_tokens": 4731219.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 5.963768005371094,
|
|
"epoch": 2.28664142779881,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00048000921708023117,
|
|
"loss": 5.7192,
|
|
"mean_token_accuracy": 0.17015954703092576,
|
|
"num_tokens": 4742637.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 5.9226116180419925,
|
|
"epoch": 2.292049756625203,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004798320999650911,
|
|
"loss": 5.6193,
|
|
"mean_token_accuracy": 0.1703588470816612,
|
|
"num_tokens": 4753341.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 6.100405216217041,
|
|
"epoch": 2.2974580854515954,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.00047965423848962636,
|
|
"loss": 5.7774,
|
|
"mean_token_accuracy": 0.17631693184375763,
|
|
"num_tokens": 4765339.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 5.9548952102661135,
|
|
"epoch": 2.302866414277988,
|
|
"grad_norm": 0.79296875,
|
|
"learning_rate": 0.000479475633300186,
|
|
"loss": 5.6155,
|
|
"mean_token_accuracy": 0.1795394852757454,
|
|
"num_tokens": 4776149.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 5.871384334564209,
|
|
"epoch": 2.3082747431043806,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.00047929628504582173,
|
|
"loss": 5.6531,
|
|
"mean_token_accuracy": 0.17741299122571946,
|
|
"num_tokens": 4788120.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 5.998767900466919,
|
|
"epoch": 2.3136830719307735,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004791161943782856,
|
|
"loss": 5.7328,
|
|
"mean_token_accuracy": 0.16924816071987153,
|
|
"num_tokens": 4799922.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 5.823422861099243,
|
|
"epoch": 2.319091400757166,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.00047893536195202726,
|
|
"loss": 5.6683,
|
|
"mean_token_accuracy": 0.169066621363163,
|
|
"num_tokens": 4810501.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 5.91726450920105,
|
|
"epoch": 2.3244997295835588,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004787537884241923,
|
|
"loss": 5.601,
|
|
"mean_token_accuracy": 0.1761727049946785,
|
|
"num_tokens": 4820919.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 5.978335571289063,
|
|
"epoch": 2.329908058409951,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.00047857147445461943,
|
|
"loss": 5.7127,
|
|
"mean_token_accuracy": 0.17215983420610428,
|
|
"num_tokens": 4833298.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 5.849052619934082,
|
|
"epoch": 2.335316387236344,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004783884207058379,
|
|
"loss": 5.5755,
|
|
"mean_token_accuracy": 0.17292249351739883,
|
|
"num_tokens": 4845199.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 5.862699031829834,
|
|
"epoch": 2.3407247160627365,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00047820462784306555,
|
|
"loss": 5.6373,
|
|
"mean_token_accuracy": 0.17088074684143068,
|
|
"num_tokens": 4856549.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 5.923096704483032,
|
|
"epoch": 2.3461330448891293,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.000478020096534206,
|
|
"loss": 5.6859,
|
|
"mean_token_accuracy": 0.16967353075742722,
|
|
"num_tokens": 4868446.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 6.029429006576538,
|
|
"epoch": 2.3515413737155217,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.00047783482744984644,
|
|
"loss": 5.7257,
|
|
"mean_token_accuracy": 0.16710098087787628,
|
|
"num_tokens": 4879047.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 5.930259323120117,
|
|
"epoch": 2.3569497025419146,
|
|
"grad_norm": 0.796875,
|
|
"learning_rate": 0.00047764882126325527,
|
|
"loss": 5.6235,
|
|
"mean_token_accuracy": 0.1699826642870903,
|
|
"num_tokens": 4891591.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 5.901684713363648,
|
|
"epoch": 2.362358031368307,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 0.0004774620786503793,
|
|
"loss": 5.6281,
|
|
"mean_token_accuracy": 0.17609986066818237,
|
|
"num_tokens": 4902710.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 5.9096949100494385,
|
|
"epoch": 2.3677663601947,
|
|
"grad_norm": 0.80078125,
|
|
"learning_rate": 0.0004772746002898417,
|
|
"loss": 5.7105,
|
|
"mean_token_accuracy": 0.17462405413389206,
|
|
"num_tokens": 4914668.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 6.050159168243408,
|
|
"epoch": 2.3731746890210923,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00047708638686293903,
|
|
"loss": 5.8589,
|
|
"mean_token_accuracy": 0.16697658449411393,
|
|
"num_tokens": 4926684.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 6.019222831726074,
|
|
"epoch": 2.378583017847485,
|
|
"grad_norm": 0.7734375,
|
|
"learning_rate": 0.00047689743905363963,
|
|
"loss": 5.7189,
|
|
"mean_token_accuracy": 0.170701701939106,
|
|
"num_tokens": 4938577.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 5.828266859054565,
|
|
"epoch": 2.383991346673878,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004767077575485801,
|
|
"loss": 5.5812,
|
|
"mean_token_accuracy": 0.1782996416091919,
|
|
"num_tokens": 4949070.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 5.878254413604736,
|
|
"epoch": 2.3893996755002704,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004765173430370635,
|
|
"loss": 5.6504,
|
|
"mean_token_accuracy": 0.17152102589607238,
|
|
"num_tokens": 4959558.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 5.859272575378418,
|
|
"epoch": 2.394808004326663,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004763261962110567,
|
|
"loss": 5.5225,
|
|
"mean_token_accuracy": 0.17889102399349213,
|
|
"num_tokens": 4970310.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 5.802045440673828,
|
|
"epoch": 2.4002163331530557,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.0004761343177651875,
|
|
"loss": 5.5992,
|
|
"mean_token_accuracy": 0.1725127875804901,
|
|
"num_tokens": 4982300.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 5.910652208328247,
|
|
"epoch": 2.4056246619794486,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.0004759417083967428,
|
|
"loss": 5.6301,
|
|
"mean_token_accuracy": 0.17524359822273256,
|
|
"num_tokens": 4994425.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 5.834094285964966,
|
|
"epoch": 2.411032990805841,
|
|
"grad_norm": 0.80078125,
|
|
"learning_rate": 0.00047574836880566544,
|
|
"loss": 5.6279,
|
|
"mean_token_accuracy": 0.1814741834998131,
|
|
"num_tokens": 5005872.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 5.86660532951355,
|
|
"epoch": 2.4164413196322334,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004755542996945519,
|
|
"loss": 5.6812,
|
|
"mean_token_accuracy": 0.17532453089952468,
|
|
"num_tokens": 5016903.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 5.97542290687561,
|
|
"epoch": 2.4218496484586263,
|
|
"grad_norm": 0.80859375,
|
|
"learning_rate": 0.00047535950176864976,
|
|
"loss": 5.722,
|
|
"mean_token_accuracy": 0.17611724883317947,
|
|
"num_tokens": 5028407.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 5.881137847900391,
|
|
"epoch": 2.427257977285019,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.00047516397573585525,
|
|
"loss": 5.7158,
|
|
"mean_token_accuracy": 0.17345101535320281,
|
|
"num_tokens": 5039693.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 5.993742322921753,
|
|
"epoch": 2.4326663061114115,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.00047496772230671034,
|
|
"loss": 5.6622,
|
|
"mean_token_accuracy": 0.17378358244895936,
|
|
"num_tokens": 5051279.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 5.909454488754273,
|
|
"epoch": 2.4380746349378044,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004747707421944004,
|
|
"loss": 5.6179,
|
|
"mean_token_accuracy": 0.17530350983142853,
|
|
"num_tokens": 5062698.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 5.898091793060303,
|
|
"epoch": 2.443482963764197,
|
|
"grad_norm": 0.79296875,
|
|
"learning_rate": 0.0004745730361147517,
|
|
"loss": 5.594,
|
|
"mean_token_accuracy": 0.1770256981253624,
|
|
"num_tokens": 5074927.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 5.894674301147461,
|
|
"epoch": 2.4488912925905897,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004743746047862286,
|
|
"loss": 5.7309,
|
|
"mean_token_accuracy": 0.1637778252363205,
|
|
"num_tokens": 5086298.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 5.981807708740234,
|
|
"epoch": 2.454299621416982,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 0.0004741754489299309,
|
|
"loss": 5.668,
|
|
"mean_token_accuracy": 0.1721787706017494,
|
|
"num_tokens": 5098161.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 5.831097507476807,
|
|
"epoch": 2.459707950243375,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.00047397556926959155,
|
|
"loss": 5.6566,
|
|
"mean_token_accuracy": 0.1823459669947624,
|
|
"num_tokens": 5110743.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 5.993427467346192,
|
|
"epoch": 2.4651162790697674,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.00047377496653157374,
|
|
"loss": 5.7046,
|
|
"mean_token_accuracy": 0.1776488870382309,
|
|
"num_tokens": 5121838.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 5.861187219619751,
|
|
"epoch": 2.4705246078961602,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00047357364144486806,
|
|
"loss": 5.6353,
|
|
"mean_token_accuracy": 0.17126872688531875,
|
|
"num_tokens": 5135047.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 5.844826650619507,
|
|
"epoch": 2.4759329367225527,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.00047337159474109057,
|
|
"loss": 5.6107,
|
|
"mean_token_accuracy": 0.17242614775896073,
|
|
"num_tokens": 5146329.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 5.949182653427124,
|
|
"epoch": 2.4813412655489455,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.00047316882715447934,
|
|
"loss": 5.6846,
|
|
"mean_token_accuracy": 0.17800829112529754,
|
|
"num_tokens": 5157511.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 5.8185004711151125,
|
|
"epoch": 2.486749594375338,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004729653394218922,
|
|
"loss": 5.6579,
|
|
"mean_token_accuracy": 0.17028010934591292,
|
|
"num_tokens": 5168456.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 6.005807256698608,
|
|
"epoch": 2.492157923201731,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.00047276113228280427,
|
|
"loss": 5.7136,
|
|
"mean_token_accuracy": 0.17623213529586793,
|
|
"num_tokens": 5179294.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 5.961325216293335,
|
|
"epoch": 2.497566252028123,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.0004725562064793045,
|
|
"loss": 5.6651,
|
|
"mean_token_accuracy": 0.1779454082250595,
|
|
"num_tokens": 5190195.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 5.758653068542481,
|
|
"epoch": 2.502974580854516,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00047235056275609397,
|
|
"loss": 5.5326,
|
|
"mean_token_accuracy": 0.1819295510649681,
|
|
"num_tokens": 5201289.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 5.838653421401977,
|
|
"epoch": 2.5083829096809085,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004721442018604822,
|
|
"loss": 5.57,
|
|
"mean_token_accuracy": 0.183014440536499,
|
|
"num_tokens": 5212295.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 5.829467916488648,
|
|
"epoch": 2.5137912385073014,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.0004719371245423853,
|
|
"loss": 5.5381,
|
|
"mean_token_accuracy": 0.18126602470874786,
|
|
"num_tokens": 5223144.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 5.765709447860718,
|
|
"epoch": 2.5191995673336938,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.00047172933155432273,
|
|
"loss": 5.5491,
|
|
"mean_token_accuracy": 0.17753212600946428,
|
|
"num_tokens": 5232840.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 5.896287298202514,
|
|
"epoch": 2.5246078961600866,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.00047152082365141466,
|
|
"loss": 5.6568,
|
|
"mean_token_accuracy": 0.1709638550877571,
|
|
"num_tokens": 5243796.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 5.951115417480469,
|
|
"epoch": 2.530016224986479,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004713116015913794,
|
|
"loss": 5.7522,
|
|
"mean_token_accuracy": 0.17281907349824904,
|
|
"num_tokens": 5255990.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 5.9733569622039795,
|
|
"epoch": 2.535424553812872,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.00047110166613453016,
|
|
"loss": 5.7016,
|
|
"mean_token_accuracy": 0.17224203944206237,
|
|
"num_tokens": 5268134.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 5.7451770305633545,
|
|
"epoch": 2.5408328826392643,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.00047089101804377305,
|
|
"loss": 5.5472,
|
|
"mean_token_accuracy": 0.17507804930210114,
|
|
"num_tokens": 5278662.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 5.86171588897705,
|
|
"epoch": 2.546241211465657,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004706796580846037,
|
|
"loss": 5.6181,
|
|
"mean_token_accuracy": 0.17575339525938033,
|
|
"num_tokens": 5288382.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 5.922559595108032,
|
|
"epoch": 2.5516495402920496,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.00047046758702510465,
|
|
"loss": 5.6495,
|
|
"mean_token_accuracy": 0.1694240838289261,
|
|
"num_tokens": 5299769.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 5.962764835357666,
|
|
"epoch": 2.5570578691184425,
|
|
"grad_norm": 0.8203125,
|
|
"learning_rate": 0.00047025480563594264,
|
|
"loss": 5.6233,
|
|
"mean_token_accuracy": 0.17845901548862458,
|
|
"num_tokens": 5310662.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 5.814746141433716,
|
|
"epoch": 2.5624661979448353,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.00047004131469036576,
|
|
"loss": 5.593,
|
|
"mean_token_accuracy": 0.1722265288233757,
|
|
"num_tokens": 5321785.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 5.901667499542237,
|
|
"epoch": 2.5678745267712277,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.00046982711496420057,
|
|
"loss": 5.6242,
|
|
"mean_token_accuracy": 0.1716166153550148,
|
|
"num_tokens": 5333557.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 5.924446439743042,
|
|
"epoch": 2.57328285559762,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.00046961220723584925,
|
|
"loss": 5.6765,
|
|
"mean_token_accuracy": 0.1697666823863983,
|
|
"num_tokens": 5344767.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 5.874181699752808,
|
|
"epoch": 2.578691184424013,
|
|
"grad_norm": 0.7734375,
|
|
"learning_rate": 0.00046939659228628725,
|
|
"loss": 5.5978,
|
|
"mean_token_accuracy": 0.1760467231273651,
|
|
"num_tokens": 5356673.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 5.874102354049683,
|
|
"epoch": 2.584099513250406,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00046918027089905956,
|
|
"loss": 5.7119,
|
|
"mean_token_accuracy": 0.16885269433259964,
|
|
"num_tokens": 5367296.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 5.8442058086395265,
|
|
"epoch": 2.5895078420767983,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.00046896324386027873,
|
|
"loss": 5.5834,
|
|
"mean_token_accuracy": 0.1757770448923111,
|
|
"num_tokens": 5378081.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 5.8963714122772215,
|
|
"epoch": 2.5949161709031907,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.00046874551195862145,
|
|
"loss": 5.6618,
|
|
"mean_token_accuracy": 0.17249250411987305,
|
|
"num_tokens": 5390244.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 5.818678188323974,
|
|
"epoch": 2.6003244997295836,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.000468527075985326,
|
|
"loss": 5.525,
|
|
"mean_token_accuracy": 0.18108315616846085,
|
|
"num_tokens": 5400741.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 5.944147539138794,
|
|
"epoch": 2.6057328285559764,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.00046830793673418924,
|
|
"loss": 5.7406,
|
|
"mean_token_accuracy": 0.17808103561401367,
|
|
"num_tokens": 5411440.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 5.851105117797852,
|
|
"epoch": 2.611141157382369,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00046808809500156355,
|
|
"loss": 5.5795,
|
|
"mean_token_accuracy": 0.17286153137683868,
|
|
"num_tokens": 5422472.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 5.849776124954223,
|
|
"epoch": 2.6165494862087613,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0004678675515863545,
|
|
"loss": 5.6645,
|
|
"mean_token_accuracy": 0.16712078154087068,
|
|
"num_tokens": 5433958.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 5.861989068984985,
|
|
"epoch": 2.621957815035154,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004676463072900173,
|
|
"loss": 5.577,
|
|
"mean_token_accuracy": 0.18329984545707703,
|
|
"num_tokens": 5444343.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 5.826681184768677,
|
|
"epoch": 2.627366143861547,
|
|
"grad_norm": 0.80078125,
|
|
"learning_rate": 0.00046742436291655415,
|
|
"loss": 5.5761,
|
|
"mean_token_accuracy": 0.17418364882469178,
|
|
"num_tokens": 5456260.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 5.866898536682129,
|
|
"epoch": 2.6327744726879394,
|
|
"grad_norm": 0.78515625,
|
|
"learning_rate": 0.00046720171927251155,
|
|
"loss": 5.6709,
|
|
"mean_token_accuracy": 0.17819034904241562,
|
|
"num_tokens": 5468281.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 5.729748725891113,
|
|
"epoch": 2.638182801514332,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 0.0004669783771669769,
|
|
"loss": 5.5208,
|
|
"mean_token_accuracy": 0.19250661432743071,
|
|
"num_tokens": 5480068.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 5.784274673461914,
|
|
"epoch": 2.6435911303407247,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.000466754337411576,
|
|
"loss": 5.5069,
|
|
"mean_token_accuracy": 0.1778206929564476,
|
|
"num_tokens": 5490932.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 5.7609247207641605,
|
|
"epoch": 2.6489994591671175,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.00046652960082046985,
|
|
"loss": 5.6247,
|
|
"mean_token_accuracy": 0.17345626205205916,
|
|
"num_tokens": 5501787.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 5.870206212997436,
|
|
"epoch": 2.65440778799351,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.00046630416821035164,
|
|
"loss": 5.6032,
|
|
"mean_token_accuracy": 0.17429625540971755,
|
|
"num_tokens": 5513297.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 5.817053031921387,
|
|
"epoch": 2.6598161168199024,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004660780404004441,
|
|
"loss": 5.5823,
|
|
"mean_token_accuracy": 0.17704997807741166,
|
|
"num_tokens": 5523797.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 5.779816389083862,
|
|
"epoch": 2.6652244456462952,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.00046585121821249623,
|
|
"loss": 5.6018,
|
|
"mean_token_accuracy": 0.18384530395269394,
|
|
"num_tokens": 5535808.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 5.941058111190796,
|
|
"epoch": 2.670632774472688,
|
|
"grad_norm": 0.77734375,
|
|
"learning_rate": 0.00046562370247078025,
|
|
"loss": 5.693,
|
|
"mean_token_accuracy": 0.16583001911640166,
|
|
"num_tokens": 5548179.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 5.916377401351928,
|
|
"epoch": 2.6760411032990805,
|
|
"grad_norm": 0.80859375,
|
|
"learning_rate": 0.000465395494002089,
|
|
"loss": 5.7105,
|
|
"mean_token_accuracy": 0.16976383179426194,
|
|
"num_tokens": 5559144.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 5.826733827590942,
|
|
"epoch": 2.6814494321254734,
|
|
"grad_norm": 0.796875,
|
|
"learning_rate": 0.0004651665936357325,
|
|
"loss": 5.592,
|
|
"mean_token_accuracy": 0.1779371291399002,
|
|
"num_tokens": 5571103.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 5.8336090564727785,
|
|
"epoch": 2.686857760951866,
|
|
"grad_norm": 0.8203125,
|
|
"learning_rate": 0.0004649370022035354,
|
|
"loss": 5.5323,
|
|
"mean_token_accuracy": 0.17698664963245392,
|
|
"num_tokens": 5583239.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 5.794463396072388,
|
|
"epoch": 2.6922660897782587,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004647067205398332,
|
|
"loss": 5.5076,
|
|
"mean_token_accuracy": 0.17905741482973098,
|
|
"num_tokens": 5593602.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 5.913798856735229,
|
|
"epoch": 2.697674418604651,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.00046447574948147027,
|
|
"loss": 5.6733,
|
|
"mean_token_accuracy": 0.17029502838850022,
|
|
"num_tokens": 5604696.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 5.788999462127686,
|
|
"epoch": 2.703082747431044,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.00046424408986779587,
|
|
"loss": 5.6405,
|
|
"mean_token_accuracy": 0.16970723271369934,
|
|
"num_tokens": 5615668.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"epoch": 2.703082747431044,
|
|
"eval_entropy": 5.768812253258445,
|
|
"eval_loss": 5.997067451477051,
|
|
"eval_mean_token_accuracy": 0.16798656996000896,
|
|
"eval_num_tokens": 5615668.0,
|
|
"eval_runtime": 3.5301,
|
|
"eval_samples_per_second": 995.734,
|
|
"eval_steps_per_second": 124.644,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 5.871507740020752,
|
|
"epoch": 2.7084910762574363,
|
|
"grad_norm": 0.79296875,
|
|
"learning_rate": 0.0004640117425406615,
|
|
"loss": 5.5861,
|
|
"mean_token_accuracy": 0.17402273416519165,
|
|
"num_tokens": 5627344.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 5.816973114013672,
|
|
"epoch": 2.713899405083829,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.000463778708344418,
|
|
"loss": 5.6162,
|
|
"mean_token_accuracy": 0.1698262020945549,
|
|
"num_tokens": 5637472.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 5.838492155075073,
|
|
"epoch": 2.7193077339102216,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00046354498812591207,
|
|
"loss": 5.593,
|
|
"mean_token_accuracy": 0.17822807133197785,
|
|
"num_tokens": 5647766.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 5.756899166107178,
|
|
"epoch": 2.7247160627366145,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.00046331058273448364,
|
|
"loss": 5.5315,
|
|
"mean_token_accuracy": 0.17876374572515488,
|
|
"num_tokens": 5658343.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 5.830268621444702,
|
|
"epoch": 2.730124391563007,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004630754930219624,
|
|
"loss": 5.5274,
|
|
"mean_token_accuracy": 0.17779684215784072,
|
|
"num_tokens": 5669640.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 5.722399377822876,
|
|
"epoch": 2.7355327203893998,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0004628397198426648,
|
|
"loss": 5.5105,
|
|
"mean_token_accuracy": 0.17078663855791093,
|
|
"num_tokens": 5682421.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 5.80373067855835,
|
|
"epoch": 2.740941049215792,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.00046260326405339126,
|
|
"loss": 5.4883,
|
|
"mean_token_accuracy": 0.18236415833234787,
|
|
"num_tokens": 5692797.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 5.822386312484741,
|
|
"epoch": 2.746349378042185,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004623661265134226,
|
|
"loss": 5.6664,
|
|
"mean_token_accuracy": 0.17654704302549362,
|
|
"num_tokens": 5702779.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 5.918497848510742,
|
|
"epoch": 2.7517577068685775,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004621283080845172,
|
|
"loss": 5.663,
|
|
"mean_token_accuracy": 0.17752265632152558,
|
|
"num_tokens": 5714147.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 5.775480699539185,
|
|
"epoch": 2.7571660356949703,
|
|
"grad_norm": 0.76953125,
|
|
"learning_rate": 0.00046188980963090783,
|
|
"loss": 5.4802,
|
|
"mean_token_accuracy": 0.18461580723524093,
|
|
"num_tokens": 5725718.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 5.782088947296143,
|
|
"epoch": 2.7625743645213627,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.00046165063201929825,
|
|
"loss": 5.5608,
|
|
"mean_token_accuracy": 0.18376427441835402,
|
|
"num_tokens": 5737441.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 5.866076374053955,
|
|
"epoch": 2.7679826933477556,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00046141077611886074,
|
|
"loss": 5.6701,
|
|
"mean_token_accuracy": 0.17962786555290222,
|
|
"num_tokens": 5748735.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 5.876386070251465,
|
|
"epoch": 2.773391022174148,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.0004611702428012319,
|
|
"loss": 5.7241,
|
|
"mean_token_accuracy": 0.17184043377637864,
|
|
"num_tokens": 5760262.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 5.827700901031494,
|
|
"epoch": 2.778799351000541,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.00046092903294051057,
|
|
"loss": 5.6277,
|
|
"mean_token_accuracy": 0.17197844982147217,
|
|
"num_tokens": 5771909.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 5.746778774261474,
|
|
"epoch": 2.7842076798269337,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.00046068714741325385,
|
|
"loss": 5.5074,
|
|
"mean_token_accuracy": 0.18136420100927353,
|
|
"num_tokens": 5782028.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 5.926895189285278,
|
|
"epoch": 2.789616008653326,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004604445870984743,
|
|
"loss": 5.6703,
|
|
"mean_token_accuracy": 0.17582763284444808,
|
|
"num_tokens": 5793641.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 5.88707127571106,
|
|
"epoch": 2.7950243374797186,
|
|
"grad_norm": 0.79296875,
|
|
"learning_rate": 0.0004602013528776367,
|
|
"loss": 5.6228,
|
|
"mean_token_accuracy": 0.17337359935045243,
|
|
"num_tokens": 5805392.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 5.856297540664673,
|
|
"epoch": 2.8004326663061114,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004599574456346548,
|
|
"loss": 5.6008,
|
|
"mean_token_accuracy": 0.17427357882261277,
|
|
"num_tokens": 5816362.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 5.776535749435425,
|
|
"epoch": 2.8058409951325043,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.00045971286625588796,
|
|
"loss": 5.5224,
|
|
"mean_token_accuracy": 0.18254357427358628,
|
|
"num_tokens": 5827518.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 5.925326108932495,
|
|
"epoch": 2.8112493239588967,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004594676156301383,
|
|
"loss": 5.6717,
|
|
"mean_token_accuracy": 0.17464789897203445,
|
|
"num_tokens": 5838455.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 5.769072771072388,
|
|
"epoch": 2.816657652785289,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.00045922169464864725,
|
|
"loss": 5.533,
|
|
"mean_token_accuracy": 0.18585131019353868,
|
|
"num_tokens": 5850065.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 5.866156673431396,
|
|
"epoch": 2.822065981611682,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.000458975104205092,
|
|
"loss": 5.6765,
|
|
"mean_token_accuracy": 0.16976690292358398,
|
|
"num_tokens": 5860444.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 5.939409828186035,
|
|
"epoch": 2.827474310438075,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.00045872784519558304,
|
|
"loss": 5.6545,
|
|
"mean_token_accuracy": 0.17222582846879958,
|
|
"num_tokens": 5873984.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 5.821524620056152,
|
|
"epoch": 2.8328826392644673,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.0004584799185186599,
|
|
"loss": 5.6664,
|
|
"mean_token_accuracy": 0.17595746219158173,
|
|
"num_tokens": 5885807.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 5.802651500701904,
|
|
"epoch": 2.8382909680908597,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.00045823132507528903,
|
|
"loss": 5.5426,
|
|
"mean_token_accuracy": 0.1819417729973793,
|
|
"num_tokens": 5897122.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 5.7716185569763185,
|
|
"epoch": 2.8436992969172525,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0004579820657688593,
|
|
"loss": 5.5491,
|
|
"mean_token_accuracy": 0.18012744039297104,
|
|
"num_tokens": 5909189.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 5.883975315093994,
|
|
"epoch": 2.8491076257436454,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004577321415051798,
|
|
"loss": 5.638,
|
|
"mean_token_accuracy": 0.1742127478122711,
|
|
"num_tokens": 5920135.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 5.859697246551514,
|
|
"epoch": 2.854515954570038,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.00045748155319247564,
|
|
"loss": 5.621,
|
|
"mean_token_accuracy": 0.18195434361696244,
|
|
"num_tokens": 5932279.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 5.780061674118042,
|
|
"epoch": 2.8599242833964302,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0004572303017413856,
|
|
"loss": 5.5421,
|
|
"mean_token_accuracy": 0.17683554887771608,
|
|
"num_tokens": 5942082.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 5.877644157409668,
|
|
"epoch": 2.865332612222823,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004569783880649578,
|
|
"loss": 5.598,
|
|
"mean_token_accuracy": 0.175541952252388,
|
|
"num_tokens": 5952856.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 5.702970361709594,
|
|
"epoch": 2.870740941049216,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004567258130786474,
|
|
"loss": 5.4833,
|
|
"mean_token_accuracy": 0.1861475631594658,
|
|
"num_tokens": 5963937.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 5.803145885467529,
|
|
"epoch": 2.8761492698756084,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00045647257770031215,
|
|
"loss": 5.5817,
|
|
"mean_token_accuracy": 0.18028615266084672,
|
|
"num_tokens": 5974852.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 5.8312476634979244,
|
|
"epoch": 2.8815575987020012,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.00045621868285021023,
|
|
"loss": 5.5892,
|
|
"mean_token_accuracy": 0.18228398263454437,
|
|
"num_tokens": 5985616.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 5.847126150131226,
|
|
"epoch": 2.8869659275283936,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00045596412945099615,
|
|
"loss": 5.5403,
|
|
"mean_token_accuracy": 0.18726042956113814,
|
|
"num_tokens": 5997199.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 5.786398935317993,
|
|
"epoch": 2.8923742563547865,
|
|
"grad_norm": 0.81640625,
|
|
"learning_rate": 0.00045570891842771727,
|
|
"loss": 5.5987,
|
|
"mean_token_accuracy": 0.17823826968669892,
|
|
"num_tokens": 6008654.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 5.820050621032715,
|
|
"epoch": 2.897782585181179,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.00045545305070781135,
|
|
"loss": 5.6461,
|
|
"mean_token_accuracy": 0.17028469890356063,
|
|
"num_tokens": 6020662.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 5.89305157661438,
|
|
"epoch": 2.903190914007572,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.00045519652722110206,
|
|
"loss": 5.5322,
|
|
"mean_token_accuracy": 0.19578891694545747,
|
|
"num_tokens": 6031793.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 5.720826578140259,
|
|
"epoch": 2.908599242833964,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00045493934889979655,
|
|
"loss": 5.5161,
|
|
"mean_token_accuracy": 0.18341525197029113,
|
|
"num_tokens": 6042181.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 5.687424802780152,
|
|
"epoch": 2.914007571660357,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.00045468151667848116,
|
|
"loss": 5.4899,
|
|
"mean_token_accuracy": 0.18657083064317703,
|
|
"num_tokens": 6052774.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 5.909038496017456,
|
|
"epoch": 2.9194159004867495,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.000454423031494119,
|
|
"loss": 5.493,
|
|
"mean_token_accuracy": 0.1903107300400734,
|
|
"num_tokens": 6063273.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 5.680450582504273,
|
|
"epoch": 2.9248242293131423,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004541638942860458,
|
|
"loss": 5.4966,
|
|
"mean_token_accuracy": 0.18520134687423706,
|
|
"num_tokens": 6075121.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 5.692252206802368,
|
|
"epoch": 2.9302325581395348,
|
|
"grad_norm": 0.8203125,
|
|
"learning_rate": 0.0004539041059959667,
|
|
"loss": 5.4711,
|
|
"mean_token_accuracy": 0.18130189329385757,
|
|
"num_tokens": 6085666.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 5.8206319332122805,
|
|
"epoch": 2.9356408869659276,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.00045364366756795295,
|
|
"loss": 5.4994,
|
|
"mean_token_accuracy": 0.1884692370891571,
|
|
"num_tokens": 6096619.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 5.84162654876709,
|
|
"epoch": 2.94104921579232,
|
|
"grad_norm": 0.80859375,
|
|
"learning_rate": 0.00045338257994843845,
|
|
"loss": 5.6515,
|
|
"mean_token_accuracy": 0.1753935068845749,
|
|
"num_tokens": 6108800.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 5.802308082580566,
|
|
"epoch": 2.946457544618713,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.0004531208440862162,
|
|
"loss": 5.5638,
|
|
"mean_token_accuracy": 0.1790132462978363,
|
|
"num_tokens": 6119789.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 5.744951248168945,
|
|
"epoch": 2.9518658734451053,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.00045285846093243485,
|
|
"loss": 5.5479,
|
|
"mean_token_accuracy": 0.17773659080266951,
|
|
"num_tokens": 6131786.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 5.885506057739258,
|
|
"epoch": 2.957274202271498,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.00045259543144059557,
|
|
"loss": 5.6498,
|
|
"mean_token_accuracy": 0.17503989338874817,
|
|
"num_tokens": 6143012.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 5.773147869110107,
|
|
"epoch": 2.9626825310978906,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0004523317565665479,
|
|
"loss": 5.5156,
|
|
"mean_token_accuracy": 0.1824488028883934,
|
|
"num_tokens": 6153972.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 5.858189439773559,
|
|
"epoch": 2.9680908599242835,
|
|
"grad_norm": 0.81640625,
|
|
"learning_rate": 0.00045206743726848716,
|
|
"loss": 5.6961,
|
|
"mean_token_accuracy": 0.17325770556926728,
|
|
"num_tokens": 6166202.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 5.743167352676392,
|
|
"epoch": 2.973499188750676,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.0004518024745069502,
|
|
"loss": 5.4152,
|
|
"mean_token_accuracy": 0.19447794556617737,
|
|
"num_tokens": 6177253.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 5.760520029067993,
|
|
"epoch": 2.9789075175770687,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004515368692448123,
|
|
"loss": 5.682,
|
|
"mean_token_accuracy": 0.1684027850627899,
|
|
"num_tokens": 6188857.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 5.894824504852295,
|
|
"epoch": 2.9843158464034616,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00045127062244728353,
|
|
"loss": 5.6525,
|
|
"mean_token_accuracy": 0.1678069680929184,
|
|
"num_tokens": 6199945.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 5.790930557250976,
|
|
"epoch": 2.989724175229854,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004510037350819055,
|
|
"loss": 5.6274,
|
|
"mean_token_accuracy": 0.17852601408958435,
|
|
"num_tokens": 6209866.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 5.840263795852661,
|
|
"epoch": 2.9951325040562464,
|
|
"grad_norm": 0.73828125,
|
|
"learning_rate": 0.00045073620811854744,
|
|
"loss": 5.601,
|
|
"mean_token_accuracy": 0.1708589166402817,
|
|
"num_tokens": 6221636.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 5.786147753397624,
|
|
"epoch": 3.0,
|
|
"grad_norm": 1.890625,
|
|
"learning_rate": 0.00045046804252940294,
|
|
"loss": 5.4912,
|
|
"mean_token_accuracy": 0.17953668700324166,
|
|
"num_tokens": 6230667.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 5.775953054428101,
|
|
"epoch": 3.005408328826393,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.00045019923928898644,
|
|
"loss": 5.3137,
|
|
"mean_token_accuracy": 0.19720368832349777,
|
|
"num_tokens": 6242242.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 5.84576735496521,
|
|
"epoch": 3.0108166576527853,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004499297993741294,
|
|
"loss": 5.3426,
|
|
"mean_token_accuracy": 0.19166723936796187,
|
|
"num_tokens": 6253337.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 5.7105354309082035,
|
|
"epoch": 3.016224986479178,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004496597237639773,
|
|
"loss": 5.2496,
|
|
"mean_token_accuracy": 0.19138242602348327,
|
|
"num_tokens": 6264582.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 5.744516038894654,
|
|
"epoch": 3.0216333153055706,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.00044938901343998536,
|
|
"loss": 5.2704,
|
|
"mean_token_accuracy": 0.20015280097723007,
|
|
"num_tokens": 6276172.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 5.702597761154175,
|
|
"epoch": 3.0270416441319634,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004491176693859156,
|
|
"loss": 5.1496,
|
|
"mean_token_accuracy": 0.20553621798753738,
|
|
"num_tokens": 6286038.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 5.674466514587403,
|
|
"epoch": 3.032449972958356,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004488456925878329,
|
|
"loss": 5.2351,
|
|
"mean_token_accuracy": 0.19469217211008072,
|
|
"num_tokens": 6296546.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 5.761750411987305,
|
|
"epoch": 3.0378583017847487,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.00044857308403410163,
|
|
"loss": 5.3232,
|
|
"mean_token_accuracy": 0.19953671097755432,
|
|
"num_tokens": 6308001.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 5.761521577835083,
|
|
"epoch": 3.043266630611141,
|
|
"grad_norm": 0.765625,
|
|
"learning_rate": 0.00044829984471538183,
|
|
"loss": 5.3335,
|
|
"mean_token_accuracy": 0.19271712601184846,
|
|
"num_tokens": 6320896.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 5.754070854187011,
|
|
"epoch": 3.048674959437534,
|
|
"grad_norm": 0.78515625,
|
|
"learning_rate": 0.0004480259756246258,
|
|
"loss": 5.3847,
|
|
"mean_token_accuracy": 0.18835910707712172,
|
|
"num_tokens": 6332710.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 5.758888149261475,
|
|
"epoch": 3.0540832882639264,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00044775147775707465,
|
|
"loss": 5.2553,
|
|
"mean_token_accuracy": 0.19560375809669495,
|
|
"num_tokens": 6343563.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 5.734629487991333,
|
|
"epoch": 3.0594916170903192,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.000447476352110254,
|
|
"loss": 5.3814,
|
|
"mean_token_accuracy": 0.19167284071445465,
|
|
"num_tokens": 6354064.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 5.698854875564575,
|
|
"epoch": 3.0648999459167117,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.0004472005996839712,
|
|
"loss": 5.2934,
|
|
"mean_token_accuracy": 0.19390958547592163,
|
|
"num_tokens": 6366551.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 5.695221996307373,
|
|
"epoch": 3.0703082747431045,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004469242214803112,
|
|
"loss": 5.2232,
|
|
"mean_token_accuracy": 0.19430456012487413,
|
|
"num_tokens": 6377086.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 5.7700074195861815,
|
|
"epoch": 3.075716603569497,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004466472185036331,
|
|
"loss": 5.328,
|
|
"mean_token_accuracy": 0.1926795169711113,
|
|
"num_tokens": 6388795.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 5.757317638397216,
|
|
"epoch": 3.08112493239589,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004463695917605663,
|
|
"loss": 5.3364,
|
|
"mean_token_accuracy": 0.19379594773054123,
|
|
"num_tokens": 6400174.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 5.687938547134399,
|
|
"epoch": 3.086533261222282,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.000446091342260007,
|
|
"loss": 5.2597,
|
|
"mean_token_accuracy": 0.19438011348247528,
|
|
"num_tokens": 6411297.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 5.699021053314209,
|
|
"epoch": 3.091941590048675,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004458124710131145,
|
|
"loss": 5.2971,
|
|
"mean_token_accuracy": 0.1954619586467743,
|
|
"num_tokens": 6423487.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 5.821798658370971,
|
|
"epoch": 3.0973499188750675,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.00044553297903330766,
|
|
"loss": 5.5021,
|
|
"mean_token_accuracy": 0.18111297935247422,
|
|
"num_tokens": 6435460.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 5.81842942237854,
|
|
"epoch": 3.1027582477014604,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.00044525286733626085,
|
|
"loss": 5.4305,
|
|
"mean_token_accuracy": 0.19182252883911133,
|
|
"num_tokens": 6445664.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 5.742527818679809,
|
|
"epoch": 3.1081665765278528,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004449721369399008,
|
|
"loss": 5.2679,
|
|
"mean_token_accuracy": 0.19546485245227813,
|
|
"num_tokens": 6457182.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 5.655294704437256,
|
|
"epoch": 3.1135749053542456,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.00044469078886440227,
|
|
"loss": 5.1412,
|
|
"mean_token_accuracy": 0.20516275316476823,
|
|
"num_tokens": 6468378.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 5.666563653945923,
|
|
"epoch": 3.118983234180638,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.00044440882413218484,
|
|
"loss": 5.3755,
|
|
"mean_token_accuracy": 0.19009971916675567,
|
|
"num_tokens": 6479625.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 5.69419755935669,
|
|
"epoch": 3.124391563007031,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00044412624376790905,
|
|
"loss": 5.2755,
|
|
"mean_token_accuracy": 0.19653896391391754,
|
|
"num_tokens": 6490713.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 5.609663391113282,
|
|
"epoch": 3.1297998918334233,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0004438430487984726,
|
|
"loss": 5.3221,
|
|
"mean_token_accuracy": 0.18166651725769042,
|
|
"num_tokens": 6502161.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 5.712747716903687,
|
|
"epoch": 3.135208220659816,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004435592402530066,
|
|
"loss": 5.2324,
|
|
"mean_token_accuracy": 0.19632606953382492,
|
|
"num_tokens": 6513639.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 5.653345727920533,
|
|
"epoch": 3.1406165494862086,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.000443274819162872,
|
|
"loss": 5.2653,
|
|
"mean_token_accuracy": 0.19660175293684007,
|
|
"num_tokens": 6524893.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 5.592651271820069,
|
|
"epoch": 3.1460248783126015,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004429897865616557,
|
|
"loss": 5.2409,
|
|
"mean_token_accuracy": 0.19942384511232375,
|
|
"num_tokens": 6535797.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 5.70519790649414,
|
|
"epoch": 3.151433207138994,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0004427041434851668,
|
|
"loss": 5.3629,
|
|
"mean_token_accuracy": 0.1893122002482414,
|
|
"num_tokens": 6547117.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 5.791764926910401,
|
|
"epoch": 3.1568415359653867,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.00044241789097143293,
|
|
"loss": 5.3986,
|
|
"mean_token_accuracy": 0.19074662029743195,
|
|
"num_tokens": 6558771.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 5.658868408203125,
|
|
"epoch": 3.162249864791779,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.00044213103006069635,
|
|
"loss": 5.3344,
|
|
"mean_token_accuracy": 0.1919943004846573,
|
|
"num_tokens": 6569922.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 5.756223583221436,
|
|
"epoch": 3.167658193618172,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00044184356179541035,
|
|
"loss": 5.3583,
|
|
"mean_token_accuracy": 0.19234999716281892,
|
|
"num_tokens": 6580561.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 5.651391172409058,
|
|
"epoch": 3.1730665224445644,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0004415554872202352,
|
|
"loss": 5.1829,
|
|
"mean_token_accuracy": 0.19829604923725128,
|
|
"num_tokens": 6591636.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 5.701524543762207,
|
|
"epoch": 3.1784748512709573,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.00044126680738203466,
|
|
"loss": 5.4093,
|
|
"mean_token_accuracy": 0.1898488447070122,
|
|
"num_tokens": 6603736.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 5.842616081237793,
|
|
"epoch": 3.1838831800973497,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004409775233298718,
|
|
"loss": 5.3315,
|
|
"mean_token_accuracy": 0.19555475413799286,
|
|
"num_tokens": 6615415.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 5.66992416381836,
|
|
"epoch": 3.1892915089237426,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004406876361150056,
|
|
"loss": 5.3804,
|
|
"mean_token_accuracy": 0.1857382357120514,
|
|
"num_tokens": 6625970.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 5.7360669612884525,
|
|
"epoch": 3.194699837750135,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004403971467908869,
|
|
"loss": 5.3318,
|
|
"mean_token_accuracy": 0.18936026096343994,
|
|
"num_tokens": 6636326.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 5.660694217681884,
|
|
"epoch": 3.200108166576528,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.0004401060564131545,
|
|
"loss": 5.3429,
|
|
"mean_token_accuracy": 0.18854644447565078,
|
|
"num_tokens": 6646825.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 5.723381900787354,
|
|
"epoch": 3.2055164954029207,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004398143660396315,
|
|
"loss": 5.3293,
|
|
"mean_token_accuracy": 0.19304709732532502,
|
|
"num_tokens": 6657748.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 5.667957448959351,
|
|
"epoch": 3.210924824229313,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004395220767303215,
|
|
"loss": 5.2539,
|
|
"mean_token_accuracy": 0.1999385714530945,
|
|
"num_tokens": 6668422.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 5.734276485443115,
|
|
"epoch": 3.2163331530557056,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004392291895474045,
|
|
"loss": 5.3345,
|
|
"mean_token_accuracy": 0.18912182450294496,
|
|
"num_tokens": 6678898.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 5.777085161209106,
|
|
"epoch": 3.2217414818820984,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004389357055552331,
|
|
"loss": 5.3798,
|
|
"mean_token_accuracy": 0.1966140851378441,
|
|
"num_tokens": 6688691.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 5.637375545501709,
|
|
"epoch": 3.2271498107084913,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00043864162582032897,
|
|
"loss": 5.2633,
|
|
"mean_token_accuracy": 0.2004222169518471,
|
|
"num_tokens": 6700279.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 5.660527753829956,
|
|
"epoch": 3.2325581395348837,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004383469514113784,
|
|
"loss": 5.1931,
|
|
"mean_token_accuracy": 0.20055626332759857,
|
|
"num_tokens": 6710440.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 5.670192289352417,
|
|
"epoch": 3.2379664683612766,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004380516833992289,
|
|
"loss": 5.3683,
|
|
"mean_token_accuracy": 0.18947068601846695,
|
|
"num_tokens": 6722347.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 5.681786155700683,
|
|
"epoch": 3.243374797187669,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000437755822856885,
|
|
"loss": 5.4017,
|
|
"mean_token_accuracy": 0.1939360350370407,
|
|
"num_tokens": 6732936.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 3.243374797187669,
|
|
"eval_entropy": 5.564755200256001,
|
|
"eval_loss": 5.958420753479004,
|
|
"eval_mean_token_accuracy": 0.17200287141075188,
|
|
"eval_num_tokens": 6732936.0,
|
|
"eval_runtime": 3.5257,
|
|
"eval_samples_per_second": 996.964,
|
|
"eval_steps_per_second": 124.798,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"entropy": 5.69419264793396,
|
|
"epoch": 3.248783126014062,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004374593708595049,
|
|
"loss": 5.4094,
|
|
"mean_token_accuracy": 0.19353188574314117,
|
|
"num_tokens": 6744552.0,
|
|
"step": 3005
|
|
},
|
|
{
|
|
"entropy": 5.771547460556031,
|
|
"epoch": 3.2541914548404542,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00043716232848439555,
|
|
"loss": 5.3354,
|
|
"mean_token_accuracy": 0.19333928674459458,
|
|
"num_tokens": 6755749.0,
|
|
"step": 3010
|
|
},
|
|
{
|
|
"entropy": 5.671427249908447,
|
|
"epoch": 3.259599783666847,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00043686469681100985,
|
|
"loss": 5.264,
|
|
"mean_token_accuracy": 0.195198193192482,
|
|
"num_tokens": 6766119.0,
|
|
"step": 3015
|
|
},
|
|
{
|
|
"entropy": 5.613597774505616,
|
|
"epoch": 3.2650081124932395,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00043656647692094186,
|
|
"loss": 5.3204,
|
|
"mean_token_accuracy": 0.18876274228096007,
|
|
"num_tokens": 6776491.0,
|
|
"step": 3020
|
|
},
|
|
{
|
|
"entropy": 5.795355272293091,
|
|
"epoch": 3.2704164413196324,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004362676698979233,
|
|
"loss": 5.4191,
|
|
"mean_token_accuracy": 0.1886429339647293,
|
|
"num_tokens": 6788795.0,
|
|
"step": 3025
|
|
},
|
|
{
|
|
"entropy": 5.788425588607788,
|
|
"epoch": 3.275824770146025,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.00043596827682781974,
|
|
"loss": 5.4312,
|
|
"mean_token_accuracy": 0.19093211144208908,
|
|
"num_tokens": 6799706.0,
|
|
"step": 3030
|
|
},
|
|
{
|
|
"entropy": 5.713137865066528,
|
|
"epoch": 3.2812330989724177,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0004356682987986262,
|
|
"loss": 5.3456,
|
|
"mean_token_accuracy": 0.1885154128074646,
|
|
"num_tokens": 6812454.0,
|
|
"step": 3035
|
|
},
|
|
{
|
|
"entropy": 5.7139589309692385,
|
|
"epoch": 3.28664142779881,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004353677369004635,
|
|
"loss": 5.2795,
|
|
"mean_token_accuracy": 0.1961999014019966,
|
|
"num_tokens": 6823950.0,
|
|
"step": 3040
|
|
},
|
|
{
|
|
"entropy": 5.77326455116272,
|
|
"epoch": 3.292049756625203,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00043506659222557426,
|
|
"loss": 5.4954,
|
|
"mean_token_accuracy": 0.18418579697608947,
|
|
"num_tokens": 6835028.0,
|
|
"step": 3045
|
|
},
|
|
{
|
|
"entropy": 5.659905576705933,
|
|
"epoch": 3.2974580854515954,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.000434764865868319,
|
|
"loss": 5.3572,
|
|
"mean_token_accuracy": 0.1901393324136734,
|
|
"num_tokens": 6846316.0,
|
|
"step": 3050
|
|
},
|
|
{
|
|
"entropy": 5.65548620223999,
|
|
"epoch": 3.302866414277988,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000434462558925172,
|
|
"loss": 5.2931,
|
|
"mean_token_accuracy": 0.2012960657477379,
|
|
"num_tokens": 6856463.0,
|
|
"step": 3055
|
|
},
|
|
{
|
|
"entropy": 5.724829435348511,
|
|
"epoch": 3.3082747431043806,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.00043415967249471734,
|
|
"loss": 5.3243,
|
|
"mean_token_accuracy": 0.1975801855325699,
|
|
"num_tokens": 6867869.0,
|
|
"step": 3060
|
|
},
|
|
{
|
|
"entropy": 5.6754532814025875,
|
|
"epoch": 3.3136830719307735,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.000433856207677645,
|
|
"loss": 5.3017,
|
|
"mean_token_accuracy": 0.19414029717445375,
|
|
"num_tokens": 6878545.0,
|
|
"step": 3065
|
|
},
|
|
{
|
|
"entropy": 5.621609354019165,
|
|
"epoch": 3.319091400757166,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.000433552165576747,
|
|
"loss": 5.3517,
|
|
"mean_token_accuracy": 0.19075303226709367,
|
|
"num_tokens": 6889399.0,
|
|
"step": 3070
|
|
},
|
|
{
|
|
"entropy": 5.752235174179077,
|
|
"epoch": 3.3244997295835588,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00043324754729691275,
|
|
"loss": 5.3317,
|
|
"mean_token_accuracy": 0.1996409922838211,
|
|
"num_tokens": 6900428.0,
|
|
"step": 3075
|
|
},
|
|
{
|
|
"entropy": 5.820627021789551,
|
|
"epoch": 3.329908058409951,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.00043294235394512594,
|
|
"loss": 5.4489,
|
|
"mean_token_accuracy": 0.18546947836875916,
|
|
"num_tokens": 6911112.0,
|
|
"step": 3080
|
|
},
|
|
{
|
|
"entropy": 5.611655569076538,
|
|
"epoch": 3.335316387236344,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004326365866304599,
|
|
"loss": 5.2095,
|
|
"mean_token_accuracy": 0.2032056301832199,
|
|
"num_tokens": 6922383.0,
|
|
"step": 3085
|
|
},
|
|
{
|
|
"entropy": 5.669274568557739,
|
|
"epoch": 3.3407247160627365,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.0004323302464640738,
|
|
"loss": 5.3961,
|
|
"mean_token_accuracy": 0.18529124259948732,
|
|
"num_tokens": 6935153.0,
|
|
"step": 3090
|
|
},
|
|
{
|
|
"entropy": 5.8112133026123045,
|
|
"epoch": 3.3461330448891293,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.00043202333455920843,
|
|
"loss": 5.4514,
|
|
"mean_token_accuracy": 0.18703726530075074,
|
|
"num_tokens": 6946814.0,
|
|
"step": 3095
|
|
},
|
|
{
|
|
"entropy": 5.68588342666626,
|
|
"epoch": 3.3515413737155217,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004317158520311824,
|
|
"loss": 5.2956,
|
|
"mean_token_accuracy": 0.19330791085958482,
|
|
"num_tokens": 6957230.0,
|
|
"step": 3100
|
|
},
|
|
{
|
|
"entropy": 5.581646871566773,
|
|
"epoch": 3.3569497025419146,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004314077999973879,
|
|
"loss": 5.3052,
|
|
"mean_token_accuracy": 0.18773041367530824,
|
|
"num_tokens": 6968060.0,
|
|
"step": 3105
|
|
},
|
|
{
|
|
"entropy": 5.675249195098877,
|
|
"epoch": 3.362358031368307,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004310991795772869,
|
|
"loss": 5.2227,
|
|
"mean_token_accuracy": 0.19939267337322236,
|
|
"num_tokens": 6977782.0,
|
|
"step": 3110
|
|
},
|
|
{
|
|
"entropy": 5.698951387405396,
|
|
"epoch": 3.3677663601947,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004307899918924065,
|
|
"loss": 5.308,
|
|
"mean_token_accuracy": 0.19284514635801314,
|
|
"num_tokens": 6988252.0,
|
|
"step": 3115
|
|
},
|
|
{
|
|
"entropy": 5.711275100708008,
|
|
"epoch": 3.3731746890210923,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004304802380663356,
|
|
"loss": 5.3178,
|
|
"mean_token_accuracy": 0.19082191735506057,
|
|
"num_tokens": 6998523.0,
|
|
"step": 3120
|
|
},
|
|
{
|
|
"entropy": 5.6842108249664305,
|
|
"epoch": 3.378583017847485,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.00043016991922472015,
|
|
"loss": 5.3499,
|
|
"mean_token_accuracy": 0.18608995378017426,
|
|
"num_tokens": 7010058.0,
|
|
"step": 3125
|
|
},
|
|
{
|
|
"entropy": 5.721271753311157,
|
|
"epoch": 3.383991346673878,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.00042985903649525977,
|
|
"loss": 5.4426,
|
|
"mean_token_accuracy": 0.18295274376869203,
|
|
"num_tokens": 7021289.0,
|
|
"step": 3130
|
|
},
|
|
{
|
|
"entropy": 5.824668645858765,
|
|
"epoch": 3.3893996755002704,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00042954759100770275,
|
|
"loss": 5.4721,
|
|
"mean_token_accuracy": 0.18034785389900207,
|
|
"num_tokens": 7032649.0,
|
|
"step": 3135
|
|
},
|
|
{
|
|
"entropy": 5.740284824371338,
|
|
"epoch": 3.394808004326663,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00042923558389384303,
|
|
"loss": 5.3462,
|
|
"mean_token_accuracy": 0.19291579276323317,
|
|
"num_tokens": 7044120.0,
|
|
"step": 3140
|
|
},
|
|
{
|
|
"entropy": 5.675812435150147,
|
|
"epoch": 3.4002163331530557,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004289230162875149,
|
|
"loss": 5.3373,
|
|
"mean_token_accuracy": 0.1933383896946907,
|
|
"num_tokens": 7054526.0,
|
|
"step": 3145
|
|
},
|
|
{
|
|
"entropy": 5.553069257736206,
|
|
"epoch": 3.4056246619794486,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00042860988932458987,
|
|
"loss": 5.2047,
|
|
"mean_token_accuracy": 0.20529062896966935,
|
|
"num_tokens": 7064744.0,
|
|
"step": 3150
|
|
},
|
|
{
|
|
"entropy": 5.7465972900390625,
|
|
"epoch": 3.411032990805841,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.0004282962041429721,
|
|
"loss": 5.3288,
|
|
"mean_token_accuracy": 0.19247181862592697,
|
|
"num_tokens": 7076360.0,
|
|
"step": 3155
|
|
},
|
|
{
|
|
"entropy": 5.702290821075439,
|
|
"epoch": 3.4164413196322334,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00042798196188259427,
|
|
"loss": 5.3507,
|
|
"mean_token_accuracy": 0.197810398042202,
|
|
"num_tokens": 7088114.0,
|
|
"step": 3160
|
|
},
|
|
{
|
|
"entropy": 5.711839437484741,
|
|
"epoch": 3.4218496484586263,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.0004276671636854133,
|
|
"loss": 5.382,
|
|
"mean_token_accuracy": 0.18595191538333894,
|
|
"num_tokens": 7099125.0,
|
|
"step": 3165
|
|
},
|
|
{
|
|
"entropy": 5.714252567291259,
|
|
"epoch": 3.427257977285019,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.00042735181069540677,
|
|
"loss": 5.3241,
|
|
"mean_token_accuracy": 0.19867067486047746,
|
|
"num_tokens": 7109875.0,
|
|
"step": 3170
|
|
},
|
|
{
|
|
"entropy": 5.679116201400757,
|
|
"epoch": 3.4326663061114115,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.0004270359040585681,
|
|
"loss": 5.3546,
|
|
"mean_token_accuracy": 0.19345322698354722,
|
|
"num_tokens": 7121388.0,
|
|
"step": 3175
|
|
},
|
|
{
|
|
"entropy": 5.695175409317017,
|
|
"epoch": 3.4380746349378044,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00042671944492290277,
|
|
"loss": 5.3867,
|
|
"mean_token_accuracy": 0.18891119360923767,
|
|
"num_tokens": 7134161.0,
|
|
"step": 3180
|
|
},
|
|
{
|
|
"entropy": 5.680855226516724,
|
|
"epoch": 3.443482963764197,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004264024344384239,
|
|
"loss": 5.4019,
|
|
"mean_token_accuracy": 0.19104203283786775,
|
|
"num_tokens": 7146580.0,
|
|
"step": 3185
|
|
},
|
|
{
|
|
"entropy": 5.723768138885498,
|
|
"epoch": 3.4488912925905897,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00042608487375714847,
|
|
"loss": 5.3258,
|
|
"mean_token_accuracy": 0.18677957803010942,
|
|
"num_tokens": 7156969.0,
|
|
"step": 3190
|
|
},
|
|
{
|
|
"entropy": 5.632949686050415,
|
|
"epoch": 3.454299621416982,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00042576676403309265,
|
|
"loss": 5.2689,
|
|
"mean_token_accuracy": 0.2009991616010666,
|
|
"num_tokens": 7168687.0,
|
|
"step": 3195
|
|
},
|
|
{
|
|
"entropy": 5.720569324493408,
|
|
"epoch": 3.459707950243375,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004254481064222679,
|
|
"loss": 5.3618,
|
|
"mean_token_accuracy": 0.1992029070854187,
|
|
"num_tokens": 7179996.0,
|
|
"step": 3200
|
|
},
|
|
{
|
|
"entropy": 5.7404012203216555,
|
|
"epoch": 3.4651162790697674,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.00042512890208267677,
|
|
"loss": 5.3843,
|
|
"mean_token_accuracy": 0.1858539655804634,
|
|
"num_tokens": 7191214.0,
|
|
"step": 3205
|
|
},
|
|
{
|
|
"entropy": 5.71372799873352,
|
|
"epoch": 3.4705246078961602,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.00042480915217430853,
|
|
"loss": 5.4083,
|
|
"mean_token_accuracy": 0.1886027604341507,
|
|
"num_tokens": 7203491.0,
|
|
"step": 3210
|
|
},
|
|
{
|
|
"entropy": 5.724027013778686,
|
|
"epoch": 3.4759329367225527,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.00042448885785913507,
|
|
"loss": 5.3434,
|
|
"mean_token_accuracy": 0.18529798090457916,
|
|
"num_tokens": 7215018.0,
|
|
"step": 3215
|
|
},
|
|
{
|
|
"entropy": 5.675080108642578,
|
|
"epoch": 3.4813412655489455,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.00042416802030110677,
|
|
"loss": 5.3058,
|
|
"mean_token_accuracy": 0.19982714653015138,
|
|
"num_tokens": 7226209.0,
|
|
"step": 3220
|
|
},
|
|
{
|
|
"entropy": 5.681215381622314,
|
|
"epoch": 3.486749594375338,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.00042384664066614787,
|
|
"loss": 5.3055,
|
|
"mean_token_accuracy": 0.19416655153036116,
|
|
"num_tokens": 7236505.0,
|
|
"step": 3225
|
|
},
|
|
{
|
|
"entropy": 5.5543177127838135,
|
|
"epoch": 3.492157923201731,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.0004235247201221528,
|
|
"loss": 5.2026,
|
|
"mean_token_accuracy": 0.20201916843652726,
|
|
"num_tokens": 7247414.0,
|
|
"step": 3230
|
|
},
|
|
{
|
|
"entropy": 5.709533786773681,
|
|
"epoch": 3.497566252028123,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004232022598389816,
|
|
"loss": 5.414,
|
|
"mean_token_accuracy": 0.18973794132471083,
|
|
"num_tokens": 7258666.0,
|
|
"step": 3235
|
|
},
|
|
{
|
|
"entropy": 5.733735084533691,
|
|
"epoch": 3.502974580854516,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.0004228792609884557,
|
|
"loss": 5.3322,
|
|
"mean_token_accuracy": 0.19570078551769257,
|
|
"num_tokens": 7270201.0,
|
|
"step": 3240
|
|
},
|
|
{
|
|
"entropy": 5.686567735671997,
|
|
"epoch": 3.5083829096809085,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.00042255572474435364,
|
|
"loss": 5.3867,
|
|
"mean_token_accuracy": 0.18971844911575317,
|
|
"num_tokens": 7282254.0,
|
|
"step": 3245
|
|
},
|
|
{
|
|
"entropy": 5.726072883605957,
|
|
"epoch": 3.5137912385073014,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0004222316522824069,
|
|
"loss": 5.4209,
|
|
"mean_token_accuracy": 0.1875574067234993,
|
|
"num_tokens": 7294579.0,
|
|
"step": 3250
|
|
},
|
|
{
|
|
"entropy": 5.674247169494629,
|
|
"epoch": 3.5191995673336938,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004219070447802955,
|
|
"loss": 5.3811,
|
|
"mean_token_accuracy": 0.18960184305906297,
|
|
"num_tokens": 7305822.0,
|
|
"step": 3255
|
|
},
|
|
{
|
|
"entropy": 5.775310277938843,
|
|
"epoch": 3.5246078961600866,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.00042158190341764384,
|
|
"loss": 5.4021,
|
|
"mean_token_accuracy": 0.18817628771066666,
|
|
"num_tokens": 7317543.0,
|
|
"step": 3260
|
|
},
|
|
{
|
|
"entropy": 5.6781822681427006,
|
|
"epoch": 3.530016224986479,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00042125622937601645,
|
|
"loss": 5.3363,
|
|
"mean_token_accuracy": 0.1896511659026146,
|
|
"num_tokens": 7327804.0,
|
|
"step": 3265
|
|
},
|
|
{
|
|
"entropy": 5.694125604629517,
|
|
"epoch": 3.535424553812872,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004209300238389135,
|
|
"loss": 5.2936,
|
|
"mean_token_accuracy": 0.1985452502965927,
|
|
"num_tokens": 7337045.0,
|
|
"step": 3270
|
|
},
|
|
{
|
|
"entropy": 5.674904012680054,
|
|
"epoch": 3.5408328826392643,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004206032879917667,
|
|
"loss": 5.3897,
|
|
"mean_token_accuracy": 0.19581838101148605,
|
|
"num_tokens": 7348184.0,
|
|
"step": 3275
|
|
},
|
|
{
|
|
"entropy": 5.6754368305206295,
|
|
"epoch": 3.546241211465657,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004202760230219348,
|
|
"loss": 5.2863,
|
|
"mean_token_accuracy": 0.19495007544755935,
|
|
"num_tokens": 7359571.0,
|
|
"step": 3280
|
|
},
|
|
{
|
|
"entropy": 5.638453149795533,
|
|
"epoch": 3.5516495402920496,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00041994823011869955,
|
|
"loss": 5.3175,
|
|
"mean_token_accuracy": 0.19527029544115065,
|
|
"num_tokens": 7370429.0,
|
|
"step": 3285
|
|
},
|
|
{
|
|
"entropy": 5.66474928855896,
|
|
"epoch": 3.5570578691184425,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.00041961991047326116,
|
|
"loss": 5.2618,
|
|
"mean_token_accuracy": 0.19621153324842452,
|
|
"num_tokens": 7381569.0,
|
|
"step": 3290
|
|
},
|
|
{
|
|
"entropy": 5.630235958099365,
|
|
"epoch": 3.5624661979448353,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.0004192910652787338,
|
|
"loss": 5.2537,
|
|
"mean_token_accuracy": 0.19903503805398942,
|
|
"num_tokens": 7392746.0,
|
|
"step": 3295
|
|
},
|
|
{
|
|
"entropy": 5.699338436126709,
|
|
"epoch": 3.5678745267712277,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0004189616957301419,
|
|
"loss": 5.3764,
|
|
"mean_token_accuracy": 0.18029830902814864,
|
|
"num_tokens": 7405520.0,
|
|
"step": 3300
|
|
},
|
|
{
|
|
"entropy": 5.607436227798462,
|
|
"epoch": 3.57328285559762,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004186318030244149,
|
|
"loss": 5.3223,
|
|
"mean_token_accuracy": 0.19025534987449647,
|
|
"num_tokens": 7417788.0,
|
|
"step": 3305
|
|
},
|
|
{
|
|
"entropy": 5.680609178543091,
|
|
"epoch": 3.578691184424013,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00041830138836038383,
|
|
"loss": 5.4068,
|
|
"mean_token_accuracy": 0.1853346049785614,
|
|
"num_tokens": 7428912.0,
|
|
"step": 3310
|
|
},
|
|
{
|
|
"entropy": 5.7111168384552,
|
|
"epoch": 3.584099513250406,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004179704529387763,
|
|
"loss": 5.2737,
|
|
"mean_token_accuracy": 0.19536608308553696,
|
|
"num_tokens": 7439797.0,
|
|
"step": 3315
|
|
},
|
|
{
|
|
"entropy": 5.596440601348877,
|
|
"epoch": 3.5895078420767983,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004176389979622125,
|
|
"loss": 5.2704,
|
|
"mean_token_accuracy": 0.19136498868465424,
|
|
"num_tokens": 7451755.0,
|
|
"step": 3320
|
|
},
|
|
{
|
|
"entropy": 5.7032252788543705,
|
|
"epoch": 3.5949161709031907,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004173070246352003,
|
|
"loss": 5.3657,
|
|
"mean_token_accuracy": 0.19019594490528108,
|
|
"num_tokens": 7463393.0,
|
|
"step": 3325
|
|
},
|
|
{
|
|
"entropy": 5.645083808898926,
|
|
"epoch": 3.6003244997295836,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00041697453416413184,
|
|
"loss": 5.3032,
|
|
"mean_token_accuracy": 0.19092003405094146,
|
|
"num_tokens": 7474404.0,
|
|
"step": 3330
|
|
},
|
|
{
|
|
"entropy": 5.614504718780518,
|
|
"epoch": 3.6057328285559764,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.000416641527757278,
|
|
"loss": 5.3051,
|
|
"mean_token_accuracy": 0.1955208122730255,
|
|
"num_tokens": 7485495.0,
|
|
"step": 3335
|
|
},
|
|
{
|
|
"entropy": 5.602577114105225,
|
|
"epoch": 3.611141157382369,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00041630800662478514,
|
|
"loss": 5.2578,
|
|
"mean_token_accuracy": 0.1999647542834282,
|
|
"num_tokens": 7497305.0,
|
|
"step": 3340
|
|
},
|
|
{
|
|
"entropy": 5.642804288864136,
|
|
"epoch": 3.6165494862087613,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.00041597397197866957,
|
|
"loss": 5.2042,
|
|
"mean_token_accuracy": 0.2022366017103195,
|
|
"num_tokens": 7509008.0,
|
|
"step": 3345
|
|
},
|
|
{
|
|
"entropy": 5.56644115447998,
|
|
"epoch": 3.621957815035154,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004156394250328141,
|
|
"loss": 5.288,
|
|
"mean_token_accuracy": 0.19770007729530334,
|
|
"num_tokens": 7520079.0,
|
|
"step": 3350
|
|
},
|
|
{
|
|
"entropy": 5.598038148880005,
|
|
"epoch": 3.627366143861547,
|
|
"grad_norm": 0.80859375,
|
|
"learning_rate": 0.000415304367002963,
|
|
"loss": 5.29,
|
|
"mean_token_accuracy": 0.19738789498806,
|
|
"num_tokens": 7531294.0,
|
|
"step": 3355
|
|
},
|
|
{
|
|
"entropy": 5.634009170532226,
|
|
"epoch": 3.6327744726879394,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.00041496879910671796,
|
|
"loss": 5.3231,
|
|
"mean_token_accuracy": 0.1974948063492775,
|
|
"num_tokens": 7542508.0,
|
|
"step": 3360
|
|
},
|
|
{
|
|
"entropy": 5.683190965652466,
|
|
"epoch": 3.638182801514332,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004146327225635336,
|
|
"loss": 5.2837,
|
|
"mean_token_accuracy": 0.19979180991649628,
|
|
"num_tokens": 7553288.0,
|
|
"step": 3365
|
|
},
|
|
{
|
|
"entropy": 5.681000757217407,
|
|
"epoch": 3.6435911303407247,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004142961385947127,
|
|
"loss": 5.4655,
|
|
"mean_token_accuracy": 0.19177876263856888,
|
|
"num_tokens": 7564532.0,
|
|
"step": 3370
|
|
},
|
|
{
|
|
"entropy": 5.6698637962341305,
|
|
"epoch": 3.6489994591671175,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.00041395904842340215,
|
|
"loss": 5.2805,
|
|
"mean_token_accuracy": 0.1996176928281784,
|
|
"num_tokens": 7575109.0,
|
|
"step": 3375
|
|
},
|
|
{
|
|
"entropy": 5.593832540512085,
|
|
"epoch": 3.65440778799351,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0004136214532745887,
|
|
"loss": 5.2754,
|
|
"mean_token_accuracy": 0.1988567069172859,
|
|
"num_tokens": 7585871.0,
|
|
"step": 3380
|
|
},
|
|
{
|
|
"entropy": 5.717426919937134,
|
|
"epoch": 3.6598161168199024,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004132833543750936,
|
|
"loss": 5.3678,
|
|
"mean_token_accuracy": 0.19920488446950912,
|
|
"num_tokens": 7596821.0,
|
|
"step": 3385
|
|
},
|
|
{
|
|
"entropy": 5.580592441558838,
|
|
"epoch": 3.6652244456462952,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00041294475295356917,
|
|
"loss": 5.2927,
|
|
"mean_token_accuracy": 0.1922448143362999,
|
|
"num_tokens": 7608101.0,
|
|
"step": 3390
|
|
},
|
|
{
|
|
"entropy": 5.6851887702941895,
|
|
"epoch": 3.670632774472688,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004126056502404937,
|
|
"loss": 5.2963,
|
|
"mean_token_accuracy": 0.1860350713133812,
|
|
"num_tokens": 7618722.0,
|
|
"step": 3395
|
|
},
|
|
{
|
|
"entropy": 5.605044364929199,
|
|
"epoch": 3.6760411032990805,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004122660474681672,
|
|
"loss": 5.306,
|
|
"mean_token_accuracy": 0.19814881831407546,
|
|
"num_tokens": 7630061.0,
|
|
"step": 3400
|
|
},
|
|
{
|
|
"entropy": 5.663322687149048,
|
|
"epoch": 3.6814494321254734,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.00041192594587070706,
|
|
"loss": 5.3021,
|
|
"mean_token_accuracy": 0.19622568786144257,
|
|
"num_tokens": 7641830.0,
|
|
"step": 3405
|
|
},
|
|
{
|
|
"entropy": 5.652032899856567,
|
|
"epoch": 3.686857760951866,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.00041158534668404325,
|
|
"loss": 5.3313,
|
|
"mean_token_accuracy": 0.1957012414932251,
|
|
"num_tokens": 7652750.0,
|
|
"step": 3410
|
|
},
|
|
{
|
|
"entropy": 5.595523262023926,
|
|
"epoch": 3.6922660897782587,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.00041124425114591386,
|
|
"loss": 5.2408,
|
|
"mean_token_accuracy": 0.20421086102724076,
|
|
"num_tokens": 7663405.0,
|
|
"step": 3415
|
|
},
|
|
{
|
|
"entropy": 5.53755521774292,
|
|
"epoch": 3.697674418604651,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.00041090266049586107,
|
|
"loss": 5.2454,
|
|
"mean_token_accuracy": 0.19019141644239426,
|
|
"num_tokens": 7673958.0,
|
|
"step": 3420
|
|
},
|
|
{
|
|
"entropy": 5.595367527008056,
|
|
"epoch": 3.703082747431044,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.0004105605759752258,
|
|
"loss": 5.2498,
|
|
"mean_token_accuracy": 0.19534795880317687,
|
|
"num_tokens": 7685475.0,
|
|
"step": 3425
|
|
},
|
|
{
|
|
"entropy": 5.550297975540161,
|
|
"epoch": 3.7084910762574363,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00041021799882714436,
|
|
"loss": 5.2423,
|
|
"mean_token_accuracy": 0.19693287312984467,
|
|
"num_tokens": 7697296.0,
|
|
"step": 3430
|
|
},
|
|
{
|
|
"entropy": 5.658037567138672,
|
|
"epoch": 3.713899405083829,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004098749302965426,
|
|
"loss": 5.3877,
|
|
"mean_token_accuracy": 0.19491476863622664,
|
|
"num_tokens": 7709414.0,
|
|
"step": 3435
|
|
},
|
|
{
|
|
"entropy": 5.672684621810913,
|
|
"epoch": 3.7193077339102216,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004095313716301324,
|
|
"loss": 5.3649,
|
|
"mean_token_accuracy": 0.18477234989404678,
|
|
"num_tokens": 7720965.0,
|
|
"step": 3440
|
|
},
|
|
{
|
|
"entropy": 5.627675580978393,
|
|
"epoch": 3.7247160627366145,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004091873240764069,
|
|
"loss": 5.3158,
|
|
"mean_token_accuracy": 0.18891929984092712,
|
|
"num_tokens": 7733044.0,
|
|
"step": 3445
|
|
},
|
|
{
|
|
"entropy": 5.743028783798218,
|
|
"epoch": 3.730124391563007,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00040884278888563566,
|
|
"loss": 5.3172,
|
|
"mean_token_accuracy": 0.1957472950220108,
|
|
"num_tokens": 7743878.0,
|
|
"step": 3450
|
|
},
|
|
{
|
|
"entropy": 5.59826340675354,
|
|
"epoch": 3.7355327203893998,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00040849776730986014,
|
|
"loss": 5.3012,
|
|
"mean_token_accuracy": 0.1986134171485901,
|
|
"num_tokens": 7755445.0,
|
|
"step": 3455
|
|
},
|
|
{
|
|
"entropy": 5.688895797729492,
|
|
"epoch": 3.740941049215792,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004081522606028899,
|
|
"loss": 5.3763,
|
|
"mean_token_accuracy": 0.18677901327610016,
|
|
"num_tokens": 7765990.0,
|
|
"step": 3460
|
|
},
|
|
{
|
|
"entropy": 5.577515697479248,
|
|
"epoch": 3.746349378042185,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 0.00040780627002029674,
|
|
"loss": 5.1684,
|
|
"mean_token_accuracy": 0.20449962615966796,
|
|
"num_tokens": 7777064.0,
|
|
"step": 3465
|
|
},
|
|
{
|
|
"entropy": 5.500732517242431,
|
|
"epoch": 3.7517577068685775,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004074597968194116,
|
|
"loss": 5.1971,
|
|
"mean_token_accuracy": 0.19851535111665725,
|
|
"num_tokens": 7788618.0,
|
|
"step": 3470
|
|
},
|
|
{
|
|
"entropy": 5.651279878616333,
|
|
"epoch": 3.7571660356949703,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004071128422593186,
|
|
"loss": 5.3285,
|
|
"mean_token_accuracy": 0.1883489266037941,
|
|
"num_tokens": 7798484.0,
|
|
"step": 3475
|
|
},
|
|
{
|
|
"entropy": 5.662864398956299,
|
|
"epoch": 3.7625743645213627,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004067654076008515,
|
|
"loss": 5.2988,
|
|
"mean_token_accuracy": 0.19100126326084138,
|
|
"num_tokens": 7810270.0,
|
|
"step": 3480
|
|
},
|
|
{
|
|
"entropy": 5.569951772689819,
|
|
"epoch": 3.7679826933477556,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.00040641749410658877,
|
|
"loss": 5.3869,
|
|
"mean_token_accuracy": 0.18685653656721116,
|
|
"num_tokens": 7822616.0,
|
|
"step": 3485
|
|
},
|
|
{
|
|
"entropy": 5.70475025177002,
|
|
"epoch": 3.773391022174148,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004060691030408487,
|
|
"loss": 5.3052,
|
|
"mean_token_accuracy": 0.19388724565505983,
|
|
"num_tokens": 7834933.0,
|
|
"step": 3490
|
|
},
|
|
{
|
|
"entropy": 5.707274723052978,
|
|
"epoch": 3.778799351000541,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00040572023566968536,
|
|
"loss": 5.339,
|
|
"mean_token_accuracy": 0.19112299680709838,
|
|
"num_tokens": 7847686.0,
|
|
"step": 3495
|
|
},
|
|
{
|
|
"entropy": 5.60497555732727,
|
|
"epoch": 3.7842076798269337,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.00040537089326088345,
|
|
"loss": 5.2943,
|
|
"mean_token_accuracy": 0.19179562628269195,
|
|
"num_tokens": 7859087.0,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"epoch": 3.7842076798269337,
|
|
"eval_entropy": 5.523230028152466,
|
|
"eval_loss": 5.890049934387207,
|
|
"eval_mean_token_accuracy": 0.1753515634601089,
|
|
"eval_num_tokens": 7859087.0,
|
|
"eval_runtime": 3.5412,
|
|
"eval_samples_per_second": 992.606,
|
|
"eval_steps_per_second": 124.252,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"entropy": 5.651162815093994,
|
|
"epoch": 3.789616008653326,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004050210770839541,
|
|
"loss": 5.3867,
|
|
"mean_token_accuracy": 0.1891338646411896,
|
|
"num_tokens": 7870684.0,
|
|
"step": 3505
|
|
},
|
|
{
|
|
"entropy": 5.663752794265747,
|
|
"epoch": 3.7950243374797186,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004046707884101303,
|
|
"loss": 5.2793,
|
|
"mean_token_accuracy": 0.19717731028795243,
|
|
"num_tokens": 7880017.0,
|
|
"step": 3510
|
|
},
|
|
{
|
|
"entropy": 5.494886875152588,
|
|
"epoch": 3.8004326663061114,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00040432002851236164,
|
|
"loss": 5.2312,
|
|
"mean_token_accuracy": 0.1946556493639946,
|
|
"num_tokens": 7891751.0,
|
|
"step": 3515
|
|
},
|
|
{
|
|
"entropy": 5.730977964401245,
|
|
"epoch": 3.8058409951325043,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004039687986653106,
|
|
"loss": 5.3718,
|
|
"mean_token_accuracy": 0.19073515385389328,
|
|
"num_tokens": 7902602.0,
|
|
"step": 3520
|
|
},
|
|
{
|
|
"entropy": 5.607116603851319,
|
|
"epoch": 3.8112493239588967,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00040361710014534714,
|
|
"loss": 5.2796,
|
|
"mean_token_accuracy": 0.19078059792518615,
|
|
"num_tokens": 7913548.0,
|
|
"step": 3525
|
|
},
|
|
{
|
|
"entropy": 5.673289680480957,
|
|
"epoch": 3.816657652785289,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004032649342305445,
|
|
"loss": 5.2957,
|
|
"mean_token_accuracy": 0.19233772307634353,
|
|
"num_tokens": 7925042.0,
|
|
"step": 3530
|
|
},
|
|
{
|
|
"entropy": 5.711692953109742,
|
|
"epoch": 3.822065981611682,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004029123022006745,
|
|
"loss": 5.3335,
|
|
"mean_token_accuracy": 0.18915214389562607,
|
|
"num_tokens": 7937113.0,
|
|
"step": 3535
|
|
},
|
|
{
|
|
"entropy": 5.652947950363159,
|
|
"epoch": 3.827474310438075,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004025592053372027,
|
|
"loss": 5.2694,
|
|
"mean_token_accuracy": 0.2026963084936142,
|
|
"num_tokens": 7947796.0,
|
|
"step": 3540
|
|
},
|
|
{
|
|
"entropy": 5.623374080657959,
|
|
"epoch": 3.8328826392644673,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004022056449232839,
|
|
"loss": 5.2995,
|
|
"mean_token_accuracy": 0.19341660887002946,
|
|
"num_tokens": 7959642.0,
|
|
"step": 3545
|
|
},
|
|
{
|
|
"entropy": 5.640952634811401,
|
|
"epoch": 3.8382909680908597,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004018516222437575,
|
|
"loss": 5.2798,
|
|
"mean_token_accuracy": 0.19453998208045958,
|
|
"num_tokens": 7969996.0,
|
|
"step": 3550
|
|
},
|
|
{
|
|
"entropy": 5.581825494766235,
|
|
"epoch": 3.8436992969172525,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004014971385851428,
|
|
"loss": 5.2408,
|
|
"mean_token_accuracy": 0.19880991876125337,
|
|
"num_tokens": 7980079.0,
|
|
"step": 3555
|
|
},
|
|
{
|
|
"entropy": 5.623220825195313,
|
|
"epoch": 3.8491076257436454,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.000401142195235634,
|
|
"loss": 5.325,
|
|
"mean_token_accuracy": 0.19438304305076598,
|
|
"num_tokens": 7990453.0,
|
|
"step": 3560
|
|
},
|
|
{
|
|
"entropy": 5.727757740020752,
|
|
"epoch": 3.854515954570038,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004007867934850963,
|
|
"loss": 5.3987,
|
|
"mean_token_accuracy": 0.1930052548646927,
|
|
"num_tokens": 8001810.0,
|
|
"step": 3565
|
|
},
|
|
{
|
|
"entropy": 5.698228025436402,
|
|
"epoch": 3.8599242833964302,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004004309346250603,
|
|
"loss": 5.3031,
|
|
"mean_token_accuracy": 0.20141524076461792,
|
|
"num_tokens": 8011821.0,
|
|
"step": 3570
|
|
},
|
|
{
|
|
"entropy": 5.57270450592041,
|
|
"epoch": 3.865332612222823,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0004000746199487181,
|
|
"loss": 5.2581,
|
|
"mean_token_accuracy": 0.2031795024871826,
|
|
"num_tokens": 8024106.0,
|
|
"step": 3575
|
|
},
|
|
{
|
|
"entropy": 5.552900791168213,
|
|
"epoch": 3.870740941049216,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.000399717850750918,
|
|
"loss": 5.2672,
|
|
"mean_token_accuracy": 0.1990682989358902,
|
|
"num_tokens": 8035630.0,
|
|
"step": 3580
|
|
},
|
|
{
|
|
"entropy": 5.662913036346436,
|
|
"epoch": 3.8761492698756084,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.00039936062832816,
|
|
"loss": 5.312,
|
|
"mean_token_accuracy": 0.19794627726078035,
|
|
"num_tokens": 8047130.0,
|
|
"step": 3585
|
|
},
|
|
{
|
|
"entropy": 5.587694454193115,
|
|
"epoch": 3.8815575987020012,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0003990029539785913,
|
|
"loss": 5.2595,
|
|
"mean_token_accuracy": 0.19099850654602052,
|
|
"num_tokens": 8059484.0,
|
|
"step": 3590
|
|
},
|
|
{
|
|
"entropy": 5.670936346054077,
|
|
"epoch": 3.8869659275283936,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00039864482900200135,
|
|
"loss": 5.363,
|
|
"mean_token_accuracy": 0.1892279639840126,
|
|
"num_tokens": 8069467.0,
|
|
"step": 3595
|
|
},
|
|
{
|
|
"entropy": 5.741135978698731,
|
|
"epoch": 3.8923742563547865,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0003982862546998169,
|
|
"loss": 5.364,
|
|
"mean_token_accuracy": 0.19609155356884003,
|
|
"num_tokens": 8081390.0,
|
|
"step": 3600
|
|
},
|
|
{
|
|
"entropy": 5.568446207046509,
|
|
"epoch": 3.897782585181179,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0003979272323750981,
|
|
"loss": 5.176,
|
|
"mean_token_accuracy": 0.2089977264404297,
|
|
"num_tokens": 8091202.0,
|
|
"step": 3605
|
|
},
|
|
{
|
|
"entropy": 5.581784439086914,
|
|
"epoch": 3.903190914007572,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0003975677633325327,
|
|
"loss": 5.291,
|
|
"mean_token_accuracy": 0.1949166789650917,
|
|
"num_tokens": 8101900.0,
|
|
"step": 3610
|
|
},
|
|
{
|
|
"entropy": 5.627234554290771,
|
|
"epoch": 3.908599242833964,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0003972078488784321,
|
|
"loss": 5.3106,
|
|
"mean_token_accuracy": 0.18876059502363204,
|
|
"num_tokens": 8114500.0,
|
|
"step": 3615
|
|
},
|
|
{
|
|
"entropy": 5.599054527282715,
|
|
"epoch": 3.914007571660357,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0003968474903207263,
|
|
"loss": 5.1881,
|
|
"mean_token_accuracy": 0.2103382408618927,
|
|
"num_tokens": 8125319.0,
|
|
"step": 3620
|
|
},
|
|
{
|
|
"entropy": 5.667600011825561,
|
|
"epoch": 3.9194159004867495,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0003964866889689592,
|
|
"loss": 5.3933,
|
|
"mean_token_accuracy": 0.19413406401872635,
|
|
"num_tokens": 8137077.0,
|
|
"step": 3625
|
|
},
|
|
{
|
|
"entropy": 5.561121892929077,
|
|
"epoch": 3.9248242293131423,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0003961254461342834,
|
|
"loss": 5.2379,
|
|
"mean_token_accuracy": 0.19667590111494065,
|
|
"num_tokens": 8148430.0,
|
|
"step": 3630
|
|
},
|
|
{
|
|
"entropy": 5.592974662780762,
|
|
"epoch": 3.9302325581395348,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.0003957637631294566,
|
|
"loss": 5.2628,
|
|
"mean_token_accuracy": 0.19850500375032426,
|
|
"num_tokens": 8159250.0,
|
|
"step": 3635
|
|
},
|
|
{
|
|
"entropy": 5.667021369934082,
|
|
"epoch": 3.9356408869659276,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00039540164126883554,
|
|
"loss": 5.3575,
|
|
"mean_token_accuracy": 0.1913460075855255,
|
|
"num_tokens": 8170177.0,
|
|
"step": 3640
|
|
},
|
|
{
|
|
"entropy": 5.6899487495422365,
|
|
"epoch": 3.94104921579232,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0003950390818683718,
|
|
"loss": 5.2979,
|
|
"mean_token_accuracy": 0.19830988496541976,
|
|
"num_tokens": 8181351.0,
|
|
"step": 3645
|
|
},
|
|
{
|
|
"entropy": 5.598061227798462,
|
|
"epoch": 3.946457544618713,
|
|
"grad_norm": 0.76953125,
|
|
"learning_rate": 0.0003946760862456071,
|
|
"loss": 5.3059,
|
|
"mean_token_accuracy": 0.19724066853523253,
|
|
"num_tokens": 8193472.0,
|
|
"step": 3650
|
|
},
|
|
{
|
|
"entropy": 5.636629962921143,
|
|
"epoch": 3.9518658734451053,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0003943126557196685,
|
|
"loss": 5.4095,
|
|
"mean_token_accuracy": 0.20111747235059738,
|
|
"num_tokens": 8205449.0,
|
|
"step": 3655
|
|
},
|
|
{
|
|
"entropy": 5.554954242706299,
|
|
"epoch": 3.957274202271498,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.00039394879161126333,
|
|
"loss": 5.1324,
|
|
"mean_token_accuracy": 0.2015075147151947,
|
|
"num_tokens": 8216520.0,
|
|
"step": 3660
|
|
},
|
|
{
|
|
"entropy": 5.64744815826416,
|
|
"epoch": 3.9626825310978906,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0003935844952426745,
|
|
"loss": 5.3717,
|
|
"mean_token_accuracy": 0.18534173518419267,
|
|
"num_tokens": 8228727.0,
|
|
"step": 3665
|
|
},
|
|
{
|
|
"entropy": 5.6714146614074705,
|
|
"epoch": 3.9680908599242835,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00039321976793775586,
|
|
"loss": 5.2687,
|
|
"mean_token_accuracy": 0.19557962864637374,
|
|
"num_tokens": 8240423.0,
|
|
"step": 3670
|
|
},
|
|
{
|
|
"entropy": 5.5574095249176025,
|
|
"epoch": 3.973499188750676,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0003928546110219275,
|
|
"loss": 5.2003,
|
|
"mean_token_accuracy": 0.19956401586532593,
|
|
"num_tokens": 8252757.0,
|
|
"step": 3675
|
|
},
|
|
{
|
|
"entropy": 5.443890190124511,
|
|
"epoch": 3.9789075175770687,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.00039248902582217024,
|
|
"loss": 5.1899,
|
|
"mean_token_accuracy": 0.198175610601902,
|
|
"num_tokens": 8263462.0,
|
|
"step": 3680
|
|
},
|
|
{
|
|
"entropy": 5.540591049194336,
|
|
"epoch": 3.9843158464034616,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00039212301366702153,
|
|
"loss": 5.2527,
|
|
"mean_token_accuracy": 0.19623720049858093,
|
|
"num_tokens": 8274364.0,
|
|
"step": 3685
|
|
},
|
|
{
|
|
"entropy": 5.638958740234375,
|
|
"epoch": 3.989724175229854,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.00039175657588657063,
|
|
"loss": 5.2757,
|
|
"mean_token_accuracy": 0.19483605027198792,
|
|
"num_tokens": 8286026.0,
|
|
"step": 3690
|
|
},
|
|
{
|
|
"entropy": 5.681791591644287,
|
|
"epoch": 3.9951325040562464,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0003913897138124531,
|
|
"loss": 5.3231,
|
|
"mean_token_accuracy": 0.19580954015254975,
|
|
"num_tokens": 8297990.0,
|
|
"step": 3695
|
|
},
|
|
{
|
|
"entropy": 5.670732233259413,
|
|
"epoch": 4.0,
|
|
"grad_norm": 2.0,
|
|
"learning_rate": 0.0003910224287778466,
|
|
"loss": 5.3577,
|
|
"mean_token_accuracy": 0.18696142733097076,
|
|
"num_tokens": 8307556.0,
|
|
"step": 3700
|
|
},
|
|
{
|
|
"entropy": 5.594685459136963,
|
|
"epoch": 4.005408328826393,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00039065472211746586,
|
|
"loss": 5.1159,
|
|
"mean_token_accuracy": 0.20600400269031524,
|
|
"num_tokens": 8320805.0,
|
|
"step": 3705
|
|
},
|
|
{
|
|
"entropy": 5.621187353134156,
|
|
"epoch": 4.010816657652786,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0003902865951675576,
|
|
"loss": 5.1022,
|
|
"mean_token_accuracy": 0.20499833077192306,
|
|
"num_tokens": 8332390.0,
|
|
"step": 3710
|
|
},
|
|
{
|
|
"entropy": 5.569980239868164,
|
|
"epoch": 4.016224986479178,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00038991804926589616,
|
|
"loss": 5.0016,
|
|
"mean_token_accuracy": 0.20818119943141938,
|
|
"num_tokens": 8342325.0,
|
|
"step": 3715
|
|
},
|
|
{
|
|
"entropy": 5.6370457172393795,
|
|
"epoch": 4.0216333153055706,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.000389549085751778,
|
|
"loss": 5.035,
|
|
"mean_token_accuracy": 0.21315674483776093,
|
|
"num_tokens": 8352115.0,
|
|
"step": 3720
|
|
},
|
|
{
|
|
"entropy": 5.6402758121490475,
|
|
"epoch": 4.027041644131963,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00038917970596601765,
|
|
"loss": 5.1069,
|
|
"mean_token_accuracy": 0.21209613382816314,
|
|
"num_tokens": 8364265.0,
|
|
"step": 3725
|
|
},
|
|
{
|
|
"entropy": 5.705170488357544,
|
|
"epoch": 4.032449972958356,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0003888099112509419,
|
|
"loss": 5.0779,
|
|
"mean_token_accuracy": 0.21012381911277772,
|
|
"num_tokens": 8375979.0,
|
|
"step": 3730
|
|
},
|
|
{
|
|
"entropy": 5.606539535522461,
|
|
"epoch": 4.037858301784748,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.00038843970295038564,
|
|
"loss": 5.053,
|
|
"mean_token_accuracy": 0.21184056252241135,
|
|
"num_tokens": 8387794.0,
|
|
"step": 3735
|
|
},
|
|
{
|
|
"entropy": 5.565236616134643,
|
|
"epoch": 4.043266630611141,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.00038806908240968666,
|
|
"loss": 4.9683,
|
|
"mean_token_accuracy": 0.21545963287353515,
|
|
"num_tokens": 8398796.0,
|
|
"step": 3740
|
|
},
|
|
{
|
|
"entropy": 5.570811939239502,
|
|
"epoch": 4.048674959437534,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.000387698050975681,
|
|
"loss": 5.0575,
|
|
"mean_token_accuracy": 0.21631220132112502,
|
|
"num_tokens": 8408788.0,
|
|
"step": 3745
|
|
},
|
|
{
|
|
"entropy": 5.640817546844483,
|
|
"epoch": 4.054083288263927,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0003873266099966976,
|
|
"loss": 5.077,
|
|
"mean_token_accuracy": 0.20749575942754744,
|
|
"num_tokens": 8420373.0,
|
|
"step": 3750
|
|
},
|
|
{
|
|
"entropy": 5.547187423706054,
|
|
"epoch": 4.059491617090319,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.00038695476082255395,
|
|
"loss": 5.0635,
|
|
"mean_token_accuracy": 0.21502292454242705,
|
|
"num_tokens": 8430399.0,
|
|
"step": 3755
|
|
},
|
|
{
|
|
"entropy": 5.638990688323974,
|
|
"epoch": 4.064899945916712,
|
|
"grad_norm": 0.81640625,
|
|
"learning_rate": 0.0003865825048045507,
|
|
"loss": 5.1343,
|
|
"mean_token_accuracy": 0.20479972064495086,
|
|
"num_tokens": 8442649.0,
|
|
"step": 3760
|
|
},
|
|
{
|
|
"entropy": 5.746016788482666,
|
|
"epoch": 4.0703082747431045,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0003862098432954672,
|
|
"loss": 5.1662,
|
|
"mean_token_accuracy": 0.20147386342287063,
|
|
"num_tokens": 8453985.0,
|
|
"step": 3765
|
|
},
|
|
{
|
|
"entropy": 5.646367692947388,
|
|
"epoch": 4.075716603569497,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0003858367776495561,
|
|
"loss": 5.1866,
|
|
"mean_token_accuracy": 0.20267382115125657,
|
|
"num_tokens": 8465566.0,
|
|
"step": 3770
|
|
},
|
|
{
|
|
"entropy": 5.561671733856201,
|
|
"epoch": 4.081124932395889,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00038546330922253886,
|
|
"loss": 4.9771,
|
|
"mean_token_accuracy": 0.22253454774618148,
|
|
"num_tokens": 8477426.0,
|
|
"step": 3775
|
|
},
|
|
{
|
|
"entropy": 5.550774335861206,
|
|
"epoch": 4.086533261222282,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0003850894393716007,
|
|
"loss": 5.0868,
|
|
"mean_token_accuracy": 0.20939560383558273,
|
|
"num_tokens": 8489654.0,
|
|
"step": 3780
|
|
},
|
|
{
|
|
"entropy": 5.518771600723267,
|
|
"epoch": 4.091941590048675,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0003847151694553855,
|
|
"loss": 5.0456,
|
|
"mean_token_accuracy": 0.2071119427680969,
|
|
"num_tokens": 8500398.0,
|
|
"step": 3785
|
|
},
|
|
{
|
|
"entropy": 5.551463937759399,
|
|
"epoch": 4.097349918875068,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0003843405008339908,
|
|
"loss": 4.9941,
|
|
"mean_token_accuracy": 0.20867300927639007,
|
|
"num_tokens": 8511974.0,
|
|
"step": 3790
|
|
},
|
|
{
|
|
"entropy": 5.594903993606567,
|
|
"epoch": 4.10275824770146,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00038396543486896354,
|
|
"loss": 5.0787,
|
|
"mean_token_accuracy": 0.21112921983003616,
|
|
"num_tokens": 8523140.0,
|
|
"step": 3795
|
|
},
|
|
{
|
|
"entropy": 5.536909532546997,
|
|
"epoch": 4.108166576527853,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.00038358997292329406,
|
|
"loss": 5.0614,
|
|
"mean_token_accuracy": 0.21375230997800826,
|
|
"num_tokens": 8534261.0,
|
|
"step": 3800
|
|
},
|
|
{
|
|
"entropy": 5.646812105178833,
|
|
"epoch": 4.113574905354246,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00038321411636141214,
|
|
"loss": 5.1343,
|
|
"mean_token_accuracy": 0.20292503237724305,
|
|
"num_tokens": 8544620.0,
|
|
"step": 3805
|
|
},
|
|
{
|
|
"entropy": 5.565939712524414,
|
|
"epoch": 4.1189832341806385,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0003828378665491811,
|
|
"loss": 5.086,
|
|
"mean_token_accuracy": 0.20964807122945786,
|
|
"num_tokens": 8555427.0,
|
|
"step": 3810
|
|
},
|
|
{
|
|
"entropy": 5.575354290008545,
|
|
"epoch": 4.1243915630070305,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.000382461224853894,
|
|
"loss": 5.0381,
|
|
"mean_token_accuracy": 0.21237004548311234,
|
|
"num_tokens": 8566956.0,
|
|
"step": 3815
|
|
},
|
|
{
|
|
"entropy": 5.618250799179077,
|
|
"epoch": 4.129799891833423,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0003820841926442674,
|
|
"loss": 5.2046,
|
|
"mean_token_accuracy": 0.20202476978302003,
|
|
"num_tokens": 8579373.0,
|
|
"step": 3820
|
|
},
|
|
{
|
|
"entropy": 5.526768350601197,
|
|
"epoch": 4.135208220659816,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0003817067712904372,
|
|
"loss": 5.0021,
|
|
"mean_token_accuracy": 0.2116176024079323,
|
|
"num_tokens": 8590979.0,
|
|
"step": 3825
|
|
},
|
|
{
|
|
"entropy": 5.543919229507447,
|
|
"epoch": 4.140616549486209,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00038132896216395355,
|
|
"loss": 5.0554,
|
|
"mean_token_accuracy": 0.20776492953300477,
|
|
"num_tokens": 8602514.0,
|
|
"step": 3830
|
|
},
|
|
{
|
|
"entropy": 5.5103600025177,
|
|
"epoch": 4.146024878312601,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003809507666377758,
|
|
"loss": 5.0071,
|
|
"mean_token_accuracy": 0.21537818014621735,
|
|
"num_tokens": 8613766.0,
|
|
"step": 3835
|
|
},
|
|
{
|
|
"entropy": 5.615996313095093,
|
|
"epoch": 4.151433207138994,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00038057218608626714,
|
|
"loss": 5.1436,
|
|
"mean_token_accuracy": 0.1966433897614479,
|
|
"num_tokens": 8625245.0,
|
|
"step": 3840
|
|
},
|
|
{
|
|
"entropy": 5.632936191558838,
|
|
"epoch": 4.156841535965387,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0003801932218851903,
|
|
"loss": 5.0766,
|
|
"mean_token_accuracy": 0.21580771505832672,
|
|
"num_tokens": 8635653.0,
|
|
"step": 3845
|
|
},
|
|
{
|
|
"entropy": 5.504722547531128,
|
|
"epoch": 4.16224986479178,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000379813875411702,
|
|
"loss": 5.0172,
|
|
"mean_token_accuracy": 0.20979683250188827,
|
|
"num_tokens": 8646671.0,
|
|
"step": 3850
|
|
},
|
|
{
|
|
"entropy": 5.506032943725586,
|
|
"epoch": 4.167658193618172,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00037943414804434826,
|
|
"loss": 4.9794,
|
|
"mean_token_accuracy": 0.21566243916749955,
|
|
"num_tokens": 8656441.0,
|
|
"step": 3855
|
|
},
|
|
{
|
|
"entropy": 5.541570949554443,
|
|
"epoch": 4.173066522444564,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0003790540411630592,
|
|
"loss": 5.0362,
|
|
"mean_token_accuracy": 0.21489110291004182,
|
|
"num_tokens": 8666632.0,
|
|
"step": 3860
|
|
},
|
|
{
|
|
"entropy": 5.5245226383209225,
|
|
"epoch": 4.178474851270957,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.00037867355614914404,
|
|
"loss": 5.1033,
|
|
"mean_token_accuracy": 0.21371882557868957,
|
|
"num_tokens": 8678269.0,
|
|
"step": 3865
|
|
},
|
|
{
|
|
"entropy": 5.582470369338989,
|
|
"epoch": 4.18388318009735,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0003782926943852864,
|
|
"loss": 5.1114,
|
|
"mean_token_accuracy": 0.20679743885993956,
|
|
"num_tokens": 8690004.0,
|
|
"step": 3870
|
|
},
|
|
{
|
|
"entropy": 5.634443712234497,
|
|
"epoch": 4.189291508923742,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00037791145725553853,
|
|
"loss": 5.139,
|
|
"mean_token_accuracy": 0.20164074897766113,
|
|
"num_tokens": 8702306.0,
|
|
"step": 3875
|
|
},
|
|
{
|
|
"entropy": 5.575648450851441,
|
|
"epoch": 4.194699837750135,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0003775298461453172,
|
|
"loss": 5.1502,
|
|
"mean_token_accuracy": 0.20574383735656737,
|
|
"num_tokens": 8714070.0,
|
|
"step": 3880
|
|
},
|
|
{
|
|
"entropy": 5.568425273895263,
|
|
"epoch": 4.200108166576528,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.000377147862441398,
|
|
"loss": 5.0829,
|
|
"mean_token_accuracy": 0.21182692646980286,
|
|
"num_tokens": 8724433.0,
|
|
"step": 3885
|
|
},
|
|
{
|
|
"entropy": 5.552147626876831,
|
|
"epoch": 4.205516495402921,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00037676550753191075,
|
|
"loss": 5.0323,
|
|
"mean_token_accuracy": 0.2166525885462761,
|
|
"num_tokens": 8735485.0,
|
|
"step": 3890
|
|
},
|
|
{
|
|
"entropy": 5.55499906539917,
|
|
"epoch": 4.210924824229313,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0003763827828063339,
|
|
"loss": 5.0562,
|
|
"mean_token_accuracy": 0.19902590066194534,
|
|
"num_tokens": 8746356.0,
|
|
"step": 3895
|
|
},
|
|
{
|
|
"entropy": 5.5377504348754885,
|
|
"epoch": 4.2163331530557056,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00037599968965549,
|
|
"loss": 5.0751,
|
|
"mean_token_accuracy": 0.20463165044784545,
|
|
"num_tokens": 8757667.0,
|
|
"step": 3900
|
|
},
|
|
{
|
|
"entropy": 5.524814605712891,
|
|
"epoch": 4.221741481882098,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0003756162294715405,
|
|
"loss": 5.0316,
|
|
"mean_token_accuracy": 0.21452374905347824,
|
|
"num_tokens": 8768565.0,
|
|
"step": 3905
|
|
},
|
|
{
|
|
"entropy": 5.509193229675293,
|
|
"epoch": 4.227149810708491,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.00037523240364798063,
|
|
"loss": 5.0591,
|
|
"mean_token_accuracy": 0.21244459748268127,
|
|
"num_tokens": 8779608.0,
|
|
"step": 3910
|
|
},
|
|
{
|
|
"entropy": 5.587919759750366,
|
|
"epoch": 4.232558139534884,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.00037484821357963423,
|
|
"loss": 5.1131,
|
|
"mean_token_accuracy": 0.2109584853053093,
|
|
"num_tokens": 8791492.0,
|
|
"step": 3915
|
|
},
|
|
{
|
|
"entropy": 5.54430832862854,
|
|
"epoch": 4.237966468361276,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.000374463660662649,
|
|
"loss": 5.0908,
|
|
"mean_token_accuracy": 0.2086695611476898,
|
|
"num_tokens": 8803280.0,
|
|
"step": 3920
|
|
},
|
|
{
|
|
"entropy": 5.554611492156982,
|
|
"epoch": 4.243374797187669,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.000374078746294491,
|
|
"loss": 5.1382,
|
|
"mean_token_accuracy": 0.20818226635456086,
|
|
"num_tokens": 8814626.0,
|
|
"step": 3925
|
|
},
|
|
{
|
|
"entropy": 5.5655842304229735,
|
|
"epoch": 4.248783126014062,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00037369347187393996,
|
|
"loss": 4.9754,
|
|
"mean_token_accuracy": 0.22211375534534455,
|
|
"num_tokens": 8825520.0,
|
|
"step": 3930
|
|
},
|
|
{
|
|
"entropy": 5.470966577529907,
|
|
"epoch": 4.254191454840455,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0003733078388010842,
|
|
"loss": 4.9957,
|
|
"mean_token_accuracy": 0.2146727666258812,
|
|
"num_tokens": 8835980.0,
|
|
"step": 3935
|
|
},
|
|
{
|
|
"entropy": 5.611467504501343,
|
|
"epoch": 4.259599783666847,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0003729218484773149,
|
|
"loss": 5.1723,
|
|
"mean_token_accuracy": 0.19957948327064515,
|
|
"num_tokens": 8846714.0,
|
|
"step": 3940
|
|
},
|
|
{
|
|
"entropy": 5.60156774520874,
|
|
"epoch": 4.2650081124932395,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003725355023053221,
|
|
"loss": 5.1058,
|
|
"mean_token_accuracy": 0.2070922315120697,
|
|
"num_tokens": 8858215.0,
|
|
"step": 3945
|
|
},
|
|
{
|
|
"entropy": 5.52307276725769,
|
|
"epoch": 4.270416441319632,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0003721488016890884,
|
|
"loss": 5.0119,
|
|
"mean_token_accuracy": 0.21522187143564225,
|
|
"num_tokens": 8870587.0,
|
|
"step": 3950
|
|
},
|
|
{
|
|
"entropy": 5.427456998825074,
|
|
"epoch": 4.275824770146025,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00037176174803388496,
|
|
"loss": 5.0259,
|
|
"mean_token_accuracy": 0.2161127045750618,
|
|
"num_tokens": 8881335.0,
|
|
"step": 3955
|
|
},
|
|
{
|
|
"entropy": 5.459418058395386,
|
|
"epoch": 4.281233098972417,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.00037137434274626545,
|
|
"loss": 5.026,
|
|
"mean_token_accuracy": 0.21049747616052628,
|
|
"num_tokens": 8893294.0,
|
|
"step": 3960
|
|
},
|
|
{
|
|
"entropy": 5.628967714309693,
|
|
"epoch": 4.28664142779881,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0003709865872340618,
|
|
"loss": 5.1275,
|
|
"mean_token_accuracy": 0.2062242344021797,
|
|
"num_tokens": 8903914.0,
|
|
"step": 3965
|
|
},
|
|
{
|
|
"entropy": 5.57897982597351,
|
|
"epoch": 4.292049756625203,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0003705984829063782,
|
|
"loss": 5.0929,
|
|
"mean_token_accuracy": 0.20352874249219893,
|
|
"num_tokens": 8915591.0,
|
|
"step": 3970
|
|
},
|
|
{
|
|
"entropy": 5.537923669815063,
|
|
"epoch": 4.297458085451596,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0003702100311735869,
|
|
"loss": 5.0834,
|
|
"mean_token_accuracy": 0.20173001438379287,
|
|
"num_tokens": 8927031.0,
|
|
"step": 3975
|
|
},
|
|
{
|
|
"entropy": 5.520787096023559,
|
|
"epoch": 4.302866414277988,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.00036982123344732223,
|
|
"loss": 5.02,
|
|
"mean_token_accuracy": 0.21347687095403672,
|
|
"num_tokens": 8937220.0,
|
|
"step": 3980
|
|
},
|
|
{
|
|
"entropy": 5.468257761001587,
|
|
"epoch": 4.308274743104381,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00036943209114047613,
|
|
"loss": 5.0216,
|
|
"mean_token_accuracy": 0.20727793425321578,
|
|
"num_tokens": 8948032.0,
|
|
"step": 3985
|
|
},
|
|
{
|
|
"entropy": 5.512820482254028,
|
|
"epoch": 4.3136830719307735,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0003690426056671926,
|
|
"loss": 5.1196,
|
|
"mean_token_accuracy": 0.2027975559234619,
|
|
"num_tokens": 8960735.0,
|
|
"step": 3990
|
|
},
|
|
{
|
|
"entropy": 5.467561721801758,
|
|
"epoch": 4.319091400757166,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.0003686527784428627,
|
|
"loss": 4.9983,
|
|
"mean_token_accuracy": 0.21924445480108262,
|
|
"num_tokens": 8971801.0,
|
|
"step": 3995
|
|
},
|
|
{
|
|
"entropy": 5.584245300292968,
|
|
"epoch": 4.324499729583558,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00036826261088411955,
|
|
"loss": 5.1176,
|
|
"mean_token_accuracy": 0.20276290774345399,
|
|
"num_tokens": 8983041.0,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"epoch": 4.324499729583558,
|
|
"eval_entropy": 5.478987052223899,
|
|
"eval_loss": 5.882231712341309,
|
|
"eval_mean_token_accuracy": 0.17786235040561718,
|
|
"eval_num_tokens": 8983041.0,
|
|
"eval_runtime": 3.5299,
|
|
"eval_samples_per_second": 995.782,
|
|
"eval_steps_per_second": 124.65,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"entropy": 5.531061363220215,
|
|
"epoch": 4.329908058409951,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003678721044088329,
|
|
"loss": 5.0794,
|
|
"mean_token_accuracy": 0.20491101294755937,
|
|
"num_tokens": 8993935.0,
|
|
"step": 4005
|
|
},
|
|
{
|
|
"entropy": 5.514832782745361,
|
|
"epoch": 4.335316387236344,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0003674812604361043,
|
|
"loss": 5.0313,
|
|
"mean_token_accuracy": 0.20588540881872178,
|
|
"num_tokens": 9003896.0,
|
|
"step": 4010
|
|
},
|
|
{
|
|
"entropy": 5.559433460235596,
|
|
"epoch": 4.340724716062737,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0003670900803862615,
|
|
"loss": 5.1159,
|
|
"mean_token_accuracy": 0.21055326759815216,
|
|
"num_tokens": 9014323.0,
|
|
"step": 4015
|
|
},
|
|
{
|
|
"entropy": 5.556778240203857,
|
|
"epoch": 4.346133044889129,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0003666985656808537,
|
|
"loss": 5.1484,
|
|
"mean_token_accuracy": 0.21008240729570388,
|
|
"num_tokens": 9024252.0,
|
|
"step": 4020
|
|
},
|
|
{
|
|
"entropy": 5.585116338729859,
|
|
"epoch": 4.351541373715522,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00036630671774264635,
|
|
"loss": 5.1581,
|
|
"mean_token_accuracy": 0.20255700051784514,
|
|
"num_tokens": 9035178.0,
|
|
"step": 4025
|
|
},
|
|
{
|
|
"entropy": 5.625777292251587,
|
|
"epoch": 4.356949702541915,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003659145379956158,
|
|
"loss": 5.0538,
|
|
"mean_token_accuracy": 0.21197223663330078,
|
|
"num_tokens": 9045807.0,
|
|
"step": 4030
|
|
},
|
|
{
|
|
"entropy": 5.633418416976928,
|
|
"epoch": 4.3623580313683075,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00036552202786494393,
|
|
"loss": 5.128,
|
|
"mean_token_accuracy": 0.2087592974305153,
|
|
"num_tokens": 9056927.0,
|
|
"step": 4035
|
|
},
|
|
{
|
|
"entropy": 5.499518632888794,
|
|
"epoch": 4.367766360194699,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.00036512918877701367,
|
|
"loss": 5.1006,
|
|
"mean_token_accuracy": 0.21158147752285003,
|
|
"num_tokens": 9070141.0,
|
|
"step": 4040
|
|
},
|
|
{
|
|
"entropy": 5.5149468898773195,
|
|
"epoch": 4.373174689021092,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00036473602215940316,
|
|
"loss": 4.9999,
|
|
"mean_token_accuracy": 0.21572860479354858,
|
|
"num_tokens": 9081595.0,
|
|
"step": 4045
|
|
},
|
|
{
|
|
"entropy": 5.537314939498901,
|
|
"epoch": 4.378583017847485,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0003643425294408807,
|
|
"loss": 5.094,
|
|
"mean_token_accuracy": 0.20537076890468597,
|
|
"num_tokens": 9092158.0,
|
|
"step": 4050
|
|
},
|
|
{
|
|
"entropy": 5.559205722808838,
|
|
"epoch": 4.383991346673878,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00036394871205139984,
|
|
"loss": 5.0904,
|
|
"mean_token_accuracy": 0.20810774564743043,
|
|
"num_tokens": 9102958.0,
|
|
"step": 4055
|
|
},
|
|
{
|
|
"entropy": 5.55505781173706,
|
|
"epoch": 4.38939967550027,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0003635545714220938,
|
|
"loss": 5.0501,
|
|
"mean_token_accuracy": 0.2073355570435524,
|
|
"num_tokens": 9113793.0,
|
|
"step": 4060
|
|
},
|
|
{
|
|
"entropy": 5.590800666809082,
|
|
"epoch": 4.394808004326663,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0003631601089852706,
|
|
"loss": 5.1609,
|
|
"mean_token_accuracy": 0.20814414769411088,
|
|
"num_tokens": 9125149.0,
|
|
"step": 4065
|
|
},
|
|
{
|
|
"entropy": 5.476324653625488,
|
|
"epoch": 4.400216333153056,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0003627653261744075,
|
|
"loss": 4.9524,
|
|
"mean_token_accuracy": 0.22145135551691056,
|
|
"num_tokens": 9136222.0,
|
|
"step": 4070
|
|
},
|
|
{
|
|
"entropy": 5.571965456008911,
|
|
"epoch": 4.405624661979449,
|
|
"grad_norm": 0.796875,
|
|
"learning_rate": 0.00036237022442414636,
|
|
"loss": 5.1411,
|
|
"mean_token_accuracy": 0.20361388623714446,
|
|
"num_tokens": 9147933.0,
|
|
"step": 4075
|
|
},
|
|
{
|
|
"entropy": 5.511048698425293,
|
|
"epoch": 4.411032990805841,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0003619748051702878,
|
|
"loss": 5.1336,
|
|
"mean_token_accuracy": 0.2095090851187706,
|
|
"num_tokens": 9159421.0,
|
|
"step": 4080
|
|
},
|
|
{
|
|
"entropy": 5.557892847061157,
|
|
"epoch": 4.416441319632233,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0003615790698497862,
|
|
"loss": 5.0512,
|
|
"mean_token_accuracy": 0.21234578341245652,
|
|
"num_tokens": 9170676.0,
|
|
"step": 4085
|
|
},
|
|
{
|
|
"entropy": 5.5702451229095455,
|
|
"epoch": 4.421849648458626,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00036118301990074463,
|
|
"loss": 5.0636,
|
|
"mean_token_accuracy": 0.21222918927669526,
|
|
"num_tokens": 9181267.0,
|
|
"step": 4090
|
|
},
|
|
{
|
|
"entropy": 5.501087951660156,
|
|
"epoch": 4.427257977285019,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0003607866567624097,
|
|
"loss": 5.1011,
|
|
"mean_token_accuracy": 0.20349736213684083,
|
|
"num_tokens": 9192999.0,
|
|
"step": 4095
|
|
},
|
|
{
|
|
"entropy": 5.544148254394531,
|
|
"epoch": 4.432666306111411,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00036038998187516583,
|
|
"loss": 5.0377,
|
|
"mean_token_accuracy": 0.20985708236694336,
|
|
"num_tokens": 9203778.0,
|
|
"step": 4100
|
|
},
|
|
{
|
|
"entropy": 5.541201639175415,
|
|
"epoch": 4.438074634937804,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00035999299668053043,
|
|
"loss": 5.0731,
|
|
"mean_token_accuracy": 0.21160940825939178,
|
|
"num_tokens": 9214331.0,
|
|
"step": 4105
|
|
},
|
|
{
|
|
"entropy": 5.528553581237793,
|
|
"epoch": 4.443482963764197,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0003595957026211487,
|
|
"loss": 5.0888,
|
|
"mean_token_accuracy": 0.21348123997449875,
|
|
"num_tokens": 9224809.0,
|
|
"step": 4110
|
|
},
|
|
{
|
|
"entropy": 5.615850734710693,
|
|
"epoch": 4.44889129259059,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00035919810114078815,
|
|
"loss": 5.2159,
|
|
"mean_token_accuracy": 0.20122536420822143,
|
|
"num_tokens": 9235852.0,
|
|
"step": 4115
|
|
},
|
|
{
|
|
"entropy": 5.566235685348511,
|
|
"epoch": 4.4542996214169825,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00035880019368433354,
|
|
"loss": 5.0755,
|
|
"mean_token_accuracy": 0.2152230754494667,
|
|
"num_tokens": 9246296.0,
|
|
"step": 4120
|
|
},
|
|
{
|
|
"entropy": 5.590666007995606,
|
|
"epoch": 4.4597079502433745,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0003584019816977813,
|
|
"loss": 5.0692,
|
|
"mean_token_accuracy": 0.22279670387506484,
|
|
"num_tokens": 9257761.0,
|
|
"step": 4125
|
|
},
|
|
{
|
|
"entropy": 5.467157554626465,
|
|
"epoch": 4.465116279069767,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00035800346662823493,
|
|
"loss": 5.1232,
|
|
"mean_token_accuracy": 0.2100720837712288,
|
|
"num_tokens": 9268796.0,
|
|
"step": 4130
|
|
},
|
|
{
|
|
"entropy": 5.622251224517822,
|
|
"epoch": 4.47052460789616,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000357604649923899,
|
|
"loss": 5.1783,
|
|
"mean_token_accuracy": 0.200840924680233,
|
|
"num_tokens": 9280625.0,
|
|
"step": 4135
|
|
},
|
|
{
|
|
"entropy": 5.54943437576294,
|
|
"epoch": 4.475932936722553,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0003572055330340744,
|
|
"loss": 5.0068,
|
|
"mean_token_accuracy": 0.21755584180355073,
|
|
"num_tokens": 9291096.0,
|
|
"step": 4140
|
|
},
|
|
{
|
|
"entropy": 5.538061571121216,
|
|
"epoch": 4.481341265548945,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00035680611740915264,
|
|
"loss": 5.0488,
|
|
"mean_token_accuracy": 0.21059021651744841,
|
|
"num_tokens": 9301664.0,
|
|
"step": 4145
|
|
},
|
|
{
|
|
"entropy": 5.439249086380005,
|
|
"epoch": 4.486749594375338,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0003564064045006113,
|
|
"loss": 5.024,
|
|
"mean_token_accuracy": 0.20721930414438247,
|
|
"num_tokens": 9312907.0,
|
|
"step": 4150
|
|
},
|
|
{
|
|
"entropy": 5.576662397384643,
|
|
"epoch": 4.492157923201731,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00035600639576100786,
|
|
"loss": 5.1467,
|
|
"mean_token_accuracy": 0.20236302316188812,
|
|
"num_tokens": 9323806.0,
|
|
"step": 4155
|
|
},
|
|
{
|
|
"entropy": 5.602390193939209,
|
|
"epoch": 4.497566252028124,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00035560609264397497,
|
|
"loss": 5.1534,
|
|
"mean_token_accuracy": 0.20166585296392442,
|
|
"num_tokens": 9336394.0,
|
|
"step": 4160
|
|
},
|
|
{
|
|
"entropy": 5.504319047927856,
|
|
"epoch": 4.502974580854516,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00035520549660421515,
|
|
"loss": 5.0687,
|
|
"mean_token_accuracy": 0.21283844709396363,
|
|
"num_tokens": 9349177.0,
|
|
"step": 4165
|
|
},
|
|
{
|
|
"entropy": 5.604043436050415,
|
|
"epoch": 4.5083829096809085,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0003548046090974954,
|
|
"loss": 5.1308,
|
|
"mean_token_accuracy": 0.2067163184285164,
|
|
"num_tokens": 9360033.0,
|
|
"step": 4170
|
|
},
|
|
{
|
|
"entropy": 5.505042457580567,
|
|
"epoch": 4.513791238507301,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00035440343158064176,
|
|
"loss": 5.0874,
|
|
"mean_token_accuracy": 0.20834399461746217,
|
|
"num_tokens": 9371300.0,
|
|
"step": 4175
|
|
},
|
|
{
|
|
"entropy": 5.528640604019165,
|
|
"epoch": 4.519199567333694,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.0003540019655115341,
|
|
"loss": 5.1041,
|
|
"mean_token_accuracy": 0.21185071468353273,
|
|
"num_tokens": 9383112.0,
|
|
"step": 4180
|
|
},
|
|
{
|
|
"entropy": 5.483243942260742,
|
|
"epoch": 4.524607896160086,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.00035360021234910133,
|
|
"loss": 4.99,
|
|
"mean_token_accuracy": 0.21568265557289124,
|
|
"num_tokens": 9394062.0,
|
|
"step": 4185
|
|
},
|
|
{
|
|
"entropy": 5.555526924133301,
|
|
"epoch": 4.530016224986479,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00035319817355331537,
|
|
"loss": 5.1175,
|
|
"mean_token_accuracy": 0.2110666811466217,
|
|
"num_tokens": 9406016.0,
|
|
"step": 4190
|
|
},
|
|
{
|
|
"entropy": 5.63165397644043,
|
|
"epoch": 4.535424553812872,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0003527958505851858,
|
|
"loss": 5.1618,
|
|
"mean_token_accuracy": 0.20736465603113174,
|
|
"num_tokens": 9417286.0,
|
|
"step": 4195
|
|
},
|
|
{
|
|
"entropy": 5.510341310501099,
|
|
"epoch": 4.540832882639265,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00035239324490675546,
|
|
"loss": 5.0086,
|
|
"mean_token_accuracy": 0.21016963869333266,
|
|
"num_tokens": 9428794.0,
|
|
"step": 4200
|
|
},
|
|
{
|
|
"entropy": 5.5268713474273685,
|
|
"epoch": 4.546241211465657,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0003519903579810943,
|
|
"loss": 5.1531,
|
|
"mean_token_accuracy": 0.20583009719848633,
|
|
"num_tokens": 9440524.0,
|
|
"step": 4205
|
|
},
|
|
{
|
|
"entropy": 5.4855506896972654,
|
|
"epoch": 4.55164954029205,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0003515871912722944,
|
|
"loss": 5.0154,
|
|
"mean_token_accuracy": 0.21542346179485322,
|
|
"num_tokens": 9450796.0,
|
|
"step": 4210
|
|
},
|
|
{
|
|
"entropy": 5.59025011062622,
|
|
"epoch": 4.5570578691184425,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0003511837462454643,
|
|
"loss": 5.1522,
|
|
"mean_token_accuracy": 0.20700487792491912,
|
|
"num_tokens": 9462498.0,
|
|
"step": 4215
|
|
},
|
|
{
|
|
"entropy": 5.5992223739624025,
|
|
"epoch": 4.562466197944835,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00035078002436672443,
|
|
"loss": 5.0793,
|
|
"mean_token_accuracy": 0.21457747220993043,
|
|
"num_tokens": 9473395.0,
|
|
"step": 4220
|
|
},
|
|
{
|
|
"entropy": 5.5534467697143555,
|
|
"epoch": 4.567874526771227,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0003503760271032007,
|
|
"loss": 5.1393,
|
|
"mean_token_accuracy": 0.2054167002439499,
|
|
"num_tokens": 9485099.0,
|
|
"step": 4225
|
|
},
|
|
{
|
|
"entropy": 5.612093877792359,
|
|
"epoch": 4.57328285559762,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0003499717559230205,
|
|
"loss": 5.1223,
|
|
"mean_token_accuracy": 0.2043273314833641,
|
|
"num_tokens": 9495129.0,
|
|
"step": 4230
|
|
},
|
|
{
|
|
"entropy": 5.557339811325074,
|
|
"epoch": 4.578691184424013,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00034956721229530586,
|
|
"loss": 5.1595,
|
|
"mean_token_accuracy": 0.20534357726573943,
|
|
"num_tokens": 9506933.0,
|
|
"step": 4235
|
|
},
|
|
{
|
|
"entropy": 5.529027986526489,
|
|
"epoch": 4.584099513250406,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0003491623976901695,
|
|
"loss": 5.0387,
|
|
"mean_token_accuracy": 0.21302892416715621,
|
|
"num_tokens": 9518262.0,
|
|
"step": 4240
|
|
},
|
|
{
|
|
"entropy": 5.598697853088379,
|
|
"epoch": 4.589507842076799,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.0003487573135787085,
|
|
"loss": 5.0785,
|
|
"mean_token_accuracy": 0.2066194474697113,
|
|
"num_tokens": 9530829.0,
|
|
"step": 4245
|
|
},
|
|
{
|
|
"entropy": 5.465062379837036,
|
|
"epoch": 4.594916170903191,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00034835196143299956,
|
|
"loss": 4.9612,
|
|
"mean_token_accuracy": 0.22218543589115142,
|
|
"num_tokens": 9541762.0,
|
|
"step": 4250
|
|
},
|
|
{
|
|
"entropy": 5.507507181167602,
|
|
"epoch": 4.600324499729584,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00034794634272609313,
|
|
"loss": 5.0666,
|
|
"mean_token_accuracy": 0.21444679349660872,
|
|
"num_tokens": 9552344.0,
|
|
"step": 4255
|
|
},
|
|
{
|
|
"entropy": 5.420781087875366,
|
|
"epoch": 4.605732828555976,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0003475404589320089,
|
|
"loss": 4.9647,
|
|
"mean_token_accuracy": 0.20661905854940416,
|
|
"num_tokens": 9563806.0,
|
|
"step": 4260
|
|
},
|
|
{
|
|
"entropy": 5.557787322998047,
|
|
"epoch": 4.611141157382368,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0003471343115257291,
|
|
"loss": 5.1441,
|
|
"mean_token_accuracy": 0.20521425306797028,
|
|
"num_tokens": 9575305.0,
|
|
"step": 4265
|
|
},
|
|
{
|
|
"entropy": 5.640677499771118,
|
|
"epoch": 4.616549486208761,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00034672790198319453,
|
|
"loss": 5.167,
|
|
"mean_token_accuracy": 0.20268668085336686,
|
|
"num_tokens": 9586155.0,
|
|
"step": 4270
|
|
},
|
|
{
|
|
"entropy": 5.572687005996704,
|
|
"epoch": 4.621957815035154,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0003463212317812985,
|
|
"loss": 5.1454,
|
|
"mean_token_accuracy": 0.20789299607276918,
|
|
"num_tokens": 9597731.0,
|
|
"step": 4275
|
|
},
|
|
{
|
|
"entropy": 5.6729076385498045,
|
|
"epoch": 4.627366143861547,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003459143023978813,
|
|
"loss": 5.1317,
|
|
"mean_token_accuracy": 0.20901119112968444,
|
|
"num_tokens": 9609438.0,
|
|
"step": 4280
|
|
},
|
|
{
|
|
"entropy": 5.562269020080566,
|
|
"epoch": 4.63277447268794,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0003455071153117252,
|
|
"loss": 5.0949,
|
|
"mean_token_accuracy": 0.20159664154052734,
|
|
"num_tokens": 9620665.0,
|
|
"step": 4285
|
|
},
|
|
{
|
|
"entropy": 5.539447832107544,
|
|
"epoch": 4.638182801514332,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0003450996720025493,
|
|
"loss": 5.1311,
|
|
"mean_token_accuracy": 0.20525468587875367,
|
|
"num_tokens": 9631868.0,
|
|
"step": 4290
|
|
},
|
|
{
|
|
"entropy": 5.483764791488648,
|
|
"epoch": 4.643591130340725,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.00034469197395100323,
|
|
"loss": 4.9889,
|
|
"mean_token_accuracy": 0.2170760914683342,
|
|
"num_tokens": 9643879.0,
|
|
"step": 4295
|
|
},
|
|
{
|
|
"entropy": 5.5390242576599125,
|
|
"epoch": 4.6489994591671175,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.00034428402263866285,
|
|
"loss": 5.0369,
|
|
"mean_token_accuracy": 0.21911296546459197,
|
|
"num_tokens": 9656277.0,
|
|
"step": 4300
|
|
},
|
|
{
|
|
"entropy": 5.523605394363403,
|
|
"epoch": 4.6544077879935095,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.00034387581954802406,
|
|
"loss": 5.1341,
|
|
"mean_token_accuracy": 0.20436792373657225,
|
|
"num_tokens": 9666590.0,
|
|
"step": 4305
|
|
},
|
|
{
|
|
"entropy": 5.575499248504639,
|
|
"epoch": 4.659816116819902,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00034346736616249786,
|
|
"loss": 5.1014,
|
|
"mean_token_accuracy": 0.2100989818572998,
|
|
"num_tokens": 9677248.0,
|
|
"step": 4310
|
|
},
|
|
{
|
|
"entropy": 5.528906965255738,
|
|
"epoch": 4.665224445646295,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00034305866396640473,
|
|
"loss": 5.0628,
|
|
"mean_token_accuracy": 0.20571537613868712,
|
|
"num_tokens": 9689617.0,
|
|
"step": 4315
|
|
},
|
|
{
|
|
"entropy": 5.51939058303833,
|
|
"epoch": 4.670632774472688,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00034264971444496954,
|
|
"loss": 5.1372,
|
|
"mean_token_accuracy": 0.20264151841402053,
|
|
"num_tokens": 9702663.0,
|
|
"step": 4320
|
|
},
|
|
{
|
|
"entropy": 5.516185235977173,
|
|
"epoch": 4.676041103299081,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00034224051908431574,
|
|
"loss": 5.1636,
|
|
"mean_token_accuracy": 0.20523711591958999,
|
|
"num_tokens": 9714005.0,
|
|
"step": 4325
|
|
},
|
|
{
|
|
"entropy": 5.560047626495361,
|
|
"epoch": 4.681449432125473,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0003418310793714602,
|
|
"loss": 5.0622,
|
|
"mean_token_accuracy": 0.2134099081158638,
|
|
"num_tokens": 9724931.0,
|
|
"step": 4330
|
|
},
|
|
{
|
|
"entropy": 5.5301953792572025,
|
|
"epoch": 4.686857760951866,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003414213967943077,
|
|
"loss": 5.0612,
|
|
"mean_token_accuracy": 0.20607510209083557,
|
|
"num_tokens": 9734948.0,
|
|
"step": 4335
|
|
},
|
|
{
|
|
"entropy": 5.487113666534424,
|
|
"epoch": 4.692266089778259,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00034101147284164583,
|
|
"loss": 5.0187,
|
|
"mean_token_accuracy": 0.2138725608587265,
|
|
"num_tokens": 9746424.0,
|
|
"step": 4340
|
|
},
|
|
{
|
|
"entropy": 5.600327634811402,
|
|
"epoch": 4.6976744186046515,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00034060130900313905,
|
|
"loss": 5.1805,
|
|
"mean_token_accuracy": 0.21004009395837783,
|
|
"num_tokens": 9758434.0,
|
|
"step": 4345
|
|
},
|
|
{
|
|
"entropy": 5.516699647903442,
|
|
"epoch": 4.7030827474310435,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00034019090676932383,
|
|
"loss": 5.0387,
|
|
"mean_token_accuracy": 0.20620029866695405,
|
|
"num_tokens": 9768869.0,
|
|
"step": 4350
|
|
},
|
|
{
|
|
"entropy": 5.616180181503296,
|
|
"epoch": 4.708491076257436,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0003397802676316027,
|
|
"loss": 5.1973,
|
|
"mean_token_accuracy": 0.20477681905031203,
|
|
"num_tokens": 9780001.0,
|
|
"step": 4355
|
|
},
|
|
{
|
|
"entropy": 5.56871976852417,
|
|
"epoch": 4.713899405083829,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00033936939308223935,
|
|
"loss": 5.1394,
|
|
"mean_token_accuracy": 0.20348589569330217,
|
|
"num_tokens": 9790609.0,
|
|
"step": 4360
|
|
},
|
|
{
|
|
"entropy": 5.492574024200439,
|
|
"epoch": 4.719307733910222,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00033895828461435275,
|
|
"loss": 5.0634,
|
|
"mean_token_accuracy": 0.21887852400541305,
|
|
"num_tokens": 9802064.0,
|
|
"step": 4365
|
|
},
|
|
{
|
|
"entropy": 5.544607400894165,
|
|
"epoch": 4.724716062736614,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0003385469437219121,
|
|
"loss": 5.1384,
|
|
"mean_token_accuracy": 0.20900798141956328,
|
|
"num_tokens": 9812124.0,
|
|
"step": 4370
|
|
},
|
|
{
|
|
"entropy": 5.5435210227966305,
|
|
"epoch": 4.730124391563007,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00033813537189973105,
|
|
"loss": 4.98,
|
|
"mean_token_accuracy": 0.2272951290011406,
|
|
"num_tokens": 9824371.0,
|
|
"step": 4375
|
|
},
|
|
{
|
|
"entropy": 5.507232809066773,
|
|
"epoch": 4.7355327203894,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00033772357064346266,
|
|
"loss": 5.0107,
|
|
"mean_token_accuracy": 0.21994542926549912,
|
|
"num_tokens": 9834919.0,
|
|
"step": 4380
|
|
},
|
|
{
|
|
"entropy": 5.493790626525879,
|
|
"epoch": 4.740941049215793,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0003373115414495935,
|
|
"loss": 5.1082,
|
|
"mean_token_accuracy": 0.2011227235198021,
|
|
"num_tokens": 9845435.0,
|
|
"step": 4385
|
|
},
|
|
{
|
|
"entropy": 5.54889178276062,
|
|
"epoch": 4.746349378042185,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0003368992858154389,
|
|
"loss": 5.0641,
|
|
"mean_token_accuracy": 0.2104277119040489,
|
|
"num_tokens": 9856091.0,
|
|
"step": 4390
|
|
},
|
|
{
|
|
"entropy": 5.479821825027466,
|
|
"epoch": 4.7517577068685775,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00033648680523913643,
|
|
"loss": 4.9348,
|
|
"mean_token_accuracy": 0.218231301009655,
|
|
"num_tokens": 9866861.0,
|
|
"step": 4395
|
|
},
|
|
{
|
|
"entropy": 5.454116535186768,
|
|
"epoch": 4.75716603569497,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003360741012196417,
|
|
"loss": 5.0393,
|
|
"mean_token_accuracy": 0.20523456186056138,
|
|
"num_tokens": 9877852.0,
|
|
"step": 4400
|
|
},
|
|
{
|
|
"entropy": 5.534960031509399,
|
|
"epoch": 4.762574364521363,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000335661175256722,
|
|
"loss": 4.9843,
|
|
"mean_token_accuracy": 0.2199228972196579,
|
|
"num_tokens": 9888938.0,
|
|
"step": 4405
|
|
},
|
|
{
|
|
"entropy": 5.560315465927124,
|
|
"epoch": 4.767982693347756,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0003352480288509513,
|
|
"loss": 5.0598,
|
|
"mean_token_accuracy": 0.219546115398407,
|
|
"num_tokens": 9901782.0,
|
|
"step": 4410
|
|
},
|
|
{
|
|
"entropy": 5.521196699142456,
|
|
"epoch": 4.773391022174148,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.00033483466350370445,
|
|
"loss": 5.1658,
|
|
"mean_token_accuracy": 0.19951529055833817,
|
|
"num_tokens": 9913254.0,
|
|
"step": 4415
|
|
},
|
|
{
|
|
"entropy": 5.502304553985596,
|
|
"epoch": 4.778799351000541,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00033442108071715215,
|
|
"loss": 5.0496,
|
|
"mean_token_accuracy": 0.2143803671002388,
|
|
"num_tokens": 9923170.0,
|
|
"step": 4420
|
|
},
|
|
{
|
|
"entropy": 5.606217288970948,
|
|
"epoch": 4.784207679826934,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00033400728199425525,
|
|
"loss": 5.1021,
|
|
"mean_token_accuracy": 0.21283678859472274,
|
|
"num_tokens": 9934087.0,
|
|
"step": 4425
|
|
},
|
|
{
|
|
"entropy": 5.4965592384338375,
|
|
"epoch": 4.789616008653326,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.00033359326883875917,
|
|
"loss": 5.1654,
|
|
"mean_token_accuracy": 0.2012577623128891,
|
|
"num_tokens": 9945463.0,
|
|
"step": 4430
|
|
},
|
|
{
|
|
"entropy": 5.501139736175537,
|
|
"epoch": 4.795024337479719,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00033317904275518873,
|
|
"loss": 4.9974,
|
|
"mean_token_accuracy": 0.20785470306873322,
|
|
"num_tokens": 9955602.0,
|
|
"step": 4435
|
|
},
|
|
{
|
|
"entropy": 5.512185335159302,
|
|
"epoch": 4.800432666306111,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003327646052488425,
|
|
"loss": 5.0363,
|
|
"mean_token_accuracy": 0.21250054240226746,
|
|
"num_tokens": 9965899.0,
|
|
"step": 4440
|
|
},
|
|
{
|
|
"entropy": 5.591942167282104,
|
|
"epoch": 4.805840995132504,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0003323499578257873,
|
|
"loss": 5.2056,
|
|
"mean_token_accuracy": 0.19892056584358214,
|
|
"num_tokens": 9978047.0,
|
|
"step": 4445
|
|
},
|
|
{
|
|
"entropy": 5.663182115554809,
|
|
"epoch": 4.811249323958897,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00033193510199285286,
|
|
"loss": 5.2594,
|
|
"mean_token_accuracy": 0.19744797497987748,
|
|
"num_tokens": 9991760.0,
|
|
"step": 4450
|
|
},
|
|
{
|
|
"entropy": 5.5475287437438965,
|
|
"epoch": 4.816657652785289,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00033152003925762615,
|
|
"loss": 5.0866,
|
|
"mean_token_accuracy": 0.20573129653930664,
|
|
"num_tokens": 10002540.0,
|
|
"step": 4455
|
|
},
|
|
{
|
|
"entropy": 5.551731443405151,
|
|
"epoch": 4.822065981611682,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003311047711284462,
|
|
"loss": 5.1077,
|
|
"mean_token_accuracy": 0.2031647101044655,
|
|
"num_tokens": 10013955.0,
|
|
"step": 4460
|
|
},
|
|
{
|
|
"entropy": 5.583474063873291,
|
|
"epoch": 4.827474310438075,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0003306892991143983,
|
|
"loss": 5.0944,
|
|
"mean_token_accuracy": 0.20376263111829757,
|
|
"num_tokens": 10024845.0,
|
|
"step": 4465
|
|
},
|
|
{
|
|
"entropy": 5.531608533859253,
|
|
"epoch": 4.832882639264467,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0003302736247253087,
|
|
"loss": 5.1198,
|
|
"mean_token_accuracy": 0.21353143751621245,
|
|
"num_tokens": 10036440.0,
|
|
"step": 4470
|
|
},
|
|
{
|
|
"entropy": 5.582123136520385,
|
|
"epoch": 4.83829096809086,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0003298577494717391,
|
|
"loss": 5.1161,
|
|
"mean_token_accuracy": 0.2058424174785614,
|
|
"num_tokens": 10049891.0,
|
|
"step": 4475
|
|
},
|
|
{
|
|
"entropy": 5.505520439147949,
|
|
"epoch": 4.8436992969172525,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0003294416748649812,
|
|
"loss": 5.0116,
|
|
"mean_token_accuracy": 0.21813010573387145,
|
|
"num_tokens": 10061105.0,
|
|
"step": 4480
|
|
},
|
|
{
|
|
"entropy": 5.504058456420898,
|
|
"epoch": 4.849107625743645,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00032902540241705106,
|
|
"loss": 5.0819,
|
|
"mean_token_accuracy": 0.21307054460048674,
|
|
"num_tokens": 10072721.0,
|
|
"step": 4485
|
|
},
|
|
{
|
|
"entropy": 5.508000802993775,
|
|
"epoch": 4.854515954570038,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0003286089336406837,
|
|
"loss": 5.04,
|
|
"mean_token_accuracy": 0.21326482594013213,
|
|
"num_tokens": 10083288.0,
|
|
"step": 4490
|
|
},
|
|
{
|
|
"entropy": 5.602752351760865,
|
|
"epoch": 4.85992428339643,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00032819227004932745,
|
|
"loss": 5.1096,
|
|
"mean_token_accuracy": 0.2064409077167511,
|
|
"num_tokens": 10093879.0,
|
|
"step": 4495
|
|
},
|
|
{
|
|
"entropy": 5.508682727813721,
|
|
"epoch": 4.865332612222823,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000327775413157139,
|
|
"loss": 5.0448,
|
|
"mean_token_accuracy": 0.20702906399965287,
|
|
"num_tokens": 10104861.0,
|
|
"step": 4500
|
|
},
|
|
{
|
|
"epoch": 4.865332612222823,
|
|
"eval_entropy": 5.407308044216849,
|
|
"eval_loss": 5.845698356628418,
|
|
"eval_mean_token_accuracy": 0.18109293117780578,
|
|
"eval_num_tokens": 10104861.0,
|
|
"eval_runtime": 3.5424,
|
|
"eval_samples_per_second": 992.253,
|
|
"eval_steps_per_second": 124.208,
|
|
"step": 4500
|
|
}
|
|
],
|
|
"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": 2338775038033920.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|