2490fa797a
Model: fpadovani/eng-latn-100mb-after-ppt-shuff-dyck-10mb-ckpt500_seed3407 Source: Original Platform
1046 lines
28 KiB
JSON
1046 lines
28 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.5408328826392644,
|
|
"eval_steps": 500,
|
|
"global_step": 500,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 4.802441883087158,
|
|
"epoch": 0.005408328826392645,
|
|
"grad_norm": 13.375,
|
|
"learning_rate": 2e-06,
|
|
"loss": 14.4628,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 10942.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 4.81857442855835,
|
|
"epoch": 0.01081665765278529,
|
|
"grad_norm": 16.25,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 14.3862,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 21887.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 4.852184391021728,
|
|
"epoch": 0.016224986479177934,
|
|
"grad_norm": 18.375,
|
|
"learning_rate": 7e-06,
|
|
"loss": 14.2078,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 34345.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 5.0442163944244385,
|
|
"epoch": 0.02163331530557058,
|
|
"grad_norm": 27.0,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 13.6012,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 45445.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 6.944517469406128,
|
|
"epoch": 0.02704164413196322,
|
|
"grad_norm": 19.5,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 12.0589,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 58237.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.193405723571777,
|
|
"epoch": 0.03244997295835587,
|
|
"grad_norm": 3.921875,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 10.8364,
|
|
"mean_token_accuracy": 0.002917378023266792,
|
|
"num_tokens": 68098.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.690443992614746,
|
|
"epoch": 0.03785830178474851,
|
|
"grad_norm": 3.234375,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 10.5773,
|
|
"mean_token_accuracy": 0.014713712595403194,
|
|
"num_tokens": 79867.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.687717247009278,
|
|
"epoch": 0.04326663061114116,
|
|
"grad_norm": 2.90625,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 10.2534,
|
|
"mean_token_accuracy": 0.016428307350724935,
|
|
"num_tokens": 91257.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.658015537261964,
|
|
"epoch": 0.048674959437533805,
|
|
"grad_norm": 2.515625,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 9.9866,
|
|
"mean_token_accuracy": 0.024076672736555338,
|
|
"num_tokens": 103474.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.61853494644165,
|
|
"epoch": 0.05408328826392644,
|
|
"grad_norm": 2.53125,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 9.7234,
|
|
"mean_token_accuracy": 0.03784517608582973,
|
|
"num_tokens": 114830.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.597916889190675,
|
|
"epoch": 0.05949161709031909,
|
|
"grad_norm": 2.171875,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.5693,
|
|
"mean_token_accuracy": 0.035866692662239075,
|
|
"num_tokens": 126017.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.58803596496582,
|
|
"epoch": 0.06489994591671173,
|
|
"grad_norm": 2.171875,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.4863,
|
|
"mean_token_accuracy": 0.040941419266164306,
|
|
"num_tokens": 138509.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.57759370803833,
|
|
"epoch": 0.07030827474310439,
|
|
"grad_norm": 2.046875,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.4155,
|
|
"mean_token_accuracy": 0.040815659798681736,
|
|
"num_tokens": 148353.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.564495277404784,
|
|
"epoch": 0.07571660356949703,
|
|
"grad_norm": 2.0,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.3856,
|
|
"mean_token_accuracy": 0.03659801315516233,
|
|
"num_tokens": 160269.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.558395195007325,
|
|
"epoch": 0.08112493239588967,
|
|
"grad_norm": 2.140625,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.2438,
|
|
"mean_token_accuracy": 0.036995376273989675,
|
|
"num_tokens": 171568.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.547275161743164,
|
|
"epoch": 0.08653326122228232,
|
|
"grad_norm": 2.140625,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.1614,
|
|
"mean_token_accuracy": 0.0399360004812479,
|
|
"num_tokens": 184243.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.52041311264038,
|
|
"epoch": 0.09194159004867496,
|
|
"grad_norm": 2.09375,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.0786,
|
|
"mean_token_accuracy": 0.039298290386796,
|
|
"num_tokens": 194887.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.486145210266113,
|
|
"epoch": 0.09734991887506761,
|
|
"grad_norm": 2.0,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 8.9895,
|
|
"mean_token_accuracy": 0.04011793658137321,
|
|
"num_tokens": 206476.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.448410129547119,
|
|
"epoch": 0.10275824770146025,
|
|
"grad_norm": 1.9609375,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 8.8381,
|
|
"mean_token_accuracy": 0.04185208044946194,
|
|
"num_tokens": 216578.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.391289806365966,
|
|
"epoch": 0.10816657652785289,
|
|
"grad_norm": 1.8515625,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 8.7399,
|
|
"mean_token_accuracy": 0.0479723758995533,
|
|
"num_tokens": 226941.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.326226139068604,
|
|
"epoch": 0.11357490535424554,
|
|
"grad_norm": 1.7890625,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 8.6146,
|
|
"mean_token_accuracy": 0.05267183780670166,
|
|
"num_tokens": 238060.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.213607120513917,
|
|
"epoch": 0.11898323418063818,
|
|
"grad_norm": 1.9140625,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 8.4421,
|
|
"mean_token_accuracy": 0.055974167212843894,
|
|
"num_tokens": 249137.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 10.048615741729737,
|
|
"epoch": 0.12439156300703083,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 8.3336,
|
|
"mean_token_accuracy": 0.05477573201060295,
|
|
"num_tokens": 260712.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 9.906066799163819,
|
|
"epoch": 0.12979989183342347,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.1638,
|
|
"mean_token_accuracy": 0.05388197749853134,
|
|
"num_tokens": 271039.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 9.787034893035889,
|
|
"epoch": 0.1352082206598161,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.0076,
|
|
"mean_token_accuracy": 0.05202497914433479,
|
|
"num_tokens": 281353.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 9.50912857055664,
|
|
"epoch": 0.14061654948620878,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 7.9456,
|
|
"mean_token_accuracy": 0.054855041578412055,
|
|
"num_tokens": 292221.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 9.263109493255616,
|
|
"epoch": 0.1460248783126014,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 7.8241,
|
|
"mean_token_accuracy": 0.0536174725741148,
|
|
"num_tokens": 303353.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 8.991836738586425,
|
|
"epoch": 0.15143320713899405,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 7.7511,
|
|
"mean_token_accuracy": 0.056266501918435095,
|
|
"num_tokens": 314165.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 8.685503292083741,
|
|
"epoch": 0.1568415359653867,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 7.5805,
|
|
"mean_token_accuracy": 0.05333330743014812,
|
|
"num_tokens": 326387.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 8.49657220840454,
|
|
"epoch": 0.16224986479177933,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 7.5908,
|
|
"mean_token_accuracy": 0.05400457009673119,
|
|
"num_tokens": 339155.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 8.256601428985595,
|
|
"epoch": 0.167658193618172,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 7.5015,
|
|
"mean_token_accuracy": 0.05675696507096291,
|
|
"num_tokens": 350643.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 8.110760021209718,
|
|
"epoch": 0.17306652244456464,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 7.4385,
|
|
"mean_token_accuracy": 0.05687093436717987,
|
|
"num_tokens": 361977.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 7.999344825744629,
|
|
"epoch": 0.17847485127095727,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.4928,
|
|
"mean_token_accuracy": 0.06705345250666142,
|
|
"num_tokens": 373604.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 7.95798511505127,
|
|
"epoch": 0.1838831800973499,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.4608,
|
|
"mean_token_accuracy": 0.07175817862153053,
|
|
"num_tokens": 384306.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 7.838786649703979,
|
|
"epoch": 0.18929150892374255,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.3682,
|
|
"mean_token_accuracy": 0.07707577534019946,
|
|
"num_tokens": 396479.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 7.802298307418823,
|
|
"epoch": 0.19469983775013522,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.3187,
|
|
"mean_token_accuracy": 0.07556967809796333,
|
|
"num_tokens": 408489.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 7.777406930923462,
|
|
"epoch": 0.20010816657652786,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.3304,
|
|
"mean_token_accuracy": 0.08097317069768906,
|
|
"num_tokens": 419339.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 7.777624797821045,
|
|
"epoch": 0.2055164954029205,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.2935,
|
|
"mean_token_accuracy": 0.08264814168214799,
|
|
"num_tokens": 430486.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 7.753584861755371,
|
|
"epoch": 0.21092482422931313,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.2943,
|
|
"mean_token_accuracy": 0.08632798045873642,
|
|
"num_tokens": 440414.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 7.748952054977417,
|
|
"epoch": 0.21633315305570577,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.2391,
|
|
"mean_token_accuracy": 0.08323732763528824,
|
|
"num_tokens": 451475.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 7.659665584564209,
|
|
"epoch": 0.22174148188209844,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.3028,
|
|
"mean_token_accuracy": 0.08789955824613571,
|
|
"num_tokens": 462053.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 7.667898845672608,
|
|
"epoch": 0.22714981070849108,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.2495,
|
|
"mean_token_accuracy": 0.08465297371149064,
|
|
"num_tokens": 473207.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 7.687205266952515,
|
|
"epoch": 0.23255813953488372,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.2412,
|
|
"mean_token_accuracy": 0.08715058863162994,
|
|
"num_tokens": 484999.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 7.709909105300904,
|
|
"epoch": 0.23796646836127636,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.2455,
|
|
"mean_token_accuracy": 0.08413905426859855,
|
|
"num_tokens": 496772.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 7.729339981079102,
|
|
"epoch": 0.24337479718766902,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.2107,
|
|
"mean_token_accuracy": 0.0819280207157135,
|
|
"num_tokens": 507782.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 7.6187958240509035,
|
|
"epoch": 0.24878312601406166,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.113,
|
|
"mean_token_accuracy": 0.08953967168927193,
|
|
"num_tokens": 518950.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 7.6271685600280765,
|
|
"epoch": 0.2541914548404543,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.1849,
|
|
"mean_token_accuracy": 0.09068010970950127,
|
|
"num_tokens": 530795.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 7.659736394882202,
|
|
"epoch": 0.25959978366684694,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.1958,
|
|
"mean_token_accuracy": 0.08313242197036744,
|
|
"num_tokens": 542878.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.627575349807739,
|
|
"epoch": 0.2650081124932396,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.0933,
|
|
"mean_token_accuracy": 0.08872637823224068,
|
|
"num_tokens": 553682.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.5292261123657225,
|
|
"epoch": 0.2704164413196322,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.0941,
|
|
"mean_token_accuracy": 0.0952358789741993,
|
|
"num_tokens": 564433.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 7.531745386123657,
|
|
"epoch": 0.27582477014602486,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.1263,
|
|
"mean_token_accuracy": 0.09562606438994407,
|
|
"num_tokens": 576001.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.514750719070435,
|
|
"epoch": 0.28123309897241755,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.0509,
|
|
"mean_token_accuracy": 0.09575334340333938,
|
|
"num_tokens": 587138.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.472469186782837,
|
|
"epoch": 0.2866414277988102,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.0111,
|
|
"mean_token_accuracy": 0.097767373919487,
|
|
"num_tokens": 597572.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.394614315032959,
|
|
"epoch": 0.2920497566252028,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 6.9821,
|
|
"mean_token_accuracy": 0.10268613770604133,
|
|
"num_tokens": 608879.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.507109117507935,
|
|
"epoch": 0.29745808545159547,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.0138,
|
|
"mean_token_accuracy": 0.0923161268234253,
|
|
"num_tokens": 620439.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.481431484222412,
|
|
"epoch": 0.3028664142779881,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 7.0666,
|
|
"mean_token_accuracy": 0.0980131283402443,
|
|
"num_tokens": 632543.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.478473043441772,
|
|
"epoch": 0.30827474310438074,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.1097,
|
|
"mean_token_accuracy": 0.10091490522027016,
|
|
"num_tokens": 643369.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.441432142257691,
|
|
"epoch": 0.3136830719307734,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 6.9728,
|
|
"mean_token_accuracy": 0.1002442181110382,
|
|
"num_tokens": 655328.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.446781730651855,
|
|
"epoch": 0.319091400757166,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.0188,
|
|
"mean_token_accuracy": 0.09976664856076241,
|
|
"num_tokens": 665780.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.409105348587036,
|
|
"epoch": 0.32449972958355866,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 6.9372,
|
|
"mean_token_accuracy": 0.10388917848467827,
|
|
"num_tokens": 676566.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.41029577255249,
|
|
"epoch": 0.3299080584099513,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000152,
|
|
"loss": 6.9241,
|
|
"mean_token_accuracy": 0.09972899705171585,
|
|
"num_tokens": 687407.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.285754060745239,
|
|
"epoch": 0.335316387236344,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 6.8611,
|
|
"mean_token_accuracy": 0.10674165338277816,
|
|
"num_tokens": 698162.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.2760748863220215,
|
|
"epoch": 0.34072471606273663,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000157,
|
|
"loss": 6.8426,
|
|
"mean_token_accuracy": 0.10251836776733399,
|
|
"num_tokens": 708743.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.402875375747681,
|
|
"epoch": 0.34613304488912927,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 7.0535,
|
|
"mean_token_accuracy": 0.10290157496929168,
|
|
"num_tokens": 719828.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.430617713928223,
|
|
"epoch": 0.3515413737155219,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000162,
|
|
"loss": 7.0055,
|
|
"mean_token_accuracy": 0.10132194757461548,
|
|
"num_tokens": 732514.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.293907737731933,
|
|
"epoch": 0.35694970254191455,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 6.8334,
|
|
"mean_token_accuracy": 0.10402422174811363,
|
|
"num_tokens": 742391.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.197569942474365,
|
|
"epoch": 0.3623580313683072,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 6.834,
|
|
"mean_token_accuracy": 0.10959370359778405,
|
|
"num_tokens": 753638.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.37444748878479,
|
|
"epoch": 0.3677663601946998,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 7.0511,
|
|
"mean_token_accuracy": 0.09830658435821533,
|
|
"num_tokens": 764850.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.327763652801513,
|
|
"epoch": 0.37317468902109246,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 6.8284,
|
|
"mean_token_accuracy": 0.10949721187353134,
|
|
"num_tokens": 776192.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.142797756195068,
|
|
"epoch": 0.3785830178474851,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 6.7698,
|
|
"mean_token_accuracy": 0.1114330142736435,
|
|
"num_tokens": 787051.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.214595127105713,
|
|
"epoch": 0.3839913466738778,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.000177,
|
|
"loss": 6.81,
|
|
"mean_token_accuracy": 0.10584235936403275,
|
|
"num_tokens": 799281.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.2480724334716795,
|
|
"epoch": 0.38939967550027044,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 6.8244,
|
|
"mean_token_accuracy": 0.10464156717061997,
|
|
"num_tokens": 810003.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.165778160095215,
|
|
"epoch": 0.3948080043266631,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.000182,
|
|
"loss": 6.7971,
|
|
"mean_token_accuracy": 0.10737397745251656,
|
|
"num_tokens": 819749.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.157060241699218,
|
|
"epoch": 0.4002163331530557,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 6.798,
|
|
"mean_token_accuracy": 0.11348775699734688,
|
|
"num_tokens": 830552.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.2775880813598635,
|
|
"epoch": 0.40562466197944835,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.000187,
|
|
"loss": 6.8943,
|
|
"mean_token_accuracy": 0.11229263693094253,
|
|
"num_tokens": 840495.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.101148653030395,
|
|
"epoch": 0.411032990805841,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 6.8311,
|
|
"mean_token_accuracy": 0.11005991026759147,
|
|
"num_tokens": 852915.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.19731650352478,
|
|
"epoch": 0.41644131963223363,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000192,
|
|
"loss": 6.7841,
|
|
"mean_token_accuracy": 0.11650079712271691,
|
|
"num_tokens": 862762.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.14437313079834,
|
|
"epoch": 0.42184964845862627,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 6.7342,
|
|
"mean_token_accuracy": 0.11399752795696258,
|
|
"num_tokens": 873023.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.184113311767578,
|
|
"epoch": 0.4272579772850189,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 6.8456,
|
|
"mean_token_accuracy": 0.10619384720921517,
|
|
"num_tokens": 885281.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.0833296298980715,
|
|
"epoch": 0.43266630611141155,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 6.8406,
|
|
"mean_token_accuracy": 0.10883399248123168,
|
|
"num_tokens": 896210.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.069676637649536,
|
|
"epoch": 0.43807463493780424,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000202,
|
|
"loss": 6.7519,
|
|
"mean_token_accuracy": 0.11440604701638221,
|
|
"num_tokens": 909028.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.126840543746948,
|
|
"epoch": 0.4434829637641969,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 6.6773,
|
|
"mean_token_accuracy": 0.118293134868145,
|
|
"num_tokens": 919710.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.0679234027862545,
|
|
"epoch": 0.4488912925905895,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000207,
|
|
"loss": 6.7387,
|
|
"mean_token_accuracy": 0.1108441911637783,
|
|
"num_tokens": 930019.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.100546026229859,
|
|
"epoch": 0.45429962141698216,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 6.6661,
|
|
"mean_token_accuracy": 0.11544709652662277,
|
|
"num_tokens": 940535.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 6.917388868331909,
|
|
"epoch": 0.4597079502433748,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000212,
|
|
"loss": 6.6207,
|
|
"mean_token_accuracy": 0.11555901542305946,
|
|
"num_tokens": 951390.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.044145488739014,
|
|
"epoch": 0.46511627906976744,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 6.6635,
|
|
"mean_token_accuracy": 0.11623054891824722,
|
|
"num_tokens": 963010.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.005122280120849,
|
|
"epoch": 0.4705246078961601,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 6.629,
|
|
"mean_token_accuracy": 0.12106695622205735,
|
|
"num_tokens": 974032.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.042861032485962,
|
|
"epoch": 0.4759329367225527,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 6.6675,
|
|
"mean_token_accuracy": 0.1184677854180336,
|
|
"num_tokens": 983954.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.009529399871826,
|
|
"epoch": 0.48134126554894535,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.000222,
|
|
"loss": 6.7104,
|
|
"mean_token_accuracy": 0.11903695166110992,
|
|
"num_tokens": 994135.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 6.992412567138672,
|
|
"epoch": 0.48674959437533805,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 6.6665,
|
|
"mean_token_accuracy": 0.11242141798138619,
|
|
"num_tokens": 1006275.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 6.83262996673584,
|
|
"epoch": 0.4921579232017307,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 6.482,
|
|
"mean_token_accuracy": 0.12549787536263465,
|
|
"num_tokens": 1017767.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 6.956169319152832,
|
|
"epoch": 0.4975662520281233,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 6.6167,
|
|
"mean_token_accuracy": 0.12303209006786346,
|
|
"num_tokens": 1028286.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 6.925253438949585,
|
|
"epoch": 0.502974580854516,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 6.6247,
|
|
"mean_token_accuracy": 0.11958765834569932,
|
|
"num_tokens": 1040555.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.025730276107788,
|
|
"epoch": 0.5083829096809086,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.7262,
|
|
"mean_token_accuracy": 0.12084319666028023,
|
|
"num_tokens": 1053668.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 6.828843450546264,
|
|
"epoch": 0.5137912385073012,
|
|
"grad_norm": 1.734375,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.6761,
|
|
"mean_token_accuracy": 0.11609490439295769,
|
|
"num_tokens": 1066042.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 6.954677152633667,
|
|
"epoch": 0.5191995673336939,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.6768,
|
|
"mean_token_accuracy": 0.1181111603975296,
|
|
"num_tokens": 1077843.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 6.920350646972656,
|
|
"epoch": 0.5246078961600865,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.5678,
|
|
"mean_token_accuracy": 0.12242683097720146,
|
|
"num_tokens": 1089448.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 6.856211614608765,
|
|
"epoch": 0.5300162249864792,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.6594,
|
|
"mean_token_accuracy": 0.12442992702126503,
|
|
"num_tokens": 1100989.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.0379091739654545,
|
|
"epoch": 0.5354245538128718,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.7094,
|
|
"mean_token_accuracy": 0.12095102891325951,
|
|
"num_tokens": 1113012.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 6.903893089294433,
|
|
"epoch": 0.5408328826392644,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 6.6584,
|
|
"mean_token_accuracy": 0.11823340728878975,
|
|
"num_tokens": 1123551.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 0.5408328826392644,
|
|
"eval_entropy": 7.013918313113126,
|
|
"eval_loss": 6.79232931137085,
|
|
"eval_mean_token_accuracy": 0.11552235864272171,
|
|
"eval_num_tokens": 1123551.0,
|
|
"eval_runtime": 2.625,
|
|
"eval_samples_per_second": 1339.063,
|
|
"eval_steps_per_second": 167.621,
|
|
"step": 500
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 9240,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 10,
|
|
"save_steps": 500,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 1751032221696000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|