2490fa797a
Model: fpadovani/eng-latn-100mb-after-ppt-shuff-dyck-10mb-ckpt500_seed3407 Source: Original Platform
4079 lines
110 KiB
JSON
4079 lines
110 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 2.162249864791779,
|
|
"eval_steps": 500,
|
|
"global_step": 2000,
|
|
"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
|
|
},
|
|
{
|
|
"entropy": 6.937011671066284,
|
|
"epoch": 0.5462412114656571,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.5714,
|
|
"mean_token_accuracy": 0.1215690866112709,
|
|
"num_tokens": 1133838.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 6.832457637786865,
|
|
"epoch": 0.5516495402920497,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.6089,
|
|
"mean_token_accuracy": 0.11541855782270431,
|
|
"num_tokens": 1146005.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 6.898182916641235,
|
|
"epoch": 0.5570578691184424,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.6986,
|
|
"mean_token_accuracy": 0.11636286452412606,
|
|
"num_tokens": 1157328.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 6.86890025138855,
|
|
"epoch": 0.5624661979448351,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.5041,
|
|
"mean_token_accuracy": 0.1261405475437641,
|
|
"num_tokens": 1168592.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 6.840576457977295,
|
|
"epoch": 0.5678745267712277,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.6185,
|
|
"mean_token_accuracy": 0.12170294821262359,
|
|
"num_tokens": 1178680.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 6.846650314331055,
|
|
"epoch": 0.5732828555976204,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.5657,
|
|
"mean_token_accuracy": 0.12621748968958854,
|
|
"num_tokens": 1188825.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 6.873343515396118,
|
|
"epoch": 0.578691184424013,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.6197,
|
|
"mean_token_accuracy": 0.12520118281245232,
|
|
"num_tokens": 1201375.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 6.885719203948975,
|
|
"epoch": 0.5840995132504057,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.7389,
|
|
"mean_token_accuracy": 0.12146739810705184,
|
|
"num_tokens": 1211915.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 6.8768409252166744,
|
|
"epoch": 0.5895078420767983,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.6373,
|
|
"mean_token_accuracy": 0.12586278915405275,
|
|
"num_tokens": 1223151.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 6.8410814762115475,
|
|
"epoch": 0.5949161709031909,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.5612,
|
|
"mean_token_accuracy": 0.1256933443248272,
|
|
"num_tokens": 1234229.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 6.868610095977783,
|
|
"epoch": 0.6003244997295836,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.5732,
|
|
"mean_token_accuracy": 0.1251695491373539,
|
|
"num_tokens": 1245298.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 6.753601741790772,
|
|
"epoch": 0.6057328285559762,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.5604,
|
|
"mean_token_accuracy": 0.12367500364780426,
|
|
"num_tokens": 1256154.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 6.8078868865966795,
|
|
"epoch": 0.6111411573823688,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.4867,
|
|
"mean_token_accuracy": 0.12721323147416114,
|
|
"num_tokens": 1267404.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 6.796451997756958,
|
|
"epoch": 0.6165494862087615,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.5311,
|
|
"mean_token_accuracy": 0.12494980618357658,
|
|
"num_tokens": 1278894.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 6.720667362213135,
|
|
"epoch": 0.6219578150351541,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.5174,
|
|
"mean_token_accuracy": 0.1313262701034546,
|
|
"num_tokens": 1290408.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 6.79968957901001,
|
|
"epoch": 0.6273661438615468,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.5932,
|
|
"mean_token_accuracy": 0.12316951155662537,
|
|
"num_tokens": 1301508.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 6.8444901466369625,
|
|
"epoch": 0.6327744726879394,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.5698,
|
|
"mean_token_accuracy": 0.1288432337343693,
|
|
"num_tokens": 1312551.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 6.841915416717529,
|
|
"epoch": 0.638182801514332,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.6691,
|
|
"mean_token_accuracy": 0.12365695238113403,
|
|
"num_tokens": 1324876.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 6.743932723999023,
|
|
"epoch": 0.6435911303407247,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.5384,
|
|
"mean_token_accuracy": 0.12478130236268044,
|
|
"num_tokens": 1335401.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 6.72650032043457,
|
|
"epoch": 0.6489994591671173,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.4655,
|
|
"mean_token_accuracy": 0.12780711874365808,
|
|
"num_tokens": 1348265.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 6.636385107040406,
|
|
"epoch": 0.65440778799351,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.4219,
|
|
"mean_token_accuracy": 0.13813963606953622,
|
|
"num_tokens": 1359078.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 6.532011079788208,
|
|
"epoch": 0.6598161168199026,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.3346,
|
|
"mean_token_accuracy": 0.12941448390483856,
|
|
"num_tokens": 1369499.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 6.7241157531738285,
|
|
"epoch": 0.6652244456462953,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.4365,
|
|
"mean_token_accuracy": 0.12599155828356742,
|
|
"num_tokens": 1380329.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 6.65777325630188,
|
|
"epoch": 0.670632774472688,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.5301,
|
|
"mean_token_accuracy": 0.12588285207748412,
|
|
"num_tokens": 1393047.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 6.6903739929199215,
|
|
"epoch": 0.6760411032990806,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.5108,
|
|
"mean_token_accuracy": 0.1278846800327301,
|
|
"num_tokens": 1405600.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 6.655543231964112,
|
|
"epoch": 0.6814494321254733,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.4692,
|
|
"mean_token_accuracy": 0.12778620719909667,
|
|
"num_tokens": 1416670.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 6.649730110168457,
|
|
"epoch": 0.6868577609518659,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.4681,
|
|
"mean_token_accuracy": 0.1273884281516075,
|
|
"num_tokens": 1428929.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 6.69134488105774,
|
|
"epoch": 0.6922660897782585,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.3925,
|
|
"mean_token_accuracy": 0.12978972792625426,
|
|
"num_tokens": 1439298.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 6.6626770973205565,
|
|
"epoch": 0.6976744186046512,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.4936,
|
|
"mean_token_accuracy": 0.13316625058650972,
|
|
"num_tokens": 1449853.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 6.727707958221435,
|
|
"epoch": 0.7030827474310438,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.5531,
|
|
"mean_token_accuracy": 0.1245946742594242,
|
|
"num_tokens": 1460816.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 6.585225248336792,
|
|
"epoch": 0.7084910762574365,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.4105,
|
|
"mean_token_accuracy": 0.12925421670079232,
|
|
"num_tokens": 1473003.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 6.654827499389649,
|
|
"epoch": 0.7138994050838291,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.5262,
|
|
"mean_token_accuracy": 0.12104681655764579,
|
|
"num_tokens": 1484745.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 6.636510896682739,
|
|
"epoch": 0.7193077339102217,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.4073,
|
|
"mean_token_accuracy": 0.1321321666240692,
|
|
"num_tokens": 1495803.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 6.633009243011474,
|
|
"epoch": 0.7247160627366144,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.4088,
|
|
"mean_token_accuracy": 0.1358610361814499,
|
|
"num_tokens": 1507504.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 6.664580249786377,
|
|
"epoch": 0.730124391563007,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.5178,
|
|
"mean_token_accuracy": 0.12738374322652818,
|
|
"num_tokens": 1518765.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 6.511888456344605,
|
|
"epoch": 0.7355327203893997,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.3856,
|
|
"mean_token_accuracy": 0.13164084926247596,
|
|
"num_tokens": 1529481.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 6.642051362991333,
|
|
"epoch": 0.7409410492157923,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.4132,
|
|
"mean_token_accuracy": 0.13513489440083504,
|
|
"num_tokens": 1539612.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 6.646273422241211,
|
|
"epoch": 0.7463493780421849,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.5721,
|
|
"mean_token_accuracy": 0.12754858955740928,
|
|
"num_tokens": 1551370.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 6.6850049018859865,
|
|
"epoch": 0.7517577068685776,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.4495,
|
|
"mean_token_accuracy": 0.13616673350334169,
|
|
"num_tokens": 1561662.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 6.5182334899902346,
|
|
"epoch": 0.7571660356949702,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.4259,
|
|
"mean_token_accuracy": 0.1308984115719795,
|
|
"num_tokens": 1573479.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 6.6419507503509525,
|
|
"epoch": 0.7625743645213628,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.3767,
|
|
"mean_token_accuracy": 0.1341143622994423,
|
|
"num_tokens": 1583684.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 6.539843273162842,
|
|
"epoch": 0.7679826933477556,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.4271,
|
|
"mean_token_accuracy": 0.13350974842906,
|
|
"num_tokens": 1595694.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 6.517860841751099,
|
|
"epoch": 0.7733910221741482,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.4161,
|
|
"mean_token_accuracy": 0.12861086875200273,
|
|
"num_tokens": 1606104.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 6.653214359283448,
|
|
"epoch": 0.7787993510005409,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.4917,
|
|
"mean_token_accuracy": 0.13703173398971558,
|
|
"num_tokens": 1617368.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 6.583222913742065,
|
|
"epoch": 0.7842076798269335,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.5023,
|
|
"mean_token_accuracy": 0.12887703627347946,
|
|
"num_tokens": 1630485.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 6.571360731124878,
|
|
"epoch": 0.7896160086533262,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.3358,
|
|
"mean_token_accuracy": 0.13361840322613716,
|
|
"num_tokens": 1641411.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 6.496680021286011,
|
|
"epoch": 0.7950243374797188,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.3417,
|
|
"mean_token_accuracy": 0.1377808481454849,
|
|
"num_tokens": 1653737.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.627929592132569,
|
|
"epoch": 0.8004326663061114,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.5447,
|
|
"mean_token_accuracy": 0.12955381646752356,
|
|
"num_tokens": 1665910.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 6.509966039657593,
|
|
"epoch": 0.8058409951325041,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.3154,
|
|
"mean_token_accuracy": 0.1310278929769993,
|
|
"num_tokens": 1676927.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.544528341293335,
|
|
"epoch": 0.8112493239588967,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.3675,
|
|
"mean_token_accuracy": 0.13159878849983214,
|
|
"num_tokens": 1688167.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 6.473386955261231,
|
|
"epoch": 0.8166576527852893,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.4376,
|
|
"mean_token_accuracy": 0.136484544724226,
|
|
"num_tokens": 1699899.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 6.567677068710327,
|
|
"epoch": 0.822065981611682,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.3601,
|
|
"mean_token_accuracy": 0.1403407372534275,
|
|
"num_tokens": 1709862.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.378448438644409,
|
|
"epoch": 0.8274743104380746,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.3464,
|
|
"mean_token_accuracy": 0.13634308502078057,
|
|
"num_tokens": 1719902.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.528745365142822,
|
|
"epoch": 0.8328826392644673,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.3591,
|
|
"mean_token_accuracy": 0.13441544622182847,
|
|
"num_tokens": 1730351.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.510399198532104,
|
|
"epoch": 0.8382909680908599,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.3384,
|
|
"mean_token_accuracy": 0.13328076303005218,
|
|
"num_tokens": 1741783.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.322606420516967,
|
|
"epoch": 0.8436992969172525,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.3697,
|
|
"mean_token_accuracy": 0.13814696222543715,
|
|
"num_tokens": 1752952.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.606351613998413,
|
|
"epoch": 0.8491076257436452,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.3959,
|
|
"mean_token_accuracy": 0.13247718662023544,
|
|
"num_tokens": 1764014.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.447211217880249,
|
|
"epoch": 0.8545159545700378,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.3455,
|
|
"mean_token_accuracy": 0.13544300198554993,
|
|
"num_tokens": 1776154.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.516087865829467,
|
|
"epoch": 0.8599242833964305,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.4172,
|
|
"mean_token_accuracy": 0.13449665680527687,
|
|
"num_tokens": 1787292.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.481107139587403,
|
|
"epoch": 0.8653326122228231,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.3821,
|
|
"mean_token_accuracy": 0.13621541485190392,
|
|
"num_tokens": 1799469.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.4164855003356935,
|
|
"epoch": 0.8707409410492158,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.2457,
|
|
"mean_token_accuracy": 0.14523041546344756,
|
|
"num_tokens": 1810091.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.484155941009521,
|
|
"epoch": 0.8761492698756085,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.405,
|
|
"mean_token_accuracy": 0.13718786239624023,
|
|
"num_tokens": 1820835.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.451174211502075,
|
|
"epoch": 0.8815575987020011,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.364,
|
|
"mean_token_accuracy": 0.13937411159276963,
|
|
"num_tokens": 1831868.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.407666540145874,
|
|
"epoch": 0.8869659275283938,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.3273,
|
|
"mean_token_accuracy": 0.13826591596007348,
|
|
"num_tokens": 1842040.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.53410325050354,
|
|
"epoch": 0.8923742563547864,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.4122,
|
|
"mean_token_accuracy": 0.1363173633813858,
|
|
"num_tokens": 1853765.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.515021228790284,
|
|
"epoch": 0.897782585181179,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.4173,
|
|
"mean_token_accuracy": 0.13506757691502572,
|
|
"num_tokens": 1866081.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.419595050811767,
|
|
"epoch": 0.9031909140075717,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.3213,
|
|
"mean_token_accuracy": 0.1396598130464554,
|
|
"num_tokens": 1876754.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.540018558502197,
|
|
"epoch": 0.9085992428339643,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.4374,
|
|
"mean_token_accuracy": 0.13417667225003244,
|
|
"num_tokens": 1888299.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.325070524215699,
|
|
"epoch": 0.914007571660357,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.2112,
|
|
"mean_token_accuracy": 0.14675599336624146,
|
|
"num_tokens": 1899027.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.455481433868409,
|
|
"epoch": 0.9194159004867496,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.3336,
|
|
"mean_token_accuracy": 0.13757723793387414,
|
|
"num_tokens": 1911043.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.381244230270386,
|
|
"epoch": 0.9248242293131422,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.2676,
|
|
"mean_token_accuracy": 0.13861764669418336,
|
|
"num_tokens": 1921295.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.472774600982666,
|
|
"epoch": 0.9302325581395349,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.3636,
|
|
"mean_token_accuracy": 0.14171690344810486,
|
|
"num_tokens": 1932944.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.311262941360473,
|
|
"epoch": 0.9356408869659275,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.2982,
|
|
"mean_token_accuracy": 0.13167063891887665,
|
|
"num_tokens": 1944072.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.397356128692627,
|
|
"epoch": 0.9410492157923201,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.3051,
|
|
"mean_token_accuracy": 0.14503950625658035,
|
|
"num_tokens": 1954798.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.3590000629425045,
|
|
"epoch": 0.9464575446187128,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.2638,
|
|
"mean_token_accuracy": 0.14990901201963425,
|
|
"num_tokens": 1965304.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.482557010650635,
|
|
"epoch": 0.9518658734451054,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.3088,
|
|
"mean_token_accuracy": 0.13219384998083114,
|
|
"num_tokens": 1976797.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.331461238861084,
|
|
"epoch": 0.9572742022714981,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.335,
|
|
"mean_token_accuracy": 0.13832038938999175,
|
|
"num_tokens": 1987448.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.1668407917022705,
|
|
"epoch": 0.9626825310978907,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.1689,
|
|
"mean_token_accuracy": 0.13529722392559052,
|
|
"num_tokens": 2000948.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.391203689575195,
|
|
"epoch": 0.9680908599242833,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.2438,
|
|
"mean_token_accuracy": 0.13901959732174873,
|
|
"num_tokens": 2011807.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.355871772766113,
|
|
"epoch": 0.9734991887506761,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.3258,
|
|
"mean_token_accuracy": 0.13166970759630203,
|
|
"num_tokens": 2022835.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.3004621982574465,
|
|
"epoch": 0.9789075175770687,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.233,
|
|
"mean_token_accuracy": 0.13900147005915642,
|
|
"num_tokens": 2034331.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.364811372756958,
|
|
"epoch": 0.9843158464034614,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.242,
|
|
"mean_token_accuracy": 0.13642507493495942,
|
|
"num_tokens": 2047002.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.2332769393920895,
|
|
"epoch": 0.989724175229854,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.1739,
|
|
"mean_token_accuracy": 0.14840540438890457,
|
|
"num_tokens": 2056299.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.289453458786011,
|
|
"epoch": 0.9951325040562466,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.2991,
|
|
"mean_token_accuracy": 0.14142733812332153,
|
|
"num_tokens": 2068031.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.354425801171197,
|
|
"epoch": 1.0,
|
|
"grad_norm": 1.9296875,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.2578,
|
|
"mean_token_accuracy": 0.1445001173350546,
|
|
"num_tokens": 2076889.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.325593137741089,
|
|
"epoch": 1.0054083288263926,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.0941,
|
|
"mean_token_accuracy": 0.1489081636071205,
|
|
"num_tokens": 2087316.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.279916524887085,
|
|
"epoch": 1.0108166576527853,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.0993,
|
|
"mean_token_accuracy": 0.1413825236260891,
|
|
"num_tokens": 2098522.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.153361511230469,
|
|
"epoch": 1.016224986479178,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.0827,
|
|
"mean_token_accuracy": 0.14982683211565018,
|
|
"num_tokens": 2110015.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.294631814956665,
|
|
"epoch": 1.0216333153055706,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.1968,
|
|
"mean_token_accuracy": 0.13932642787694932,
|
|
"num_tokens": 2121333.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.309599113464356,
|
|
"epoch": 1.0270416441319632,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.0672,
|
|
"mean_token_accuracy": 0.15198333263397218,
|
|
"num_tokens": 2131385.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.302496814727784,
|
|
"epoch": 1.0324499729583558,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.1164,
|
|
"mean_token_accuracy": 0.15040989816188813,
|
|
"num_tokens": 2143274.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.08814058303833,
|
|
"epoch": 1.0378583017847485,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.0022,
|
|
"mean_token_accuracy": 0.150777405500412,
|
|
"num_tokens": 2154529.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.278039503097534,
|
|
"epoch": 1.043266630611141,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.1564,
|
|
"mean_token_accuracy": 0.1439770504832268,
|
|
"num_tokens": 2165949.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.256474876403809,
|
|
"epoch": 1.0486749594375337,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.1395,
|
|
"mean_token_accuracy": 0.14725938364863395,
|
|
"num_tokens": 2176710.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.2467734813690186,
|
|
"epoch": 1.0540832882639264,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.1028,
|
|
"mean_token_accuracy": 0.15126092433929444,
|
|
"num_tokens": 2189470.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.138709020614624,
|
|
"epoch": 1.059491617090319,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.0725,
|
|
"mean_token_accuracy": 0.15428618490695953,
|
|
"num_tokens": 2200266.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.209649991989136,
|
|
"epoch": 1.0648999459167117,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.086,
|
|
"mean_token_accuracy": 0.15003882795572282,
|
|
"num_tokens": 2210736.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.291779804229736,
|
|
"epoch": 1.0703082747431043,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.197,
|
|
"mean_token_accuracy": 0.14641276448965074,
|
|
"num_tokens": 2221057.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.208144187927246,
|
|
"epoch": 1.075716603569497,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.1826,
|
|
"mean_token_accuracy": 0.14815677255392073,
|
|
"num_tokens": 2232352.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.239853429794311,
|
|
"epoch": 1.0811249323958896,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.0704,
|
|
"mean_token_accuracy": 0.1508354812860489,
|
|
"num_tokens": 2244655.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 1.0811249323958896,
|
|
"eval_entropy": 6.042540579492395,
|
|
"eval_loss": 6.387476444244385,
|
|
"eval_mean_token_accuracy": 0.14328089801763946,
|
|
"eval_num_tokens": 2244655.0,
|
|
"eval_runtime": 2.5132,
|
|
"eval_samples_per_second": 1398.64,
|
|
"eval_steps_per_second": 175.079,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.158850622177124,
|
|
"epoch": 1.0865332612222822,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004999997383518051,
|
|
"loss": 6.0398,
|
|
"mean_token_accuracy": 0.15552402436733245,
|
|
"num_tokens": 2255699.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.152651357650757,
|
|
"epoch": 1.0919415900486749,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999986754070562,
|
|
"loss": 6.0174,
|
|
"mean_token_accuracy": 0.15807413756847383,
|
|
"num_tokens": 2266627.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.259909152984619,
|
|
"epoch": 1.0973499188750675,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004999967948166009,
|
|
"loss": 6.2279,
|
|
"mean_token_accuracy": 0.14398918375372888,
|
|
"num_tokens": 2277814.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.2683977603912355,
|
|
"epoch": 1.1027582477014604,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004999940965872734,
|
|
"loss": 6.1558,
|
|
"mean_token_accuracy": 0.15007070675492287,
|
|
"num_tokens": 2287344.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.0631325244903564,
|
|
"epoch": 1.1081665765278528,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999905807288788,
|
|
"loss": 5.9394,
|
|
"mean_token_accuracy": 0.15720777958631516,
|
|
"num_tokens": 2298332.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.224096870422363,
|
|
"epoch": 1.1135749053542456,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000499986247254194,
|
|
"loss": 6.0793,
|
|
"mean_token_accuracy": 0.1513406902551651,
|
|
"num_tokens": 2309161.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.102398872375488,
|
|
"epoch": 1.1189832341806383,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004999810961789667,
|
|
"loss": 6.07,
|
|
"mean_token_accuracy": 0.15258585810661315,
|
|
"num_tokens": 2319516.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.118391466140747,
|
|
"epoch": 1.124391563007031,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000499975127521916,
|
|
"loss": 5.961,
|
|
"mean_token_accuracy": 0.15804129093885422,
|
|
"num_tokens": 2330441.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.098996210098266,
|
|
"epoch": 1.1297998918334236,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004999683413047319,
|
|
"loss": 5.9831,
|
|
"mean_token_accuracy": 0.15822130292654038,
|
|
"num_tokens": 2340524.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.0999085903167725,
|
|
"epoch": 1.1352082206598162,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004999607375520759,
|
|
"loss": 5.9807,
|
|
"mean_token_accuracy": 0.15443639159202577,
|
|
"num_tokens": 2353379.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.159690189361572,
|
|
"epoch": 1.1406165494862088,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999523162915794,
|
|
"loss": 6.167,
|
|
"mean_token_accuracy": 0.14501615166664122,
|
|
"num_tokens": 2365708.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.102793979644775,
|
|
"epoch": 1.1460248783126015,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004999430775538459,
|
|
"loss": 6.0713,
|
|
"mean_token_accuracy": 0.1473181888461113,
|
|
"num_tokens": 2375630.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.1075451374053955,
|
|
"epoch": 1.151433207138994,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999330213724487,
|
|
"loss": 6.0719,
|
|
"mean_token_accuracy": 0.14480418860912322,
|
|
"num_tokens": 2386885.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.081627559661865,
|
|
"epoch": 1.1568415359653867,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004999221477839319,
|
|
"loss": 6.1088,
|
|
"mean_token_accuracy": 0.14639888927340508,
|
|
"num_tokens": 2397676.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.224732160568237,
|
|
"epoch": 1.1622498647917794,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004999104568278104,
|
|
"loss": 6.1147,
|
|
"mean_token_accuracy": 0.14688487648963927,
|
|
"num_tokens": 2409039.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.17805757522583,
|
|
"epoch": 1.167658193618172,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000499897948546569,
|
|
"loss": 6.1415,
|
|
"mean_token_accuracy": 0.1510010063648224,
|
|
"num_tokens": 2420103.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.0936055183410645,
|
|
"epoch": 1.1730665224445647,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004998846229856629,
|
|
"loss": 6.0593,
|
|
"mean_token_accuracy": 0.1524211898446083,
|
|
"num_tokens": 2430792.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.138086366653442,
|
|
"epoch": 1.1784748512709573,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004998704801935171,
|
|
"loss": 6.0535,
|
|
"mean_token_accuracy": 0.14968017786741256,
|
|
"num_tokens": 2442165.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.096742153167725,
|
|
"epoch": 1.18388318009735,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004998555202215267,
|
|
"loss": 5.9798,
|
|
"mean_token_accuracy": 0.1533754363656044,
|
|
"num_tokens": 2454231.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.096604728698731,
|
|
"epoch": 1.1892915089237426,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004998397431240561,
|
|
"loss": 6.03,
|
|
"mean_token_accuracy": 0.15928858369588852,
|
|
"num_tokens": 2464828.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.102307462692261,
|
|
"epoch": 1.1946998377501352,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004998231489584395,
|
|
"loss": 6.075,
|
|
"mean_token_accuracy": 0.14581840857863426,
|
|
"num_tokens": 2476144.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.272434854507447,
|
|
"epoch": 1.2001081665765279,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00049980573778498,
|
|
"loss": 6.252,
|
|
"mean_token_accuracy": 0.14115440845489502,
|
|
"num_tokens": 2489116.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.186100244522095,
|
|
"epoch": 1.2055164954029205,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004997875096669498,
|
|
"loss": 6.1067,
|
|
"mean_token_accuracy": 0.15317970514297485,
|
|
"num_tokens": 2501609.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.073341655731201,
|
|
"epoch": 1.2109248242293131,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004997684646705901,
|
|
"loss": 6.0154,
|
|
"mean_token_accuracy": 0.15728455781936646,
|
|
"num_tokens": 2513338.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.175504589080811,
|
|
"epoch": 1.2163331530557058,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004997486028651103,
|
|
"loss": 6.0208,
|
|
"mean_token_accuracy": 0.15885657221078872,
|
|
"num_tokens": 2523199.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.153501510620117,
|
|
"epoch": 1.2217414818820984,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004997279243226884,
|
|
"loss": 6.0323,
|
|
"mean_token_accuracy": 0.1503905773162842,
|
|
"num_tokens": 2535136.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.015810632705689,
|
|
"epoch": 1.227149810708491,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004997064291184701,
|
|
"loss": 5.9632,
|
|
"mean_token_accuracy": 0.14958094879984857,
|
|
"num_tokens": 2546428.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.139112186431885,
|
|
"epoch": 1.2325581395348837,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004996841173305691,
|
|
"loss": 6.1552,
|
|
"mean_token_accuracy": 0.153196282684803,
|
|
"num_tokens": 2558304.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.018696594238281,
|
|
"epoch": 1.2379664683612763,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004996609890400665,
|
|
"loss": 5.8364,
|
|
"mean_token_accuracy": 0.1650115966796875,
|
|
"num_tokens": 2568717.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.054702711105347,
|
|
"epoch": 1.243374797187669,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004996370443310107,
|
|
"loss": 6.0665,
|
|
"mean_token_accuracy": 0.14768287688493728,
|
|
"num_tokens": 2578748.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.168256664276123,
|
|
"epoch": 1.2487831260140616,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004996122832904165,
|
|
"loss": 6.0197,
|
|
"mean_token_accuracy": 0.15079330205917357,
|
|
"num_tokens": 2590095.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.085101318359375,
|
|
"epoch": 1.2541914548404542,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004995867060082658,
|
|
"loss": 6.041,
|
|
"mean_token_accuracy": 0.1540378987789154,
|
|
"num_tokens": 2600929.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.067340564727783,
|
|
"epoch": 1.2595997836668469,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004995603125775065,
|
|
"loss": 6.0717,
|
|
"mean_token_accuracy": 0.1497349411249161,
|
|
"num_tokens": 2611830.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.087981414794922,
|
|
"epoch": 1.2650081124932395,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004995331030940523,
|
|
"loss": 6.0094,
|
|
"mean_token_accuracy": 0.15056796371936798,
|
|
"num_tokens": 2623177.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.027497720718384,
|
|
"epoch": 1.2704164413196322,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004995050776567826,
|
|
"loss": 6.0018,
|
|
"mean_token_accuracy": 0.14762658923864364,
|
|
"num_tokens": 2635563.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.138033103942871,
|
|
"epoch": 1.2758247701460248,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004994762363675421,
|
|
"loss": 5.983,
|
|
"mean_token_accuracy": 0.1599634364247322,
|
|
"num_tokens": 2647157.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.068094396591187,
|
|
"epoch": 1.2812330989724177,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004994465793311396,
|
|
"loss": 5.9718,
|
|
"mean_token_accuracy": 0.1542554795742035,
|
|
"num_tokens": 2658175.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.021796751022339,
|
|
"epoch": 1.28664142779881,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004994161066553493,
|
|
"loss": 5.9302,
|
|
"mean_token_accuracy": 0.1576513960957527,
|
|
"num_tokens": 2669458.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.142000722885132,
|
|
"epoch": 1.292049756625203,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004993848184509086,
|
|
"loss": 6.1372,
|
|
"mean_token_accuracy": 0.15316924899816514,
|
|
"num_tokens": 2679727.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.0802788734436035,
|
|
"epoch": 1.2974580854515954,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004993527148315193,
|
|
"loss": 6.0873,
|
|
"mean_token_accuracy": 0.15363624691963196,
|
|
"num_tokens": 2691031.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.037839460372925,
|
|
"epoch": 1.3028664142779882,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004993197959138458,
|
|
"loss": 5.878,
|
|
"mean_token_accuracy": 0.1630819708108902,
|
|
"num_tokens": 2701699.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.004513549804687,
|
|
"epoch": 1.3082747431043806,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004992860618175156,
|
|
"loss": 5.9301,
|
|
"mean_token_accuracy": 0.15601511746644975,
|
|
"num_tokens": 2713038.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.00040225982666,
|
|
"epoch": 1.3136830719307735,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004992515126651184,
|
|
"loss": 5.9313,
|
|
"mean_token_accuracy": 0.15195388048887254,
|
|
"num_tokens": 2723685.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 5.98259425163269,
|
|
"epoch": 1.319091400757166,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004992161485822059,
|
|
"loss": 5.9956,
|
|
"mean_token_accuracy": 0.15625423043966294,
|
|
"num_tokens": 2735341.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.081094646453858,
|
|
"epoch": 1.3244997295835588,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004991799696972913,
|
|
"loss": 5.9448,
|
|
"mean_token_accuracy": 0.16045550853013993,
|
|
"num_tokens": 2746052.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 5.9843775749206545,
|
|
"epoch": 1.3299080584099512,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004991429761418488,
|
|
"loss": 5.9208,
|
|
"mean_token_accuracy": 0.15683557689189911,
|
|
"num_tokens": 2757155.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.003798341751098,
|
|
"epoch": 1.335316387236344,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000499105168050313,
|
|
"loss": 5.9912,
|
|
"mean_token_accuracy": 0.15128641128540038,
|
|
"num_tokens": 2769631.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.172780179977417,
|
|
"epoch": 1.3407247160627367,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004990665455600787,
|
|
"loss": 6.0889,
|
|
"mean_token_accuracy": 0.1538678914308548,
|
|
"num_tokens": 2780797.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 5.934348773956299,
|
|
"epoch": 1.3461330448891293,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004990271088114999,
|
|
"loss": 5.9352,
|
|
"mean_token_accuracy": 0.15413960218429565,
|
|
"num_tokens": 2792721.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.062621164321899,
|
|
"epoch": 1.351541373715522,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004989868579478901,
|
|
"loss": 5.9975,
|
|
"mean_token_accuracy": 0.165851429104805,
|
|
"num_tokens": 2804531.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.078337669372559,
|
|
"epoch": 1.3569497025419146,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004989457931155209,
|
|
"loss": 5.9846,
|
|
"mean_token_accuracy": 0.15587121546268462,
|
|
"num_tokens": 2816666.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.028954362869262,
|
|
"epoch": 1.3623580313683072,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004989039144636219,
|
|
"loss": 6.0218,
|
|
"mean_token_accuracy": 0.15586026459932328,
|
|
"num_tokens": 2827977.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.015844631195068,
|
|
"epoch": 1.3677663601946999,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004988612221443804,
|
|
"loss": 5.9858,
|
|
"mean_token_accuracy": 0.15803983360528945,
|
|
"num_tokens": 2839069.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.128492832183838,
|
|
"epoch": 1.3731746890210925,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004988177163129403,
|
|
"loss": 6.0716,
|
|
"mean_token_accuracy": 0.1514231324195862,
|
|
"num_tokens": 2850632.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 5.996451568603516,
|
|
"epoch": 1.3785830178474852,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004987733971274018,
|
|
"loss": 5.9498,
|
|
"mean_token_accuracy": 0.15206701457500457,
|
|
"num_tokens": 2862204.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 5.948522901535034,
|
|
"epoch": 1.3839913466738778,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004987282647488211,
|
|
"loss": 5.9237,
|
|
"mean_token_accuracy": 0.16415921002626419,
|
|
"num_tokens": 2874010.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.024138259887695,
|
|
"epoch": 1.3893996755002704,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004986823193412093,
|
|
"loss": 5.971,
|
|
"mean_token_accuracy": 0.15936093926429748,
|
|
"num_tokens": 2884315.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.162558841705322,
|
|
"epoch": 1.394808004326663,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.000498635561071532,
|
|
"loss": 6.143,
|
|
"mean_token_accuracy": 0.15060288906097413,
|
|
"num_tokens": 2896327.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 5.934901571273803,
|
|
"epoch": 1.4002163331530557,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000498587990109709,
|
|
"loss": 5.9009,
|
|
"mean_token_accuracy": 0.158343605697155,
|
|
"num_tokens": 2907007.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.026330184936524,
|
|
"epoch": 1.4056246619794484,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004985396066286133,
|
|
"loss": 5.9326,
|
|
"mean_token_accuracy": 0.16464538872241974,
|
|
"num_tokens": 2918852.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 5.981988334655762,
|
|
"epoch": 1.411032990805841,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004984904108040706,
|
|
"loss": 5.8667,
|
|
"mean_token_accuracy": 0.1570258155465126,
|
|
"num_tokens": 2929948.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 5.918032550811768,
|
|
"epoch": 1.4164413196322336,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004984404028148585,
|
|
"loss": 5.9327,
|
|
"mean_token_accuracy": 0.15995289385318756,
|
|
"num_tokens": 2941299.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 5.94100399017334,
|
|
"epoch": 1.4218496484586263,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004983895828427062,
|
|
"loss": 5.8532,
|
|
"mean_token_accuracy": 0.15711503773927687,
|
|
"num_tokens": 2952283.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.226106405258179,
|
|
"epoch": 1.427257977285019,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004983379510722935,
|
|
"loss": 6.2186,
|
|
"mean_token_accuracy": 0.15093351230025293,
|
|
"num_tokens": 2962241.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 5.982980823516845,
|
|
"epoch": 1.4326663061114115,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004982855076912506,
|
|
"loss": 5.9952,
|
|
"mean_token_accuracy": 0.15680053383111953,
|
|
"num_tokens": 2973408.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.024939918518067,
|
|
"epoch": 1.4380746349378042,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004982322528901568,
|
|
"loss": 5.9931,
|
|
"mean_token_accuracy": 0.15953508466482164,
|
|
"num_tokens": 2985877.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.096973657608032,
|
|
"epoch": 1.4434829637641968,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00049817818686254,
|
|
"loss": 6.0046,
|
|
"mean_token_accuracy": 0.15548449903726577,
|
|
"num_tokens": 2996935.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 5.963697910308838,
|
|
"epoch": 1.4488912925905895,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004981233098048765,
|
|
"loss": 5.8872,
|
|
"mean_token_accuracy": 0.16358304619789124,
|
|
"num_tokens": 3007932.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.067618751525879,
|
|
"epoch": 1.454299621416982,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004980676219165894,
|
|
"loss": 5.984,
|
|
"mean_token_accuracy": 0.15620877593755722,
|
|
"num_tokens": 3019368.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.075773096084594,
|
|
"epoch": 1.4597079502433747,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.000498011123400049,
|
|
"loss": 6.0475,
|
|
"mean_token_accuracy": 0.15290008410811423,
|
|
"num_tokens": 3029787.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 5.928175258636474,
|
|
"epoch": 1.4651162790697674,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004979538144605707,
|
|
"loss": 5.9897,
|
|
"mean_token_accuracy": 0.15913847908377649,
|
|
"num_tokens": 3042100.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.030663061141968,
|
|
"epoch": 1.47052460789616,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004978956953064154,
|
|
"loss": 6.0145,
|
|
"mean_token_accuracy": 0.1553100124001503,
|
|
"num_tokens": 3054160.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.035087013244629,
|
|
"epoch": 1.4759329367225527,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004978367661487883,
|
|
"loss": 5.9557,
|
|
"mean_token_accuracy": 0.15745455026626587,
|
|
"num_tokens": 3065401.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 5.920888233184814,
|
|
"epoch": 1.4813412655489453,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004977770272018379,
|
|
"loss": 5.9002,
|
|
"mean_token_accuracy": 0.15874640941619872,
|
|
"num_tokens": 3076137.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.074737930297852,
|
|
"epoch": 1.486749594375338,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000497716478682656,
|
|
"loss": 6.0321,
|
|
"mean_token_accuracy": 0.1597557932138443,
|
|
"num_tokens": 3087274.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.008933401107788,
|
|
"epoch": 1.4921579232017308,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004976551208112759,
|
|
"loss": 5.9202,
|
|
"mean_token_accuracy": 0.15890678018331528,
|
|
"num_tokens": 3098476.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 5.9916338443756105,
|
|
"epoch": 1.4975662520281232,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004975929538106721,
|
|
"loss": 5.949,
|
|
"mean_token_accuracy": 0.15909607857465743,
|
|
"num_tokens": 3110060.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 5.941988706588745,
|
|
"epoch": 1.502974580854516,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004975299779067598,
|
|
"loss": 5.8609,
|
|
"mean_token_accuracy": 0.16037650406360626,
|
|
"num_tokens": 3121032.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 5.976831722259521,
|
|
"epoch": 1.5083829096809085,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004974661933283936,
|
|
"loss": 5.8849,
|
|
"mean_token_accuracy": 0.16220622956752778,
|
|
"num_tokens": 3131999.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 5.8485579013824465,
|
|
"epoch": 1.5137912385073014,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004974016003073667,
|
|
"loss": 5.7999,
|
|
"mean_token_accuracy": 0.16209261864423752,
|
|
"num_tokens": 3142693.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 5.90942153930664,
|
|
"epoch": 1.5191995673336938,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004973361990784104,
|
|
"loss": 5.7695,
|
|
"mean_token_accuracy": 0.1575372710824013,
|
|
"num_tokens": 3154179.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 5.9297247409820555,
|
|
"epoch": 1.5246078961600866,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004972699898791927,
|
|
"loss": 5.8608,
|
|
"mean_token_accuracy": 0.1609016090631485,
|
|
"num_tokens": 3165314.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 5.8965638160705565,
|
|
"epoch": 1.530016224986479,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004972029729503182,
|
|
"loss": 5.9809,
|
|
"mean_token_accuracy": 0.1617598831653595,
|
|
"num_tokens": 3177646.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.036063766479492,
|
|
"epoch": 1.535424553812872,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004971351485353265,
|
|
"loss": 5.9681,
|
|
"mean_token_accuracy": 0.15694133341312408,
|
|
"num_tokens": 3188401.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 5.863464975357056,
|
|
"epoch": 1.5408328826392643,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004970665168806917,
|
|
"loss": 5.8353,
|
|
"mean_token_accuracy": 0.1589587315917015,
|
|
"num_tokens": 3200206.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.002946043014527,
|
|
"epoch": 1.5462412114656572,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004969970782358214,
|
|
"loss": 5.9443,
|
|
"mean_token_accuracy": 0.16198477745056153,
|
|
"num_tokens": 3211981.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 5.9493612289428714,
|
|
"epoch": 1.5516495402920496,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004969268328530557,
|
|
"loss": 5.9744,
|
|
"mean_token_accuracy": 0.15644426196813582,
|
|
"num_tokens": 3221997.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 5.988863134384156,
|
|
"epoch": 1.5570578691184425,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004968557809876666,
|
|
"loss": 5.9132,
|
|
"mean_token_accuracy": 0.1593450576066971,
|
|
"num_tokens": 3234333.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 5.99928731918335,
|
|
"epoch": 1.562466197944835,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004967839228978568,
|
|
"loss": 5.9548,
|
|
"mean_token_accuracy": 0.15260644033551216,
|
|
"num_tokens": 3245083.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 5.947020673751831,
|
|
"epoch": 1.5678745267712277,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004967112588447585,
|
|
"loss": 5.7966,
|
|
"mean_token_accuracy": 0.1688396155834198,
|
|
"num_tokens": 3255787.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 5.918097829818725,
|
|
"epoch": 1.5732828555976204,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004966377890924332,
|
|
"loss": 5.9637,
|
|
"mean_token_accuracy": 0.15856602489948274,
|
|
"num_tokens": 3267793.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 5.910263204574585,
|
|
"epoch": 1.578691184424013,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004965635139078704,
|
|
"loss": 5.8474,
|
|
"mean_token_accuracy": 0.169148188829422,
|
|
"num_tokens": 3279100.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.015842628479004,
|
|
"epoch": 1.5840995132504057,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004964884335609857,
|
|
"loss": 5.9925,
|
|
"mean_token_accuracy": 0.16395159661769867,
|
|
"num_tokens": 3290251.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 5.835095739364624,
|
|
"epoch": 1.5895078420767983,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004964125483246218,
|
|
"loss": 5.8594,
|
|
"mean_token_accuracy": 0.16018129587173463,
|
|
"num_tokens": 3301817.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.0568671226501465,
|
|
"epoch": 1.594916170903191,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004963358584745457,
|
|
"loss": 5.9311,
|
|
"mean_token_accuracy": 0.15858796536922454,
|
|
"num_tokens": 3313435.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 5.924640846252442,
|
|
"epoch": 1.6003244997295836,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004962583642894482,
|
|
"loss": 6.0232,
|
|
"mean_token_accuracy": 0.15672754794359206,
|
|
"num_tokens": 3325489.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.074223184585572,
|
|
"epoch": 1.6057328285559762,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004961800660509437,
|
|
"loss": 5.8956,
|
|
"mean_token_accuracy": 0.16288523972034455,
|
|
"num_tokens": 3337030.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 5.824729108810425,
|
|
"epoch": 1.6111411573823688,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000496100964043568,
|
|
"loss": 5.8671,
|
|
"mean_token_accuracy": 0.16309090554714203,
|
|
"num_tokens": 3347766.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 5.8139732837677,
|
|
"epoch": 1.6165494862087615,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000496021058554778,
|
|
"loss": 5.7893,
|
|
"mean_token_accuracy": 0.16337550133466722,
|
|
"num_tokens": 3360379.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 5.887569952011108,
|
|
"epoch": 1.6219578150351541,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004959403498749505,
|
|
"loss": 5.9175,
|
|
"mean_token_accuracy": 0.16475236117839814,
|
|
"num_tokens": 3370962.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"epoch": 1.6219578150351541,
|
|
"eval_entropy": 5.934710161252455,
|
|
"eval_loss": 6.146826267242432,
|
|
"eval_mean_token_accuracy": 0.15826588865708222,
|
|
"eval_num_tokens": 3370962.0,
|
|
"eval_runtime": 2.7496,
|
|
"eval_samples_per_second": 1278.365,
|
|
"eval_steps_per_second": 160.023,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 5.996402549743652,
|
|
"epoch": 1.6273661438615468,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004958588382973809,
|
|
"loss": 5.8312,
|
|
"mean_token_accuracy": 0.16115999966859818,
|
|
"num_tokens": 3382648.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 5.87752776145935,
|
|
"epoch": 1.6327744726879394,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004957765241182826,
|
|
"loss": 5.8939,
|
|
"mean_token_accuracy": 0.1591312602162361,
|
|
"num_tokens": 3393871.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 5.863660097122192,
|
|
"epoch": 1.638182801514332,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004956934076367854,
|
|
"loss": 5.7753,
|
|
"mean_token_accuracy": 0.16674366146326064,
|
|
"num_tokens": 3405148.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 5.8705549240112305,
|
|
"epoch": 1.6435911303407247,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000495609489154935,
|
|
"loss": 5.8111,
|
|
"mean_token_accuracy": 0.17316697388887406,
|
|
"num_tokens": 3416647.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 5.921846008300781,
|
|
"epoch": 1.6489994591671173,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004955247689776911,
|
|
"loss": 5.926,
|
|
"mean_token_accuracy": 0.15521429479122162,
|
|
"num_tokens": 3428331.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 5.833167028427124,
|
|
"epoch": 1.65440778799351,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000495439247412927,
|
|
"loss": 5.7838,
|
|
"mean_token_accuracy": 0.16418784856796265,
|
|
"num_tokens": 3440683.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 5.86015100479126,
|
|
"epoch": 1.6598161168199026,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004953529247714285,
|
|
"loss": 5.7582,
|
|
"mean_token_accuracy": 0.17050364911556243,
|
|
"num_tokens": 3452375.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 5.8904214859008786,
|
|
"epoch": 1.6652244456462952,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004952658013668922,
|
|
"loss": 5.9263,
|
|
"mean_token_accuracy": 0.15376651734113694,
|
|
"num_tokens": 3463255.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 5.954513168334961,
|
|
"epoch": 1.670632774472688,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004951778775159246,
|
|
"loss": 5.8578,
|
|
"mean_token_accuracy": 0.15923139825463295,
|
|
"num_tokens": 3473642.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 5.886024045944214,
|
|
"epoch": 1.6760411032990805,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004950891535380414,
|
|
"loss": 5.8353,
|
|
"mean_token_accuracy": 0.16338930279016495,
|
|
"num_tokens": 3484877.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 5.862248134613037,
|
|
"epoch": 1.6814494321254734,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004949996297556656,
|
|
"loss": 5.7601,
|
|
"mean_token_accuracy": 0.1690281018614769,
|
|
"num_tokens": 3495500.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 5.8944580078125,
|
|
"epoch": 1.6868577609518658,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004949093064941268,
|
|
"loss": 5.8732,
|
|
"mean_token_accuracy": 0.16452154666185378,
|
|
"num_tokens": 3508057.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 5.955048990249634,
|
|
"epoch": 1.6922660897782587,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004948181840816601,
|
|
"loss": 6.0391,
|
|
"mean_token_accuracy": 0.157832333445549,
|
|
"num_tokens": 3521110.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 5.900263166427612,
|
|
"epoch": 1.697674418604651,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004947262628494045,
|
|
"loss": 5.9507,
|
|
"mean_token_accuracy": 0.1666096642613411,
|
|
"num_tokens": 3530692.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 5.978882217407227,
|
|
"epoch": 1.703082747431044,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004946335431314018,
|
|
"loss": 5.904,
|
|
"mean_token_accuracy": 0.15454506278038024,
|
|
"num_tokens": 3542195.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 5.913450765609741,
|
|
"epoch": 1.7084910762574363,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004945400252645959,
|
|
"loss": 5.7732,
|
|
"mean_token_accuracy": 0.16944830417633056,
|
|
"num_tokens": 3552676.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 5.8387336254119875,
|
|
"epoch": 1.7138994050838292,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004944457095888308,
|
|
"loss": 5.7493,
|
|
"mean_token_accuracy": 0.16458534300327302,
|
|
"num_tokens": 3563308.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 5.85245270729065,
|
|
"epoch": 1.7193077339102216,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00049435059644685,
|
|
"loss": 5.7791,
|
|
"mean_token_accuracy": 0.16753556877374648,
|
|
"num_tokens": 3574085.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 5.887407684326172,
|
|
"epoch": 1.7247160627366145,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004942546861842947,
|
|
"loss": 5.8504,
|
|
"mean_token_accuracy": 0.16110641956329347,
|
|
"num_tokens": 3585816.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 5.863781261444092,
|
|
"epoch": 1.730124391563007,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004941579791497032,
|
|
"loss": 5.893,
|
|
"mean_token_accuracy": 0.1645564153790474,
|
|
"num_tokens": 3597776.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 5.99812536239624,
|
|
"epoch": 1.7355327203893998,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004940604756945091,
|
|
"loss": 5.9103,
|
|
"mean_token_accuracy": 0.16398233622312547,
|
|
"num_tokens": 3607991.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 5.796449804306031,
|
|
"epoch": 1.7409410492157922,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004939621761730399,
|
|
"loss": 5.8066,
|
|
"mean_token_accuracy": 0.16762966364622117,
|
|
"num_tokens": 3618204.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 5.946771907806396,
|
|
"epoch": 1.746349378042185,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004938630809425166,
|
|
"loss": 5.9245,
|
|
"mean_token_accuracy": 0.16191092431545256,
|
|
"num_tokens": 3629558.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 5.90002851486206,
|
|
"epoch": 1.7517577068685775,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004937631903630513,
|
|
"loss": 5.8598,
|
|
"mean_token_accuracy": 0.16230698376893998,
|
|
"num_tokens": 3640773.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 5.832959699630737,
|
|
"epoch": 1.7571660356949703,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004936625047976468,
|
|
"loss": 5.7704,
|
|
"mean_token_accuracy": 0.16658680737018586,
|
|
"num_tokens": 3652186.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 5.864456987380981,
|
|
"epoch": 1.7625743645213627,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004935610246121944,
|
|
"loss": 5.8168,
|
|
"mean_token_accuracy": 0.1693209633231163,
|
|
"num_tokens": 3663068.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 5.9234006881713865,
|
|
"epoch": 1.7679826933477556,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004934587501754736,
|
|
"loss": 5.968,
|
|
"mean_token_accuracy": 0.157489213347435,
|
|
"num_tokens": 3673803.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 6.049091863632202,
|
|
"epoch": 1.7733910221741482,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004933556818591498,
|
|
"loss": 5.9314,
|
|
"mean_token_accuracy": 0.16100268065929413,
|
|
"num_tokens": 3685482.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 5.813492918014527,
|
|
"epoch": 1.7787993510005409,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004932518200377736,
|
|
"loss": 5.7804,
|
|
"mean_token_accuracy": 0.17402340322732926,
|
|
"num_tokens": 3697641.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 5.849524593353271,
|
|
"epoch": 1.7842076798269335,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004931471650887789,
|
|
"loss": 5.8015,
|
|
"mean_token_accuracy": 0.16546109020709993,
|
|
"num_tokens": 3709560.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 6.003118276596069,
|
|
"epoch": 1.7896160086533262,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004930417173924822,
|
|
"loss": 5.9217,
|
|
"mean_token_accuracy": 0.15700541883707048,
|
|
"num_tokens": 3720422.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 5.727274179458618,
|
|
"epoch": 1.7950243374797188,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004929354773320806,
|
|
"loss": 5.7472,
|
|
"mean_token_accuracy": 0.1666373446583748,
|
|
"num_tokens": 3731104.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 5.875627565383911,
|
|
"epoch": 1.8004326663061114,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004928284452936506,
|
|
"loss": 5.9031,
|
|
"mean_token_accuracy": 0.1605694442987442,
|
|
"num_tokens": 3742766.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 5.919679355621338,
|
|
"epoch": 1.805840995132504,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004927206216661473,
|
|
"loss": 5.7879,
|
|
"mean_token_accuracy": 0.17063169926404953,
|
|
"num_tokens": 3753064.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 5.87092776298523,
|
|
"epoch": 1.8112493239588967,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004926120068414015,
|
|
"loss": 5.838,
|
|
"mean_token_accuracy": 0.1640581890940666,
|
|
"num_tokens": 3765183.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 5.84355616569519,
|
|
"epoch": 1.8166576527852893,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004925026012141201,
|
|
"loss": 5.825,
|
|
"mean_token_accuracy": 0.1628502182662487,
|
|
"num_tokens": 3777902.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 5.802776765823364,
|
|
"epoch": 1.822065981611682,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004923924051818832,
|
|
"loss": 5.793,
|
|
"mean_token_accuracy": 0.16486398428678511,
|
|
"num_tokens": 3788461.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 5.832666063308716,
|
|
"epoch": 1.8274743104380746,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004922814191451435,
|
|
"loss": 5.7382,
|
|
"mean_token_accuracy": 0.1673668459057808,
|
|
"num_tokens": 3799574.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 5.899340867996216,
|
|
"epoch": 1.8328826392644673,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004921696435072245,
|
|
"loss": 5.8689,
|
|
"mean_token_accuracy": 0.1626936286687851,
|
|
"num_tokens": 3811889.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 5.882519960403442,
|
|
"epoch": 1.83829096809086,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004920570786743191,
|
|
"loss": 5.8411,
|
|
"mean_token_accuracy": 0.1639084681868553,
|
|
"num_tokens": 3822949.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 5.875439453125,
|
|
"epoch": 1.8436992969172525,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004919437250554882,
|
|
"loss": 5.9286,
|
|
"mean_token_accuracy": 0.16021949350833892,
|
|
"num_tokens": 3834734.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 5.916995096206665,
|
|
"epoch": 1.8491076257436452,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004918295830626592,
|
|
"loss": 5.769,
|
|
"mean_token_accuracy": 0.17069644778966903,
|
|
"num_tokens": 3845508.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 5.7987555980682375,
|
|
"epoch": 1.8545159545700378,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004917146531106242,
|
|
"loss": 5.7626,
|
|
"mean_token_accuracy": 0.1679821029305458,
|
|
"num_tokens": 3856500.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 5.9299393653869625,
|
|
"epoch": 1.8599242833964305,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004915989356170389,
|
|
"loss": 5.84,
|
|
"mean_token_accuracy": 0.15926085114479066,
|
|
"num_tokens": 3867877.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 5.82513427734375,
|
|
"epoch": 1.865332612222823,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004914824310024213,
|
|
"loss": 5.8346,
|
|
"mean_token_accuracy": 0.17108957916498185,
|
|
"num_tokens": 3878613.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 5.831516408920288,
|
|
"epoch": 1.870740941049216,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004913651396901489,
|
|
"loss": 5.7505,
|
|
"mean_token_accuracy": 0.1657748118042946,
|
|
"num_tokens": 3888418.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 5.894839429855347,
|
|
"epoch": 1.8761492698756084,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000491247062106459,
|
|
"loss": 5.8331,
|
|
"mean_token_accuracy": 0.16697628796100616,
|
|
"num_tokens": 3899359.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 5.874541854858398,
|
|
"epoch": 1.8815575987020012,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004911281986804455,
|
|
"loss": 5.9236,
|
|
"mean_token_accuracy": 0.1603615999221802,
|
|
"num_tokens": 3911568.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 5.876946544647216,
|
|
"epoch": 1.8869659275283936,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004910085498440587,
|
|
"loss": 5.8719,
|
|
"mean_token_accuracy": 0.15552544146776198,
|
|
"num_tokens": 3921597.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 5.822331333160401,
|
|
"epoch": 1.8923742563547865,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004908881160321025,
|
|
"loss": 5.7574,
|
|
"mean_token_accuracy": 0.16052952408790588,
|
|
"num_tokens": 3933124.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 5.85541787147522,
|
|
"epoch": 1.897782585181179,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004907668976822338,
|
|
"loss": 5.7675,
|
|
"mean_token_accuracy": 0.17216205596923828,
|
|
"num_tokens": 3944388.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 5.807098245620727,
|
|
"epoch": 1.9031909140075718,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004906448952349603,
|
|
"loss": 5.7385,
|
|
"mean_token_accuracy": 0.16766344755887985,
|
|
"num_tokens": 3955024.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 5.701042556762696,
|
|
"epoch": 1.9085992428339642,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004905221091336392,
|
|
"loss": 5.7551,
|
|
"mean_token_accuracy": 0.17081061154603958,
|
|
"num_tokens": 3966029.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 5.839930725097656,
|
|
"epoch": 1.914007571660357,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004903985398244753,
|
|
"loss": 5.7179,
|
|
"mean_token_accuracy": 0.16770562827587127,
|
|
"num_tokens": 3976066.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 5.926287460327148,
|
|
"epoch": 1.9194159004867495,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004902741877565201,
|
|
"loss": 5.9192,
|
|
"mean_token_accuracy": 0.1603947177529335,
|
|
"num_tokens": 3986467.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 5.865671682357788,
|
|
"epoch": 1.9248242293131423,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000490149053381669,
|
|
"loss": 5.8872,
|
|
"mean_token_accuracy": 0.16689747869968413,
|
|
"num_tokens": 3998208.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 5.923223781585693,
|
|
"epoch": 1.9302325581395348,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004900231371546608,
|
|
"loss": 5.8281,
|
|
"mean_token_accuracy": 0.1686649963259697,
|
|
"num_tokens": 4008202.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 5.770446825027466,
|
|
"epoch": 1.9356408869659276,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004898964395330753,
|
|
"loss": 5.6637,
|
|
"mean_token_accuracy": 0.17363423705101014,
|
|
"num_tokens": 4019256.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 5.735243892669677,
|
|
"epoch": 1.94104921579232,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004897689609773318,
|
|
"loss": 5.703,
|
|
"mean_token_accuracy": 0.17118440419435502,
|
|
"num_tokens": 4029738.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 5.846836280822754,
|
|
"epoch": 1.946457544618713,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004896407019506879,
|
|
"loss": 5.9443,
|
|
"mean_token_accuracy": 0.15576391518115998,
|
|
"num_tokens": 4042277.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 5.9194403171539305,
|
|
"epoch": 1.9518658734451053,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004895116629192371,
|
|
"loss": 5.7863,
|
|
"mean_token_accuracy": 0.17253059893846512,
|
|
"num_tokens": 4054019.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 5.8024486064910885,
|
|
"epoch": 1.9572742022714982,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004893818443519074,
|
|
"loss": 5.8577,
|
|
"mean_token_accuracy": 0.16278256475925446,
|
|
"num_tokens": 4065562.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 5.93506007194519,
|
|
"epoch": 1.9626825310978906,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00048925124672046,
|
|
"loss": 5.8995,
|
|
"mean_token_accuracy": 0.16421243995428086,
|
|
"num_tokens": 4076617.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 5.859600019454956,
|
|
"epoch": 1.9680908599242835,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004891198704994868,
|
|
"loss": 5.7054,
|
|
"mean_token_accuracy": 0.17317854464054108,
|
|
"num_tokens": 4087191.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 5.792558240890503,
|
|
"epoch": 1.973499188750676,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004889877161664096,
|
|
"loss": 5.7365,
|
|
"mean_token_accuracy": 0.16982558965682984,
|
|
"num_tokens": 4099150.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 5.824551773071289,
|
|
"epoch": 1.9789075175770687,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004888547842014771,
|
|
"loss": 5.7528,
|
|
"mean_token_accuracy": 0.1684224784374237,
|
|
"num_tokens": 4110299.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 5.716506910324097,
|
|
"epoch": 1.9843158464034614,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004887210750877648,
|
|
"loss": 5.7781,
|
|
"mean_token_accuracy": 0.17140175253152848,
|
|
"num_tokens": 4121682.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 5.895252799987793,
|
|
"epoch": 1.989724175229854,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004885865893111716,
|
|
"loss": 5.7244,
|
|
"mean_token_accuracy": 0.16623667478561402,
|
|
"num_tokens": 4132914.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 5.765506172180176,
|
|
"epoch": 1.9951325040562466,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004884513273604195,
|
|
"loss": 5.8196,
|
|
"mean_token_accuracy": 0.16370895653963088,
|
|
"num_tokens": 4144213.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 5.865485191345215,
|
|
"epoch": 2.0,
|
|
"grad_norm": 2.09375,
|
|
"learning_rate": 0.0004883152897270502,
|
|
"loss": 5.849,
|
|
"mean_token_accuracy": 0.16912639306651223,
|
|
"num_tokens": 4153778.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 5.744277572631836,
|
|
"epoch": 2.005408328826393,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004881784769054253,
|
|
"loss": 5.4245,
|
|
"mean_token_accuracy": 0.17498673498630524,
|
|
"num_tokens": 4165536.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 5.830756425857544,
|
|
"epoch": 2.0108166576527853,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004880408893927226,
|
|
"loss": 5.4693,
|
|
"mean_token_accuracy": 0.17224047631025313,
|
|
"num_tokens": 4176182.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 5.637430953979492,
|
|
"epoch": 2.016224986479178,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004879025276889355,
|
|
"loss": 5.4799,
|
|
"mean_token_accuracy": 0.1830995872616768,
|
|
"num_tokens": 4187592.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 5.69356255531311,
|
|
"epoch": 2.0216333153055706,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00048776339229687077,
|
|
"loss": 5.2804,
|
|
"mean_token_accuracy": 0.19304971545934677,
|
|
"num_tokens": 4198199.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 5.668562746047973,
|
|
"epoch": 2.0270416441319634,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00048762348372214683,
|
|
"loss": 5.5222,
|
|
"mean_token_accuracy": 0.17638221979141236,
|
|
"num_tokens": 4208116.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 5.718329381942749,
|
|
"epoch": 2.032449972958356,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00048748280247319165,
|
|
"loss": 5.297,
|
|
"mean_token_accuracy": 0.18515555709600448,
|
|
"num_tokens": 4218569.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 5.681073713302612,
|
|
"epoch": 2.0378583017847487,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004873413490612412,
|
|
"loss": 5.442,
|
|
"mean_token_accuracy": 0.18598483353853226,
|
|
"num_tokens": 4228785.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 5.67579083442688,
|
|
"epoch": 2.043266630611141,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048719912400033757,
|
|
"loss": 5.4066,
|
|
"mean_token_accuracy": 0.18295987844467163,
|
|
"num_tokens": 4238814.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 5.643623304367066,
|
|
"epoch": 2.048674959437534,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.000487056127807327,
|
|
"loss": 5.3606,
|
|
"mean_token_accuracy": 0.17945607900619506,
|
|
"num_tokens": 4250544.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 5.67747540473938,
|
|
"epoch": 2.0540832882639264,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00048691236100185795,
|
|
"loss": 5.5014,
|
|
"mean_token_accuracy": 0.17847973108291626,
|
|
"num_tokens": 4261806.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 5.77795672416687,
|
|
"epoch": 2.0594916170903192,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004867678241063792,
|
|
"loss": 5.4492,
|
|
"mean_token_accuracy": 0.18175760209560393,
|
|
"num_tokens": 4272472.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 5.659222221374511,
|
|
"epoch": 2.0648999459167117,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048662251764613833,
|
|
"loss": 5.4238,
|
|
"mean_token_accuracy": 0.18123930245637893,
|
|
"num_tokens": 4283758.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 5.697822332382202,
|
|
"epoch": 2.0703082747431045,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004864764421491793,
|
|
"loss": 5.5471,
|
|
"mean_token_accuracy": 0.18098790496587752,
|
|
"num_tokens": 4295930.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 5.765074110031128,
|
|
"epoch": 2.075716603569497,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004863295981463408,
|
|
"loss": 5.4497,
|
|
"mean_token_accuracy": 0.19227266758680345,
|
|
"num_tokens": 4306711.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 5.589128541946411,
|
|
"epoch": 2.08112493239589,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004861819861712543,
|
|
"loss": 5.4423,
|
|
"mean_token_accuracy": 0.18505682647228242,
|
|
"num_tokens": 4317694.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 5.839343738555908,
|
|
"epoch": 2.086533261222282,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048603360676034203,
|
|
"loss": 5.5189,
|
|
"mean_token_accuracy": 0.17528488487005234,
|
|
"num_tokens": 4329552.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 5.62336049079895,
|
|
"epoch": 2.091941590048675,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00048588446045281505,
|
|
"loss": 5.4743,
|
|
"mean_token_accuracy": 0.18153496235609054,
|
|
"num_tokens": 4340479.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 5.624028635025025,
|
|
"epoch": 2.0973499188750675,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004857345477906714,
|
|
"loss": 5.351,
|
|
"mean_token_accuracy": 0.18419340401887893,
|
|
"num_tokens": 4351147.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 5.557870864868164,
|
|
"epoch": 2.1027582477014604,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00048558386931869417,
|
|
"loss": 5.3554,
|
|
"mean_token_accuracy": 0.18372577875852586,
|
|
"num_tokens": 4362962.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 5.652933120727539,
|
|
"epoch": 2.1081665765278528,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004854324255844492,
|
|
"loss": 5.4538,
|
|
"mean_token_accuracy": 0.18239828944206238,
|
|
"num_tokens": 4373723.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 5.602408456802368,
|
|
"epoch": 2.1135749053542456,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048528021713828336,
|
|
"loss": 5.4357,
|
|
"mean_token_accuracy": 0.18019760996103287,
|
|
"num_tokens": 4383434.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 5.613941431045532,
|
|
"epoch": 2.118983234180638,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004851272445333225,
|
|
"loss": 5.3704,
|
|
"mean_token_accuracy": 0.18846011757850648,
|
|
"num_tokens": 4394847.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 5.6594537734985355,
|
|
"epoch": 2.124391563007031,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00048497350832546937,
|
|
"loss": 5.4874,
|
|
"mean_token_accuracy": 0.18262122869491576,
|
|
"num_tokens": 4406367.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 5.743469142913819,
|
|
"epoch": 2.1297998918334233,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004848190090734018,
|
|
"loss": 5.478,
|
|
"mean_token_accuracy": 0.18379419893026352,
|
|
"num_tokens": 4417636.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 5.661316585540772,
|
|
"epoch": 2.135208220659816,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00048466374733857035,
|
|
"loss": 5.4855,
|
|
"mean_token_accuracy": 0.18583688586950303,
|
|
"num_tokens": 4428473.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 5.614023733139038,
|
|
"epoch": 2.1406165494862086,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004845077236851966,
|
|
"loss": 5.4258,
|
|
"mean_token_accuracy": 0.17982188165187835,
|
|
"num_tokens": 4439780.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 5.61946473121643,
|
|
"epoch": 2.1460248783126015,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00048435093868027076,
|
|
"loss": 5.3629,
|
|
"mean_token_accuracy": 0.1909126728773117,
|
|
"num_tokens": 4450929.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 5.619580936431885,
|
|
"epoch": 2.151433207138994,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00048419339289354995,
|
|
"loss": 5.4006,
|
|
"mean_token_accuracy": 0.18762292116880416,
|
|
"num_tokens": 4462967.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 5.6974211692810055,
|
|
"epoch": 2.1568415359653867,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000484035086897556,
|
|
"loss": 5.4979,
|
|
"mean_token_accuracy": 0.18536723852157594,
|
|
"num_tokens": 4473970.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 5.697462368011474,
|
|
"epoch": 2.162249864791779,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004838760212675732,
|
|
"loss": 5.5075,
|
|
"mean_token_accuracy": 0.17887972742319108,
|
|
"num_tokens": 4485660.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 2.162249864791779,
|
|
"eval_entropy": 5.622631918300282,
|
|
"eval_loss": 6.0652570724487305,
|
|
"eval_mean_token_accuracy": 0.16360555895688858,
|
|
"eval_num_tokens": 4485660.0,
|
|
"eval_runtime": 2.5188,
|
|
"eval_samples_per_second": 1395.505,
|
|
"eval_steps_per_second": 174.686,
|
|
"step": 2000
|
|
}
|
|
],
|
|
"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": 6984519736320000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|