2490fa797a
Model: fpadovani/eng-latn-100mb-after-ppt-shuff-dyck-10mb-ckpt500_seed3407 Source: Original Platform
8123 lines
221 KiB
JSON
8123 lines
221 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 4.324499729583558,
|
|
"eval_steps": 500,
|
|
"global_step": 4000,
|
|
"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
|
|
},
|
|
{
|
|
"entropy": 5.58496732711792,
|
|
"epoch": 2.167658193618172,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004837161965816464,
|
|
"loss": 5.4804,
|
|
"mean_token_accuracy": 0.18340193331241608,
|
|
"num_tokens": 4497261.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 5.701287508010864,
|
|
"epoch": 2.1730665224445644,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00048355561342057886,
|
|
"loss": 5.385,
|
|
"mean_token_accuracy": 0.1885501429438591,
|
|
"num_tokens": 4508120.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 5.645246505737305,
|
|
"epoch": 2.1784748512709573,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00048339427236793024,
|
|
"loss": 5.4366,
|
|
"mean_token_accuracy": 0.18911443650722504,
|
|
"num_tokens": 4519286.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 5.677903699874878,
|
|
"epoch": 2.1838831800973497,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004832321740100143,
|
|
"loss": 5.5222,
|
|
"mean_token_accuracy": 0.18122092485427857,
|
|
"num_tokens": 4531912.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 5.730960941314697,
|
|
"epoch": 2.1892915089237426,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048306931893589666,
|
|
"loss": 5.5212,
|
|
"mean_token_accuracy": 0.18324814438819886,
|
|
"num_tokens": 4544370.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 5.675964832305908,
|
|
"epoch": 2.194699837750135,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00048290570773739327,
|
|
"loss": 5.4797,
|
|
"mean_token_accuracy": 0.1786174088716507,
|
|
"num_tokens": 4556013.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 5.592317390441894,
|
|
"epoch": 2.200108166576528,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048274134100906747,
|
|
"loss": 5.4337,
|
|
"mean_token_accuracy": 0.18240178972482682,
|
|
"num_tokens": 4566200.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 5.671278762817383,
|
|
"epoch": 2.2055164954029207,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00048257621934822835,
|
|
"loss": 5.4688,
|
|
"mean_token_accuracy": 0.1822907105088234,
|
|
"num_tokens": 4576899.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 5.762304210662842,
|
|
"epoch": 2.210924824229313,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004824103433549285,
|
|
"loss": 5.4746,
|
|
"mean_token_accuracy": 0.17777852416038514,
|
|
"num_tokens": 4588983.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 5.58847827911377,
|
|
"epoch": 2.2163331530557056,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004822437136319616,
|
|
"loss": 5.453,
|
|
"mean_token_accuracy": 0.19082760214805602,
|
|
"num_tokens": 4599758.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 5.686668348312378,
|
|
"epoch": 2.2217414818820984,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004820763307848604,
|
|
"loss": 5.4722,
|
|
"mean_token_accuracy": 0.1838390588760376,
|
|
"num_tokens": 4610927.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 5.596033191680908,
|
|
"epoch": 2.2271498107084913,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048190819542189456,
|
|
"loss": 5.4989,
|
|
"mean_token_accuracy": 0.17988231480121614,
|
|
"num_tokens": 4623719.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 5.759393978118896,
|
|
"epoch": 2.2325581395348837,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048173930815406847,
|
|
"loss": 5.5535,
|
|
"mean_token_accuracy": 0.18174540549516677,
|
|
"num_tokens": 4635389.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 5.61832447052002,
|
|
"epoch": 2.2379664683612766,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004815696695951187,
|
|
"loss": 5.5003,
|
|
"mean_token_accuracy": 0.17951763421297073,
|
|
"num_tokens": 4645093.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 5.676296377182007,
|
|
"epoch": 2.243374797187669,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004813992803615123,
|
|
"loss": 5.4951,
|
|
"mean_token_accuracy": 0.17510253190994263,
|
|
"num_tokens": 4656490.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 5.678513431549073,
|
|
"epoch": 2.248783126014062,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00048122814107244414,
|
|
"loss": 5.5277,
|
|
"mean_token_accuracy": 0.17667777091264725,
|
|
"num_tokens": 4667603.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 5.614256143569946,
|
|
"epoch": 2.2541914548404542,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048105625234983466,
|
|
"loss": 5.4431,
|
|
"mean_token_accuracy": 0.17409378588199614,
|
|
"num_tokens": 4678669.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 5.52167911529541,
|
|
"epoch": 2.259599783666847,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00048088361481832807,
|
|
"loss": 5.2551,
|
|
"mean_token_accuracy": 0.19513772875070573,
|
|
"num_tokens": 4689757.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 5.569588136672974,
|
|
"epoch": 2.2650081124932395,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00048071022910528934,
|
|
"loss": 5.4355,
|
|
"mean_token_accuracy": 0.17989986538887023,
|
|
"num_tokens": 4699332.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 5.637401056289673,
|
|
"epoch": 2.2704164413196324,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048053609584080274,
|
|
"loss": 5.5162,
|
|
"mean_token_accuracy": 0.18639470934867858,
|
|
"num_tokens": 4709485.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 5.672516345977783,
|
|
"epoch": 2.275824770146025,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00048036121565766876,
|
|
"loss": 5.4487,
|
|
"mean_token_accuracy": 0.18197828978300096,
|
|
"num_tokens": 4720935.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 5.52376275062561,
|
|
"epoch": 2.2812330989724177,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004801855891914026,
|
|
"loss": 5.4163,
|
|
"mean_token_accuracy": 0.18085140138864517,
|
|
"num_tokens": 4731219.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 5.70998363494873,
|
|
"epoch": 2.28664142779881,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048000921708023117,
|
|
"loss": 5.5219,
|
|
"mean_token_accuracy": 0.17984937876462936,
|
|
"num_tokens": 4742637.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 5.606550073623657,
|
|
"epoch": 2.292049756625203,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004798320999650911,
|
|
"loss": 5.4245,
|
|
"mean_token_accuracy": 0.18459555357694626,
|
|
"num_tokens": 4753341.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 5.82206768989563,
|
|
"epoch": 2.2974580854515954,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00047965423848962636,
|
|
"loss": 5.5577,
|
|
"mean_token_accuracy": 0.1851441517472267,
|
|
"num_tokens": 4765339.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 5.615011978149414,
|
|
"epoch": 2.302866414277988,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000479475633300186,
|
|
"loss": 5.4284,
|
|
"mean_token_accuracy": 0.1830722913146019,
|
|
"num_tokens": 4776149.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 5.5949499130249025,
|
|
"epoch": 2.3082747431043806,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00047929628504582173,
|
|
"loss": 5.432,
|
|
"mean_token_accuracy": 0.18645608723163604,
|
|
"num_tokens": 4788120.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 5.659571266174316,
|
|
"epoch": 2.3136830719307735,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004791161943782856,
|
|
"loss": 5.539,
|
|
"mean_token_accuracy": 0.1774466961622238,
|
|
"num_tokens": 4799922.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 5.574285650253296,
|
|
"epoch": 2.319091400757166,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00047893536195202726,
|
|
"loss": 5.4762,
|
|
"mean_token_accuracy": 0.18149880617856978,
|
|
"num_tokens": 4810501.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 5.641510629653931,
|
|
"epoch": 2.3244997295835588,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004787537884241923,
|
|
"loss": 5.415,
|
|
"mean_token_accuracy": 0.1876547023653984,
|
|
"num_tokens": 4820919.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 5.659405660629273,
|
|
"epoch": 2.329908058409951,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00047857147445461943,
|
|
"loss": 5.505,
|
|
"mean_token_accuracy": 0.18379626274108887,
|
|
"num_tokens": 4833298.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 5.554553079605102,
|
|
"epoch": 2.335316387236344,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004783884207058379,
|
|
"loss": 5.3748,
|
|
"mean_token_accuracy": 0.18279899060726165,
|
|
"num_tokens": 4845199.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 5.572480297088623,
|
|
"epoch": 2.3407247160627365,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00047820462784306555,
|
|
"loss": 5.4371,
|
|
"mean_token_accuracy": 0.18337143808603287,
|
|
"num_tokens": 4856549.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 5.644783067703247,
|
|
"epoch": 2.3461330448891293,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000478020096534206,
|
|
"loss": 5.4903,
|
|
"mean_token_accuracy": 0.1812374100089073,
|
|
"num_tokens": 4868446.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 5.68095588684082,
|
|
"epoch": 2.3515413737155217,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00047783482744984644,
|
|
"loss": 5.5517,
|
|
"mean_token_accuracy": 0.17168335169553756,
|
|
"num_tokens": 4879047.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 5.601082277297974,
|
|
"epoch": 2.3569497025419146,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00047764882126325527,
|
|
"loss": 5.4324,
|
|
"mean_token_accuracy": 0.18012793660163878,
|
|
"num_tokens": 4891591.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 5.612620687484741,
|
|
"epoch": 2.362358031368307,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004774620786503793,
|
|
"loss": 5.4312,
|
|
"mean_token_accuracy": 0.18702213764190673,
|
|
"num_tokens": 4902710.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 5.628007221221924,
|
|
"epoch": 2.3677663601947,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004772746002898417,
|
|
"loss": 5.5122,
|
|
"mean_token_accuracy": 0.1824018105864525,
|
|
"num_tokens": 4914668.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 5.759065389633179,
|
|
"epoch": 2.3731746890210923,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00047708638686293903,
|
|
"loss": 5.6667,
|
|
"mean_token_accuracy": 0.17668095529079436,
|
|
"num_tokens": 4926684.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 5.694106674194336,
|
|
"epoch": 2.378583017847485,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00047689743905363963,
|
|
"loss": 5.5174,
|
|
"mean_token_accuracy": 0.18020099997520447,
|
|
"num_tokens": 4938577.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 5.488329076766968,
|
|
"epoch": 2.383991346673878,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004767077575485801,
|
|
"loss": 5.3969,
|
|
"mean_token_accuracy": 0.18497395962476731,
|
|
"num_tokens": 4949070.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 5.669815301895142,
|
|
"epoch": 2.3893996755002704,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004765173430370635,
|
|
"loss": 5.4608,
|
|
"mean_token_accuracy": 0.18446923792362213,
|
|
"num_tokens": 4959558.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 5.530538892745971,
|
|
"epoch": 2.394808004326663,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004763261962110567,
|
|
"loss": 5.3286,
|
|
"mean_token_accuracy": 0.18791096657514572,
|
|
"num_tokens": 4970310.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 5.522717714309692,
|
|
"epoch": 2.4002163331530557,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004761343177651875,
|
|
"loss": 5.4131,
|
|
"mean_token_accuracy": 0.18075056970119477,
|
|
"num_tokens": 4982300.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 5.613452339172364,
|
|
"epoch": 2.4056246619794486,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004759417083967428,
|
|
"loss": 5.4513,
|
|
"mean_token_accuracy": 0.18079365491867067,
|
|
"num_tokens": 4994425.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 5.544624471664429,
|
|
"epoch": 2.411032990805841,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00047574836880566544,
|
|
"loss": 5.4476,
|
|
"mean_token_accuracy": 0.18794148415327072,
|
|
"num_tokens": 5005872.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 5.590450191497803,
|
|
"epoch": 2.4164413196322334,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004755542996945519,
|
|
"loss": 5.4817,
|
|
"mean_token_accuracy": 0.1800764501094818,
|
|
"num_tokens": 5016903.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 5.658929872512817,
|
|
"epoch": 2.4218496484586263,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00047535950176864976,
|
|
"loss": 5.5351,
|
|
"mean_token_accuracy": 0.18202005475759506,
|
|
"num_tokens": 5028407.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 5.690558528900146,
|
|
"epoch": 2.427257977285019,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00047516397573585525,
|
|
"loss": 5.5267,
|
|
"mean_token_accuracy": 0.17972267419099808,
|
|
"num_tokens": 5039693.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 5.657332563400269,
|
|
"epoch": 2.4326663061114115,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00047496772230671034,
|
|
"loss": 5.4869,
|
|
"mean_token_accuracy": 0.18176871538162231,
|
|
"num_tokens": 5051279.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 5.648957586288452,
|
|
"epoch": 2.4380746349378044,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004747707421944004,
|
|
"loss": 5.4116,
|
|
"mean_token_accuracy": 0.18266654312610625,
|
|
"num_tokens": 5062698.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 5.565609931945801,
|
|
"epoch": 2.443482963764197,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004745730361147517,
|
|
"loss": 5.4125,
|
|
"mean_token_accuracy": 0.18356948643922805,
|
|
"num_tokens": 5074927.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 5.5671247959136965,
|
|
"epoch": 2.4488912925905897,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004743746047862286,
|
|
"loss": 5.5551,
|
|
"mean_token_accuracy": 0.17119887322187424,
|
|
"num_tokens": 5086298.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 5.692638397216797,
|
|
"epoch": 2.454299621416982,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004741754489299309,
|
|
"loss": 5.4862,
|
|
"mean_token_accuracy": 0.17928895801305772,
|
|
"num_tokens": 5098161.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 5.521544885635376,
|
|
"epoch": 2.459707950243375,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00047397556926959155,
|
|
"loss": 5.4649,
|
|
"mean_token_accuracy": 0.1891743004322052,
|
|
"num_tokens": 5110743.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 5.695449256896973,
|
|
"epoch": 2.4651162790697674,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00047377496653157374,
|
|
"loss": 5.5057,
|
|
"mean_token_accuracy": 0.18749736845493317,
|
|
"num_tokens": 5121838.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 5.5799483299255375,
|
|
"epoch": 2.4705246078961602,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00047357364144486806,
|
|
"loss": 5.4454,
|
|
"mean_token_accuracy": 0.1842237889766693,
|
|
"num_tokens": 5135047.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 5.58011474609375,
|
|
"epoch": 2.4759329367225527,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00047337159474109057,
|
|
"loss": 5.4408,
|
|
"mean_token_accuracy": 0.17357513606548308,
|
|
"num_tokens": 5146329.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 5.652484464645386,
|
|
"epoch": 2.4813412655489455,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00047316882715447934,
|
|
"loss": 5.4817,
|
|
"mean_token_accuracy": 0.18452474921941758,
|
|
"num_tokens": 5157511.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 5.566933536529541,
|
|
"epoch": 2.486749594375338,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004729653394218922,
|
|
"loss": 5.4674,
|
|
"mean_token_accuracy": 0.18247455805540086,
|
|
"num_tokens": 5168456.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 5.741812419891358,
|
|
"epoch": 2.492157923201731,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00047276113228280427,
|
|
"loss": 5.5195,
|
|
"mean_token_accuracy": 0.18097749799489976,
|
|
"num_tokens": 5179294.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 5.582517766952515,
|
|
"epoch": 2.497566252028123,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004725562064793045,
|
|
"loss": 5.4481,
|
|
"mean_token_accuracy": 0.18376519083976744,
|
|
"num_tokens": 5190195.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 5.489432907104492,
|
|
"epoch": 2.502974580854516,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00047235056275609397,
|
|
"loss": 5.353,
|
|
"mean_token_accuracy": 0.19146738052368165,
|
|
"num_tokens": 5201289.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 5.562600231170654,
|
|
"epoch": 2.5083829096809085,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004721442018604822,
|
|
"loss": 5.3749,
|
|
"mean_token_accuracy": 0.1949205666780472,
|
|
"num_tokens": 5212295.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 5.472233724594116,
|
|
"epoch": 2.5137912385073014,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004719371245423853,
|
|
"loss": 5.3607,
|
|
"mean_token_accuracy": 0.1919402688741684,
|
|
"num_tokens": 5223144.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 5.53741044998169,
|
|
"epoch": 2.5191995673336938,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00047172933155432273,
|
|
"loss": 5.3895,
|
|
"mean_token_accuracy": 0.18744857907295226,
|
|
"num_tokens": 5232840.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 5.617996311187744,
|
|
"epoch": 2.5246078961600866,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00047152082365141466,
|
|
"loss": 5.4831,
|
|
"mean_token_accuracy": 0.18032295107841492,
|
|
"num_tokens": 5243796.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 5.575401163101196,
|
|
"epoch": 2.530016224986479,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004713116015913794,
|
|
"loss": 5.5716,
|
|
"mean_token_accuracy": 0.17697452157735824,
|
|
"num_tokens": 5255990.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 5.7042511940002445,
|
|
"epoch": 2.535424553812872,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00047110166613453016,
|
|
"loss": 5.5183,
|
|
"mean_token_accuracy": 0.1835886225104332,
|
|
"num_tokens": 5268134.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 5.471448755264282,
|
|
"epoch": 2.5408328826392643,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00047089101804377305,
|
|
"loss": 5.3734,
|
|
"mean_token_accuracy": 0.18501282930374147,
|
|
"num_tokens": 5278662.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 5.656749820709228,
|
|
"epoch": 2.546241211465657,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004706796580846037,
|
|
"loss": 5.4525,
|
|
"mean_token_accuracy": 0.181876939535141,
|
|
"num_tokens": 5288382.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 5.624006462097168,
|
|
"epoch": 2.5516495402920496,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00047046758702510465,
|
|
"loss": 5.4891,
|
|
"mean_token_accuracy": 0.17903683930635453,
|
|
"num_tokens": 5299769.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 5.657060623168945,
|
|
"epoch": 2.5570578691184425,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00047025480563594264,
|
|
"loss": 5.4275,
|
|
"mean_token_accuracy": 0.1907404288649559,
|
|
"num_tokens": 5310662.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 5.519835710525513,
|
|
"epoch": 2.5624661979448353,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00047004131469036576,
|
|
"loss": 5.4148,
|
|
"mean_token_accuracy": 0.18364199995994568,
|
|
"num_tokens": 5321785.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 5.583510684967041,
|
|
"epoch": 2.5678745267712277,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00046982711496420057,
|
|
"loss": 5.4514,
|
|
"mean_token_accuracy": 0.18158535659313202,
|
|
"num_tokens": 5333557.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 5.645202255249023,
|
|
"epoch": 2.57328285559762,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00046961220723584925,
|
|
"loss": 5.5054,
|
|
"mean_token_accuracy": 0.1754367843270302,
|
|
"num_tokens": 5344767.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 5.510797452926636,
|
|
"epoch": 2.578691184424013,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00046939659228628725,
|
|
"loss": 5.4208,
|
|
"mean_token_accuracy": 0.18470364212989807,
|
|
"num_tokens": 5356673.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 5.612638759613037,
|
|
"epoch": 2.584099513250406,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00046918027089905956,
|
|
"loss": 5.5314,
|
|
"mean_token_accuracy": 0.1752219930291176,
|
|
"num_tokens": 5367296.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 5.562789011001587,
|
|
"epoch": 2.5895078420767983,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00046896324386027873,
|
|
"loss": 5.4141,
|
|
"mean_token_accuracy": 0.1820371448993683,
|
|
"num_tokens": 5378081.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 5.68981442451477,
|
|
"epoch": 2.5949161709031907,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00046874551195862145,
|
|
"loss": 5.5106,
|
|
"mean_token_accuracy": 0.1793023467063904,
|
|
"num_tokens": 5390244.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 5.527757263183593,
|
|
"epoch": 2.6003244997295836,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000468527075985326,
|
|
"loss": 5.37,
|
|
"mean_token_accuracy": 0.18837369978427887,
|
|
"num_tokens": 5400741.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 5.661465072631836,
|
|
"epoch": 2.6057328285559764,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00046830793673418924,
|
|
"loss": 5.5568,
|
|
"mean_token_accuracy": 0.18320724964141846,
|
|
"num_tokens": 5411440.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 5.5598897457122805,
|
|
"epoch": 2.611141157382369,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00046808809500156355,
|
|
"loss": 5.4009,
|
|
"mean_token_accuracy": 0.18060970157384873,
|
|
"num_tokens": 5422472.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 5.511435890197754,
|
|
"epoch": 2.6165494862087613,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004678675515863545,
|
|
"loss": 5.4968,
|
|
"mean_token_accuracy": 0.18061671704053878,
|
|
"num_tokens": 5433958.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 5.607755517959594,
|
|
"epoch": 2.621957815035154,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004676463072900173,
|
|
"loss": 5.3889,
|
|
"mean_token_accuracy": 0.1951850712299347,
|
|
"num_tokens": 5444343.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 5.517431402206421,
|
|
"epoch": 2.627366143861547,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00046742436291655415,
|
|
"loss": 5.413,
|
|
"mean_token_accuracy": 0.1810038775205612,
|
|
"num_tokens": 5456260.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 5.637908601760865,
|
|
"epoch": 2.6327744726879394,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00046720171927251155,
|
|
"loss": 5.5089,
|
|
"mean_token_accuracy": 0.1854751944541931,
|
|
"num_tokens": 5468281.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 5.433866500854492,
|
|
"epoch": 2.638182801514332,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004669783771669769,
|
|
"loss": 5.3607,
|
|
"mean_token_accuracy": 0.19548681378364563,
|
|
"num_tokens": 5480068.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 5.497932767868042,
|
|
"epoch": 2.6435911303407247,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000466754337411576,
|
|
"loss": 5.3508,
|
|
"mean_token_accuracy": 0.1829709678888321,
|
|
"num_tokens": 5490932.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 5.507752561569214,
|
|
"epoch": 2.6489994591671175,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00046652960082046985,
|
|
"loss": 5.4517,
|
|
"mean_token_accuracy": 0.18208895921707152,
|
|
"num_tokens": 5501787.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 5.601010990142822,
|
|
"epoch": 2.65440778799351,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00046630416821035164,
|
|
"loss": 5.4431,
|
|
"mean_token_accuracy": 0.18054919689893723,
|
|
"num_tokens": 5513297.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 5.488376712799072,
|
|
"epoch": 2.6598161168199024,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004660780404004441,
|
|
"loss": 5.4152,
|
|
"mean_token_accuracy": 0.18917850404977798,
|
|
"num_tokens": 5523797.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 5.522921752929688,
|
|
"epoch": 2.6652244456462952,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00046585121821249623,
|
|
"loss": 5.4439,
|
|
"mean_token_accuracy": 0.19177929908037186,
|
|
"num_tokens": 5535808.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 5.616627883911133,
|
|
"epoch": 2.670632774472688,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00046562370247078025,
|
|
"loss": 5.5415,
|
|
"mean_token_accuracy": 0.1763715013861656,
|
|
"num_tokens": 5548179.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 5.629001092910767,
|
|
"epoch": 2.6760411032990805,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000465395494002089,
|
|
"loss": 5.5239,
|
|
"mean_token_accuracy": 0.18134028017520903,
|
|
"num_tokens": 5559144.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 5.615467834472656,
|
|
"epoch": 2.6814494321254734,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004651665936357325,
|
|
"loss": 5.4382,
|
|
"mean_token_accuracy": 0.18410995602607727,
|
|
"num_tokens": 5571103.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 5.502652788162232,
|
|
"epoch": 2.686857760951866,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004649370022035354,
|
|
"loss": 5.3663,
|
|
"mean_token_accuracy": 0.18595369011163712,
|
|
"num_tokens": 5583239.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 5.501250267028809,
|
|
"epoch": 2.6922660897782587,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004647067205398332,
|
|
"loss": 5.3551,
|
|
"mean_token_accuracy": 0.18730516284704207,
|
|
"num_tokens": 5593602.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 5.608285856246948,
|
|
"epoch": 2.697674418604651,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00046447574948147027,
|
|
"loss": 5.4929,
|
|
"mean_token_accuracy": 0.17777594923973083,
|
|
"num_tokens": 5604696.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 5.5769391536712645,
|
|
"epoch": 2.703082747431044,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00046424408986779587,
|
|
"loss": 5.4734,
|
|
"mean_token_accuracy": 0.18040006458759308,
|
|
"num_tokens": 5615668.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"epoch": 2.703082747431044,
|
|
"eval_entropy": 5.529878706281836,
|
|
"eval_loss": 5.96121072769165,
|
|
"eval_mean_token_accuracy": 0.16987773351032626,
|
|
"eval_num_tokens": 5615668.0,
|
|
"eval_runtime": 2.52,
|
|
"eval_samples_per_second": 1394.824,
|
|
"eval_steps_per_second": 174.601,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 5.541957092285156,
|
|
"epoch": 2.7084910762574363,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004640117425406615,
|
|
"loss": 5.4023,
|
|
"mean_token_accuracy": 0.18345130383968353,
|
|
"num_tokens": 5627344.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 5.566352319717407,
|
|
"epoch": 2.713899405083829,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000463778708344418,
|
|
"loss": 5.4472,
|
|
"mean_token_accuracy": 0.17923472672700883,
|
|
"num_tokens": 5637472.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 5.530811166763305,
|
|
"epoch": 2.7193077339102216,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00046354498812591207,
|
|
"loss": 5.4369,
|
|
"mean_token_accuracy": 0.1827571853995323,
|
|
"num_tokens": 5647766.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 5.498395013809204,
|
|
"epoch": 2.7247160627366145,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00046331058273448364,
|
|
"loss": 5.36,
|
|
"mean_token_accuracy": 0.182998126745224,
|
|
"num_tokens": 5658343.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 5.513824892044068,
|
|
"epoch": 2.730124391563007,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004630754930219624,
|
|
"loss": 5.3589,
|
|
"mean_token_accuracy": 0.18447651267051696,
|
|
"num_tokens": 5669640.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 5.406114482879639,
|
|
"epoch": 2.7355327203893998,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004628397198426648,
|
|
"loss": 5.3526,
|
|
"mean_token_accuracy": 0.1800822839140892,
|
|
"num_tokens": 5682421.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 5.53621916770935,
|
|
"epoch": 2.740941049215792,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00046260326405339126,
|
|
"loss": 5.3213,
|
|
"mean_token_accuracy": 0.19262181520462035,
|
|
"num_tokens": 5692797.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 5.604675006866455,
|
|
"epoch": 2.746349378042185,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004623661265134226,
|
|
"loss": 5.5046,
|
|
"mean_token_accuracy": 0.1872665211558342,
|
|
"num_tokens": 5702779.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 5.611404800415039,
|
|
"epoch": 2.7517577068685775,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004621283080845172,
|
|
"loss": 5.4756,
|
|
"mean_token_accuracy": 0.1826336145401001,
|
|
"num_tokens": 5714147.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 5.488376426696777,
|
|
"epoch": 2.7571660356949703,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00046188980963090783,
|
|
"loss": 5.3228,
|
|
"mean_token_accuracy": 0.18872454464435579,
|
|
"num_tokens": 5725718.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 5.491141700744629,
|
|
"epoch": 2.7625743645213627,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00046165063201929825,
|
|
"loss": 5.3926,
|
|
"mean_token_accuracy": 0.19102770388126372,
|
|
"num_tokens": 5737441.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 5.5697265625,
|
|
"epoch": 2.7679826933477556,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00046141077611886074,
|
|
"loss": 5.4979,
|
|
"mean_token_accuracy": 0.18586716204881668,
|
|
"num_tokens": 5748735.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 5.669818305969239,
|
|
"epoch": 2.773391022174148,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004611702428012319,
|
|
"loss": 5.5654,
|
|
"mean_token_accuracy": 0.17501835227012635,
|
|
"num_tokens": 5760262.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 5.490745782852173,
|
|
"epoch": 2.778799351000541,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00046092903294051057,
|
|
"loss": 5.4752,
|
|
"mean_token_accuracy": 0.18279167711734773,
|
|
"num_tokens": 5771909.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 5.48658595085144,
|
|
"epoch": 2.7842076798269337,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00046068714741325385,
|
|
"loss": 5.3614,
|
|
"mean_token_accuracy": 0.1850089818239212,
|
|
"num_tokens": 5782028.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 5.6564967155456545,
|
|
"epoch": 2.789616008653326,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004604445870984743,
|
|
"loss": 5.4991,
|
|
"mean_token_accuracy": 0.1830728679895401,
|
|
"num_tokens": 5793641.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 5.595106601715088,
|
|
"epoch": 2.7950243374797186,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004602013528776367,
|
|
"loss": 5.446,
|
|
"mean_token_accuracy": 0.1821898952126503,
|
|
"num_tokens": 5805392.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 5.542260694503784,
|
|
"epoch": 2.8004326663061114,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004599574456346548,
|
|
"loss": 5.4431,
|
|
"mean_token_accuracy": 0.18215932548046113,
|
|
"num_tokens": 5816362.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 5.520218229293823,
|
|
"epoch": 2.8058409951325043,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00045971286625588796,
|
|
"loss": 5.3549,
|
|
"mean_token_accuracy": 0.18514317125082017,
|
|
"num_tokens": 5827518.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 5.594694185256958,
|
|
"epoch": 2.8112493239588967,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004594676156301383,
|
|
"loss": 5.5025,
|
|
"mean_token_accuracy": 0.1832107812166214,
|
|
"num_tokens": 5838455.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 5.521741724014282,
|
|
"epoch": 2.816657652785289,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00045922169464864725,
|
|
"loss": 5.366,
|
|
"mean_token_accuracy": 0.19088837504386902,
|
|
"num_tokens": 5850065.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 5.605880641937256,
|
|
"epoch": 2.822065981611682,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000458975104205092,
|
|
"loss": 5.4979,
|
|
"mean_token_accuracy": 0.1773496687412262,
|
|
"num_tokens": 5860444.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 5.611148071289063,
|
|
"epoch": 2.827474310438075,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00045872784519558304,
|
|
"loss": 5.4611,
|
|
"mean_token_accuracy": 0.18446561694145203,
|
|
"num_tokens": 5873984.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 5.568190860748291,
|
|
"epoch": 2.8328826392644673,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004584799185186599,
|
|
"loss": 5.5027,
|
|
"mean_token_accuracy": 0.1890917494893074,
|
|
"num_tokens": 5885807.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 5.5012836933135985,
|
|
"epoch": 2.8382909680908597,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00045823132507528903,
|
|
"loss": 5.379,
|
|
"mean_token_accuracy": 0.1860368937253952,
|
|
"num_tokens": 5897122.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 5.483487939834594,
|
|
"epoch": 2.8436992969172525,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004579820657688593,
|
|
"loss": 5.3791,
|
|
"mean_token_accuracy": 0.1876796692609787,
|
|
"num_tokens": 5909189.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 5.578971242904663,
|
|
"epoch": 2.8491076257436454,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004577321415051798,
|
|
"loss": 5.4752,
|
|
"mean_token_accuracy": 0.17739072442054749,
|
|
"num_tokens": 5920135.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 5.603898239135742,
|
|
"epoch": 2.854515954570038,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00045748155319247564,
|
|
"loss": 5.4429,
|
|
"mean_token_accuracy": 0.1896052911877632,
|
|
"num_tokens": 5932279.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 5.488618469238281,
|
|
"epoch": 2.8599242833964302,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004572303017413856,
|
|
"loss": 5.3953,
|
|
"mean_token_accuracy": 0.1838653177022934,
|
|
"num_tokens": 5942082.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 5.61859245300293,
|
|
"epoch": 2.865332612222823,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004569783880649578,
|
|
"loss": 5.4225,
|
|
"mean_token_accuracy": 0.18520787805318834,
|
|
"num_tokens": 5952856.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 5.395877313613892,
|
|
"epoch": 2.870740941049216,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004567258130786474,
|
|
"loss": 5.3358,
|
|
"mean_token_accuracy": 0.19270480573177337,
|
|
"num_tokens": 5963937.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 5.5658488273620605,
|
|
"epoch": 2.8761492698756084,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00045647257770031215,
|
|
"loss": 5.4313,
|
|
"mean_token_accuracy": 0.1854991927742958,
|
|
"num_tokens": 5974852.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 5.550371170043945,
|
|
"epoch": 2.8815575987020012,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00045621868285021023,
|
|
"loss": 5.4256,
|
|
"mean_token_accuracy": 0.18868498206138612,
|
|
"num_tokens": 5985616.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 5.500873804092407,
|
|
"epoch": 2.8869659275283936,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00045596412945099615,
|
|
"loss": 5.3728,
|
|
"mean_token_accuracy": 0.1907040685415268,
|
|
"num_tokens": 5997199.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 5.544794988632202,
|
|
"epoch": 2.8923742563547865,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00045570891842771727,
|
|
"loss": 5.4422,
|
|
"mean_token_accuracy": 0.18721016645431518,
|
|
"num_tokens": 6008654.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 5.561355257034302,
|
|
"epoch": 2.897782585181179,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00045545305070781135,
|
|
"loss": 5.4919,
|
|
"mean_token_accuracy": 0.1788644313812256,
|
|
"num_tokens": 6020662.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 5.5597954273223875,
|
|
"epoch": 2.903190914007572,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00045519652722110206,
|
|
"loss": 5.3624,
|
|
"mean_token_accuracy": 0.20726934373378753,
|
|
"num_tokens": 6031793.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 5.436601400375366,
|
|
"epoch": 2.908599242833964,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00045493934889979655,
|
|
"loss": 5.3595,
|
|
"mean_token_accuracy": 0.1865930899977684,
|
|
"num_tokens": 6042181.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 5.44061861038208,
|
|
"epoch": 2.914007571660357,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00045468151667848116,
|
|
"loss": 5.3299,
|
|
"mean_token_accuracy": 0.1910756140947342,
|
|
"num_tokens": 6052774.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 5.5122593402862545,
|
|
"epoch": 2.9194159004867495,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000454423031494119,
|
|
"loss": 5.3054,
|
|
"mean_token_accuracy": 0.19900496453046798,
|
|
"num_tokens": 6063273.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 5.422540950775146,
|
|
"epoch": 2.9248242293131423,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004541638942860458,
|
|
"loss": 5.3553,
|
|
"mean_token_accuracy": 0.19035297781229019,
|
|
"num_tokens": 6075121.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 5.503797769546509,
|
|
"epoch": 2.9302325581395348,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004539041059959667,
|
|
"loss": 5.3104,
|
|
"mean_token_accuracy": 0.18728087246418,
|
|
"num_tokens": 6085666.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 5.454809856414795,
|
|
"epoch": 2.9356408869659276,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00045364366756795295,
|
|
"loss": 5.346,
|
|
"mean_token_accuracy": 0.19704123735427856,
|
|
"num_tokens": 6096619.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 5.537594604492187,
|
|
"epoch": 2.94104921579232,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00045338257994843845,
|
|
"loss": 5.482,
|
|
"mean_token_accuracy": 0.18109430372714996,
|
|
"num_tokens": 6108800.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 5.527139663696289,
|
|
"epoch": 2.946457544618713,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004531208440862162,
|
|
"loss": 5.4037,
|
|
"mean_token_accuracy": 0.18500381112098693,
|
|
"num_tokens": 6119789.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 5.539804124832154,
|
|
"epoch": 2.9518658734451053,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00045285846093243485,
|
|
"loss": 5.404,
|
|
"mean_token_accuracy": 0.18450237959623336,
|
|
"num_tokens": 6131786.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 5.624119758605957,
|
|
"epoch": 2.957274202271498,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00045259543144059557,
|
|
"loss": 5.4827,
|
|
"mean_token_accuracy": 0.18646804094314576,
|
|
"num_tokens": 6143012.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 5.489918565750122,
|
|
"epoch": 2.9626825310978906,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004523317565665479,
|
|
"loss": 5.3703,
|
|
"mean_token_accuracy": 0.1911993756890297,
|
|
"num_tokens": 6153972.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 5.5649641990661625,
|
|
"epoch": 2.9680908599242835,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00045206743726848716,
|
|
"loss": 5.5396,
|
|
"mean_token_accuracy": 0.1772095590829849,
|
|
"num_tokens": 6166202.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 5.476376914978028,
|
|
"epoch": 2.973499188750676,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004518024745069502,
|
|
"loss": 5.2628,
|
|
"mean_token_accuracy": 0.19760213047266006,
|
|
"num_tokens": 6177253.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 5.503653049468994,
|
|
"epoch": 2.9789075175770687,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004515368692448123,
|
|
"loss": 5.5533,
|
|
"mean_token_accuracy": 0.1737920507788658,
|
|
"num_tokens": 6188857.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 5.643233919143677,
|
|
"epoch": 2.9843158464034616,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00045127062244728353,
|
|
"loss": 5.4861,
|
|
"mean_token_accuracy": 0.17456735521554947,
|
|
"num_tokens": 6199945.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 5.500112676620484,
|
|
"epoch": 2.989724175229854,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004510037350819055,
|
|
"loss": 5.4789,
|
|
"mean_token_accuracy": 0.1857718214392662,
|
|
"num_tokens": 6209866.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 5.582852029800415,
|
|
"epoch": 2.9951325040562464,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00045073620811854744,
|
|
"loss": 5.4448,
|
|
"mean_token_accuracy": 0.17825615555047988,
|
|
"num_tokens": 6221636.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 5.4927910168965655,
|
|
"epoch": 3.0,
|
|
"grad_norm": 2.34375,
|
|
"learning_rate": 0.00045046804252940294,
|
|
"loss": 5.3321,
|
|
"mean_token_accuracy": 0.18318185872501797,
|
|
"num_tokens": 6230667.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 5.509675455093384,
|
|
"epoch": 3.005408328826393,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00045019923928898644,
|
|
"loss": 5.0635,
|
|
"mean_token_accuracy": 0.2097476527094841,
|
|
"num_tokens": 6242242.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 5.556061553955078,
|
|
"epoch": 3.0108166576527853,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004499297993741294,
|
|
"loss": 5.0971,
|
|
"mean_token_accuracy": 0.20223594307899476,
|
|
"num_tokens": 6253337.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 5.390759706497192,
|
|
"epoch": 3.016224986479178,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004496597237639773,
|
|
"loss": 4.9769,
|
|
"mean_token_accuracy": 0.20241611301898957,
|
|
"num_tokens": 6264582.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 5.447097969055176,
|
|
"epoch": 3.0216333153055706,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00044938901343998536,
|
|
"loss": 5.0228,
|
|
"mean_token_accuracy": 0.21276663541793822,
|
|
"num_tokens": 6276172.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 5.399270057678223,
|
|
"epoch": 3.0270416441319634,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004491176693859156,
|
|
"loss": 4.9014,
|
|
"mean_token_accuracy": 0.21582320481538772,
|
|
"num_tokens": 6286038.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 5.408985042572022,
|
|
"epoch": 3.032449972958356,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004488456925878329,
|
|
"loss": 4.9935,
|
|
"mean_token_accuracy": 0.21080573201179503,
|
|
"num_tokens": 6296546.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 5.395119285583496,
|
|
"epoch": 3.0378583017847487,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00044857308403410163,
|
|
"loss": 5.082,
|
|
"mean_token_accuracy": 0.21180229634046555,
|
|
"num_tokens": 6308001.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 5.401966953277588,
|
|
"epoch": 3.043266630611141,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00044829984471538183,
|
|
"loss": 5.0738,
|
|
"mean_token_accuracy": 0.20532848238945006,
|
|
"num_tokens": 6320896.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 5.488651847839355,
|
|
"epoch": 3.048674959437534,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004480259756246258,
|
|
"loss": 5.1271,
|
|
"mean_token_accuracy": 0.19994610846042632,
|
|
"num_tokens": 6332710.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 5.4020051002502445,
|
|
"epoch": 3.0540832882639264,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00044775147775707465,
|
|
"loss": 5.0229,
|
|
"mean_token_accuracy": 0.20203343778848648,
|
|
"num_tokens": 6343563.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 5.404378986358642,
|
|
"epoch": 3.0594916170903192,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000447476352110254,
|
|
"loss": 5.1356,
|
|
"mean_token_accuracy": 0.20250921249389647,
|
|
"num_tokens": 6354064.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 5.408130216598511,
|
|
"epoch": 3.0648999459167117,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004472005996839712,
|
|
"loss": 5.0377,
|
|
"mean_token_accuracy": 0.2066985175013542,
|
|
"num_tokens": 6366551.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 5.376387310028076,
|
|
"epoch": 3.0703082747431045,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004469242214803112,
|
|
"loss": 4.9829,
|
|
"mean_token_accuracy": 0.20562020391225816,
|
|
"num_tokens": 6377086.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 5.399147939682007,
|
|
"epoch": 3.075716603569497,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004466472185036331,
|
|
"loss": 5.0797,
|
|
"mean_token_accuracy": 0.2034212440252304,
|
|
"num_tokens": 6388795.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 5.43955364227295,
|
|
"epoch": 3.08112493239589,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004463695917605663,
|
|
"loss": 5.0639,
|
|
"mean_token_accuracy": 0.20241507440805434,
|
|
"num_tokens": 6400174.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 5.362435960769654,
|
|
"epoch": 3.086533261222282,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000446091342260007,
|
|
"loss": 5.0126,
|
|
"mean_token_accuracy": 0.20121539682149886,
|
|
"num_tokens": 6411297.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 5.44430136680603,
|
|
"epoch": 3.091941590048675,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004458124710131145,
|
|
"loss": 5.0411,
|
|
"mean_token_accuracy": 0.20596790462732315,
|
|
"num_tokens": 6423487.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 5.4517333030700685,
|
|
"epoch": 3.0973499188750675,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00044553297903330766,
|
|
"loss": 5.2464,
|
|
"mean_token_accuracy": 0.19473037272691726,
|
|
"num_tokens": 6435460.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 5.54543399810791,
|
|
"epoch": 3.1027582477014604,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00044525286733626085,
|
|
"loss": 5.1708,
|
|
"mean_token_accuracy": 0.2071981370449066,
|
|
"num_tokens": 6445664.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 5.386275911331177,
|
|
"epoch": 3.1081665765278528,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004449721369399008,
|
|
"loss": 5.0424,
|
|
"mean_token_accuracy": 0.20263140499591828,
|
|
"num_tokens": 6457182.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 5.329753780364991,
|
|
"epoch": 3.1135749053542456,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00044469078886440227,
|
|
"loss": 4.9064,
|
|
"mean_token_accuracy": 0.21530135720968246,
|
|
"num_tokens": 6468378.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 5.342884540557861,
|
|
"epoch": 3.118983234180638,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00044440882413218484,
|
|
"loss": 5.1204,
|
|
"mean_token_accuracy": 0.19751410633325578,
|
|
"num_tokens": 6479625.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 5.365936040878296,
|
|
"epoch": 3.124391563007031,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00044412624376790905,
|
|
"loss": 5.0568,
|
|
"mean_token_accuracy": 0.20249563604593276,
|
|
"num_tokens": 6490713.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 5.358120775222778,
|
|
"epoch": 3.1297998918334233,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004438430487984726,
|
|
"loss": 5.0944,
|
|
"mean_token_accuracy": 0.1902830719947815,
|
|
"num_tokens": 6502161.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 5.426725244522094,
|
|
"epoch": 3.135208220659816,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004435592402530066,
|
|
"loss": 4.9838,
|
|
"mean_token_accuracy": 0.2060424879193306,
|
|
"num_tokens": 6513639.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 5.312023496627807,
|
|
"epoch": 3.1406165494862086,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000443274819162872,
|
|
"loss": 5.0268,
|
|
"mean_token_accuracy": 0.20649791955947877,
|
|
"num_tokens": 6524893.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 5.332180023193359,
|
|
"epoch": 3.1460248783126015,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004429897865616557,
|
|
"loss": 5.0278,
|
|
"mean_token_accuracy": 0.20728944092988968,
|
|
"num_tokens": 6535797.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 5.4129204750061035,
|
|
"epoch": 3.151433207138994,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004427041434851668,
|
|
"loss": 5.1186,
|
|
"mean_token_accuracy": 0.19582765996456147,
|
|
"num_tokens": 6547117.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 5.424050331115723,
|
|
"epoch": 3.1568415359653867,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00044241789097143293,
|
|
"loss": 5.135,
|
|
"mean_token_accuracy": 0.20738081634044647,
|
|
"num_tokens": 6558771.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 5.3796463966369625,
|
|
"epoch": 3.162249864791779,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00044213103006069635,
|
|
"loss": 5.102,
|
|
"mean_token_accuracy": 0.19800227135419846,
|
|
"num_tokens": 6569922.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 5.435439729690552,
|
|
"epoch": 3.167658193618172,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00044184356179541035,
|
|
"loss": 5.1064,
|
|
"mean_token_accuracy": 0.20139265060424805,
|
|
"num_tokens": 6580561.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 5.293603086471558,
|
|
"epoch": 3.1730665224445644,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004415554872202352,
|
|
"loss": 4.942,
|
|
"mean_token_accuracy": 0.20813218504190445,
|
|
"num_tokens": 6591636.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 5.433289432525635,
|
|
"epoch": 3.1784748512709573,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00044126680738203466,
|
|
"loss": 5.1708,
|
|
"mean_token_accuracy": 0.19901786148548126,
|
|
"num_tokens": 6603736.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 5.463957738876343,
|
|
"epoch": 3.1838831800973497,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004409775233298718,
|
|
"loss": 5.0994,
|
|
"mean_token_accuracy": 0.2040825754404068,
|
|
"num_tokens": 6615415.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 5.415656089782715,
|
|
"epoch": 3.1892915089237426,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004406876361150056,
|
|
"loss": 5.14,
|
|
"mean_token_accuracy": 0.19950893074274062,
|
|
"num_tokens": 6625970.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 5.386278438568115,
|
|
"epoch": 3.194699837750135,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004403971467908869,
|
|
"loss": 5.0804,
|
|
"mean_token_accuracy": 0.20067908614873886,
|
|
"num_tokens": 6636326.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 5.374203872680664,
|
|
"epoch": 3.200108166576528,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004401060564131545,
|
|
"loss": 5.1502,
|
|
"mean_token_accuracy": 0.20141685009002686,
|
|
"num_tokens": 6646825.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 5.4502369403839115,
|
|
"epoch": 3.2055164954029207,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004398143660396315,
|
|
"loss": 5.1232,
|
|
"mean_token_accuracy": 0.2037063106894493,
|
|
"num_tokens": 6657748.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 5.297688674926758,
|
|
"epoch": 3.210924824229313,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004395220767303215,
|
|
"loss": 5.006,
|
|
"mean_token_accuracy": 0.21185962557792665,
|
|
"num_tokens": 6668422.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 5.4858331203460695,
|
|
"epoch": 3.2163331530557056,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004392291895474045,
|
|
"loss": 5.1036,
|
|
"mean_token_accuracy": 0.19856813549995422,
|
|
"num_tokens": 6678898.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 5.3894514560699465,
|
|
"epoch": 3.2217414818820984,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004389357055552331,
|
|
"loss": 5.1379,
|
|
"mean_token_accuracy": 0.20470501333475113,
|
|
"num_tokens": 6688691.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 5.338142776489258,
|
|
"epoch": 3.2271498107084913,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00043864162582032897,
|
|
"loss": 5.0264,
|
|
"mean_token_accuracy": 0.20787582248449327,
|
|
"num_tokens": 6700279.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 5.382350444793701,
|
|
"epoch": 3.2325581395348837,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004383469514113784,
|
|
"loss": 4.9765,
|
|
"mean_token_accuracy": 0.20846454799175262,
|
|
"num_tokens": 6710440.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 5.292540121078491,
|
|
"epoch": 3.2379664683612766,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004380516833992289,
|
|
"loss": 5.1315,
|
|
"mean_token_accuracy": 0.19864480793476105,
|
|
"num_tokens": 6722347.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 5.446057081222534,
|
|
"epoch": 3.243374797187669,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000437755822856885,
|
|
"loss": 5.1435,
|
|
"mean_token_accuracy": 0.20252999067306518,
|
|
"num_tokens": 6732936.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 3.243374797187669,
|
|
"eval_entropy": 5.332620696587996,
|
|
"eval_loss": 5.948505878448486,
|
|
"eval_mean_token_accuracy": 0.17215410125526515,
|
|
"eval_num_tokens": 6732936.0,
|
|
"eval_runtime": 2.7195,
|
|
"eval_samples_per_second": 1292.504,
|
|
"eval_steps_per_second": 161.793,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"entropy": 5.404446983337403,
|
|
"epoch": 3.248783126014062,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004374593708595049,
|
|
"loss": 5.1794,
|
|
"mean_token_accuracy": 0.20141776353120805,
|
|
"num_tokens": 6744552.0,
|
|
"step": 3005
|
|
},
|
|
{
|
|
"entropy": 5.437932968139648,
|
|
"epoch": 3.2541914548404542,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00043716232848439555,
|
|
"loss": 5.1234,
|
|
"mean_token_accuracy": 0.2003450572490692,
|
|
"num_tokens": 6755749.0,
|
|
"step": 3010
|
|
},
|
|
{
|
|
"entropy": 5.304684257507324,
|
|
"epoch": 3.259599783666847,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00043686469681100985,
|
|
"loss": 5.033,
|
|
"mean_token_accuracy": 0.21049189567565918,
|
|
"num_tokens": 6766119.0,
|
|
"step": 3015
|
|
},
|
|
{
|
|
"entropy": 5.330397033691407,
|
|
"epoch": 3.2650081124932395,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00043656647692094186,
|
|
"loss": 5.0929,
|
|
"mean_token_accuracy": 0.19737981408834457,
|
|
"num_tokens": 6776491.0,
|
|
"step": 3020
|
|
},
|
|
{
|
|
"entropy": 5.488622808456421,
|
|
"epoch": 3.2704164413196324,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004362676698979233,
|
|
"loss": 5.1857,
|
|
"mean_token_accuracy": 0.19461405873298646,
|
|
"num_tokens": 6788795.0,
|
|
"step": 3025
|
|
},
|
|
{
|
|
"entropy": 5.452446317672729,
|
|
"epoch": 3.275824770146025,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00043596827682781974,
|
|
"loss": 5.1871,
|
|
"mean_token_accuracy": 0.19906928539276122,
|
|
"num_tokens": 6799706.0,
|
|
"step": 3030
|
|
},
|
|
{
|
|
"entropy": 5.405957126617432,
|
|
"epoch": 3.2812330989724177,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004356682987986262,
|
|
"loss": 5.1164,
|
|
"mean_token_accuracy": 0.19947726726531984,
|
|
"num_tokens": 6812454.0,
|
|
"step": 3035
|
|
},
|
|
{
|
|
"entropy": 5.371125650405884,
|
|
"epoch": 3.28664142779881,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004353677369004635,
|
|
"loss": 5.0523,
|
|
"mean_token_accuracy": 0.20495113730430603,
|
|
"num_tokens": 6823950.0,
|
|
"step": 3040
|
|
},
|
|
{
|
|
"entropy": 5.423996782302856,
|
|
"epoch": 3.292049756625203,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00043506659222557426,
|
|
"loss": 5.2571,
|
|
"mean_token_accuracy": 0.18792027980089188,
|
|
"num_tokens": 6835028.0,
|
|
"step": 3045
|
|
},
|
|
{
|
|
"entropy": 5.38666787147522,
|
|
"epoch": 3.2974580854515954,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000434764865868319,
|
|
"loss": 5.1416,
|
|
"mean_token_accuracy": 0.20227540284395218,
|
|
"num_tokens": 6846316.0,
|
|
"step": 3050
|
|
},
|
|
{
|
|
"entropy": 5.362994909286499,
|
|
"epoch": 3.302866414277988,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000434462558925172,
|
|
"loss": 5.0634,
|
|
"mean_token_accuracy": 0.20971330106258393,
|
|
"num_tokens": 6856463.0,
|
|
"step": 3055
|
|
},
|
|
{
|
|
"entropy": 5.367487382888794,
|
|
"epoch": 3.3082747431043806,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00043415967249471734,
|
|
"loss": 5.0826,
|
|
"mean_token_accuracy": 0.20438590347766877,
|
|
"num_tokens": 6867869.0,
|
|
"step": 3060
|
|
},
|
|
{
|
|
"entropy": 5.360267257690429,
|
|
"epoch": 3.3136830719307735,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000433856207677645,
|
|
"loss": 5.0679,
|
|
"mean_token_accuracy": 0.20497159659862518,
|
|
"num_tokens": 6878545.0,
|
|
"step": 3065
|
|
},
|
|
{
|
|
"entropy": 5.340911388397217,
|
|
"epoch": 3.319091400757166,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000433552165576747,
|
|
"loss": 5.1287,
|
|
"mean_token_accuracy": 0.19842114448547363,
|
|
"num_tokens": 6889399.0,
|
|
"step": 3070
|
|
},
|
|
{
|
|
"entropy": 5.464095878601074,
|
|
"epoch": 3.3244997295835588,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00043324754729691275,
|
|
"loss": 5.1122,
|
|
"mean_token_accuracy": 0.21228110790252686,
|
|
"num_tokens": 6900428.0,
|
|
"step": 3075
|
|
},
|
|
{
|
|
"entropy": 5.447661352157593,
|
|
"epoch": 3.329908058409951,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00043294235394512594,
|
|
"loss": 5.2305,
|
|
"mean_token_accuracy": 0.1921966329216957,
|
|
"num_tokens": 6911112.0,
|
|
"step": 3080
|
|
},
|
|
{
|
|
"entropy": 5.341199159622192,
|
|
"epoch": 3.335316387236344,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004326365866304599,
|
|
"loss": 4.9905,
|
|
"mean_token_accuracy": 0.21413780748844147,
|
|
"num_tokens": 6922383.0,
|
|
"step": 3085
|
|
},
|
|
{
|
|
"entropy": 5.327231740951538,
|
|
"epoch": 3.3407247160627365,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004323302464640738,
|
|
"loss": 5.1709,
|
|
"mean_token_accuracy": 0.19588454961776733,
|
|
"num_tokens": 6935153.0,
|
|
"step": 3090
|
|
},
|
|
{
|
|
"entropy": 5.502281904220581,
|
|
"epoch": 3.3461330448891293,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00043202333455920843,
|
|
"loss": 5.2227,
|
|
"mean_token_accuracy": 0.20031266808509826,
|
|
"num_tokens": 6946814.0,
|
|
"step": 3095
|
|
},
|
|
{
|
|
"entropy": 5.34875636100769,
|
|
"epoch": 3.3515413737155217,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004317158520311824,
|
|
"loss": 5.0626,
|
|
"mean_token_accuracy": 0.20735397785902024,
|
|
"num_tokens": 6957230.0,
|
|
"step": 3100
|
|
},
|
|
{
|
|
"entropy": 5.260023832321167,
|
|
"epoch": 3.3569497025419146,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004314077999973879,
|
|
"loss": 5.0882,
|
|
"mean_token_accuracy": 0.19161992520093918,
|
|
"num_tokens": 6968060.0,
|
|
"step": 3105
|
|
},
|
|
{
|
|
"entropy": 5.357646131515503,
|
|
"epoch": 3.362358031368307,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004310991795772869,
|
|
"loss": 4.9857,
|
|
"mean_token_accuracy": 0.20824098736047744,
|
|
"num_tokens": 6977782.0,
|
|
"step": 3110
|
|
},
|
|
{
|
|
"entropy": 5.402149105072022,
|
|
"epoch": 3.3677663601947,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004307899918924065,
|
|
"loss": 5.0892,
|
|
"mean_token_accuracy": 0.1999707654118538,
|
|
"num_tokens": 6988252.0,
|
|
"step": 3115
|
|
},
|
|
{
|
|
"entropy": 5.426277208328247,
|
|
"epoch": 3.3731746890210923,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004304802380663356,
|
|
"loss": 5.0785,
|
|
"mean_token_accuracy": 0.19842932671308516,
|
|
"num_tokens": 6998523.0,
|
|
"step": 3120
|
|
},
|
|
{
|
|
"entropy": 5.365118360519409,
|
|
"epoch": 3.378583017847485,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00043016991922472015,
|
|
"loss": 5.1201,
|
|
"mean_token_accuracy": 0.2019295185804367,
|
|
"num_tokens": 7010058.0,
|
|
"step": 3125
|
|
},
|
|
{
|
|
"entropy": 5.4208269119262695,
|
|
"epoch": 3.383991346673878,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00042985903649525977,
|
|
"loss": 5.2317,
|
|
"mean_token_accuracy": 0.19186155647039413,
|
|
"num_tokens": 7021289.0,
|
|
"step": 3130
|
|
},
|
|
{
|
|
"entropy": 5.528253030776978,
|
|
"epoch": 3.3893996755002704,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00042954759100770275,
|
|
"loss": 5.2437,
|
|
"mean_token_accuracy": 0.19011763632297515,
|
|
"num_tokens": 7032649.0,
|
|
"step": 3135
|
|
},
|
|
{
|
|
"entropy": 5.41603307723999,
|
|
"epoch": 3.394808004326663,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00042923558389384303,
|
|
"loss": 5.0942,
|
|
"mean_token_accuracy": 0.2014612853527069,
|
|
"num_tokens": 7044120.0,
|
|
"step": 3140
|
|
},
|
|
{
|
|
"entropy": 5.365779972076416,
|
|
"epoch": 3.4002163331530557,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004289230162875149,
|
|
"loss": 5.1202,
|
|
"mean_token_accuracy": 0.20391982346773146,
|
|
"num_tokens": 7054526.0,
|
|
"step": 3145
|
|
},
|
|
{
|
|
"entropy": 5.313352394104004,
|
|
"epoch": 3.4056246619794486,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00042860988932458987,
|
|
"loss": 5.0013,
|
|
"mean_token_accuracy": 0.20930091589689254,
|
|
"num_tokens": 7064744.0,
|
|
"step": 3150
|
|
},
|
|
{
|
|
"entropy": 5.349811458587647,
|
|
"epoch": 3.411032990805841,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004282962041429721,
|
|
"loss": 5.1146,
|
|
"mean_token_accuracy": 0.20003893375396728,
|
|
"num_tokens": 7076360.0,
|
|
"step": 3155
|
|
},
|
|
{
|
|
"entropy": 5.385465288162232,
|
|
"epoch": 3.4164413196322334,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00042798196188259427,
|
|
"loss": 5.129,
|
|
"mean_token_accuracy": 0.20669629722833632,
|
|
"num_tokens": 7088114.0,
|
|
"step": 3160
|
|
},
|
|
{
|
|
"entropy": 5.395055150985717,
|
|
"epoch": 3.4218496484586263,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004276671636854133,
|
|
"loss": 5.167,
|
|
"mean_token_accuracy": 0.19305334538221358,
|
|
"num_tokens": 7099125.0,
|
|
"step": 3165
|
|
},
|
|
{
|
|
"entropy": 5.384315872192383,
|
|
"epoch": 3.427257977285019,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00042735181069540677,
|
|
"loss": 5.0851,
|
|
"mean_token_accuracy": 0.20747402012348176,
|
|
"num_tokens": 7109875.0,
|
|
"step": 3170
|
|
},
|
|
{
|
|
"entropy": 5.399315214157104,
|
|
"epoch": 3.4326663061114115,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004270359040585681,
|
|
"loss": 5.1363,
|
|
"mean_token_accuracy": 0.19924622923135757,
|
|
"num_tokens": 7121388.0,
|
|
"step": 3175
|
|
},
|
|
{
|
|
"entropy": 5.364181566238403,
|
|
"epoch": 3.4380746349378044,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00042671944492290277,
|
|
"loss": 5.1741,
|
|
"mean_token_accuracy": 0.1955175593495369,
|
|
"num_tokens": 7134161.0,
|
|
"step": 3180
|
|
},
|
|
{
|
|
"entropy": 5.419300699234009,
|
|
"epoch": 3.443482963764197,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004264024344384239,
|
|
"loss": 5.1653,
|
|
"mean_token_accuracy": 0.1977361559867859,
|
|
"num_tokens": 7146580.0,
|
|
"step": 3185
|
|
},
|
|
{
|
|
"entropy": 5.446963834762573,
|
|
"epoch": 3.4488912925905897,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00042608487375714847,
|
|
"loss": 5.1156,
|
|
"mean_token_accuracy": 0.19587754905223848,
|
|
"num_tokens": 7156969.0,
|
|
"step": 3190
|
|
},
|
|
{
|
|
"entropy": 5.2848235130310055,
|
|
"epoch": 3.454299621416982,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00042576676403309265,
|
|
"loss": 5.0507,
|
|
"mean_token_accuracy": 0.21107542514801025,
|
|
"num_tokens": 7168687.0,
|
|
"step": 3195
|
|
},
|
|
{
|
|
"entropy": 5.409307241439819,
|
|
"epoch": 3.459707950243375,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004254481064222679,
|
|
"loss": 5.142,
|
|
"mean_token_accuracy": 0.20575996190309526,
|
|
"num_tokens": 7179996.0,
|
|
"step": 3200
|
|
},
|
|
{
|
|
"entropy": 5.430664873123169,
|
|
"epoch": 3.4651162790697674,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00042512890208267677,
|
|
"loss": 5.1729,
|
|
"mean_token_accuracy": 0.1971002474427223,
|
|
"num_tokens": 7191214.0,
|
|
"step": 3205
|
|
},
|
|
{
|
|
"entropy": 5.372012519836426,
|
|
"epoch": 3.4705246078961602,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00042480915217430853,
|
|
"loss": 5.1848,
|
|
"mean_token_accuracy": 0.19894933700561523,
|
|
"num_tokens": 7203491.0,
|
|
"step": 3210
|
|
},
|
|
{
|
|
"entropy": 5.425632619857788,
|
|
"epoch": 3.4759329367225527,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00042448885785913507,
|
|
"loss": 5.1205,
|
|
"mean_token_accuracy": 0.19456373900175095,
|
|
"num_tokens": 7215018.0,
|
|
"step": 3215
|
|
},
|
|
{
|
|
"entropy": 5.401223468780517,
|
|
"epoch": 3.4813412655489455,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00042416802030110677,
|
|
"loss": 5.0558,
|
|
"mean_token_accuracy": 0.21026203781366348,
|
|
"num_tokens": 7226209.0,
|
|
"step": 3220
|
|
},
|
|
{
|
|
"entropy": 5.353937673568725,
|
|
"epoch": 3.486749594375338,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00042384664066614787,
|
|
"loss": 5.0836,
|
|
"mean_token_accuracy": 0.19995051324367524,
|
|
"num_tokens": 7236505.0,
|
|
"step": 3225
|
|
},
|
|
{
|
|
"entropy": 5.285505199432373,
|
|
"epoch": 3.492157923201731,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004235247201221528,
|
|
"loss": 4.9954,
|
|
"mean_token_accuracy": 0.20573049038648605,
|
|
"num_tokens": 7247414.0,
|
|
"step": 3230
|
|
},
|
|
{
|
|
"entropy": 5.361935043334961,
|
|
"epoch": 3.497566252028123,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004232022598389816,
|
|
"loss": 5.1845,
|
|
"mean_token_accuracy": 0.19810942113399505,
|
|
"num_tokens": 7258666.0,
|
|
"step": 3235
|
|
},
|
|
{
|
|
"entropy": 5.403533029556274,
|
|
"epoch": 3.502974580854516,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004228792609884557,
|
|
"loss": 5.092,
|
|
"mean_token_accuracy": 0.2030029594898224,
|
|
"num_tokens": 7270201.0,
|
|
"step": 3240
|
|
},
|
|
{
|
|
"entropy": 5.3888873100280765,
|
|
"epoch": 3.5083829096809085,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00042255572474435364,
|
|
"loss": 5.1816,
|
|
"mean_token_accuracy": 0.19738761931657792,
|
|
"num_tokens": 7282254.0,
|
|
"step": 3245
|
|
},
|
|
{
|
|
"entropy": 5.443478727340699,
|
|
"epoch": 3.5137912385073014,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004222316522824069,
|
|
"loss": 5.1906,
|
|
"mean_token_accuracy": 0.1947275459766388,
|
|
"num_tokens": 7294579.0,
|
|
"step": 3250
|
|
},
|
|
{
|
|
"entropy": 5.359054899215698,
|
|
"epoch": 3.5191995673336938,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004219070447802955,
|
|
"loss": 5.1613,
|
|
"mean_token_accuracy": 0.19853783547878265,
|
|
"num_tokens": 7305822.0,
|
|
"step": 3255
|
|
},
|
|
{
|
|
"entropy": 5.407787561416626,
|
|
"epoch": 3.5246078961600866,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00042158190341764384,
|
|
"loss": 5.1829,
|
|
"mean_token_accuracy": 0.20015691071748734,
|
|
"num_tokens": 7317543.0,
|
|
"step": 3260
|
|
},
|
|
{
|
|
"entropy": 5.403790664672852,
|
|
"epoch": 3.530016224986479,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00042125622937601645,
|
|
"loss": 5.125,
|
|
"mean_token_accuracy": 0.20045264214277267,
|
|
"num_tokens": 7327804.0,
|
|
"step": 3265
|
|
},
|
|
{
|
|
"entropy": 5.405599212646484,
|
|
"epoch": 3.535424553812872,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004209300238389135,
|
|
"loss": 5.0782,
|
|
"mean_token_accuracy": 0.20477028638124467,
|
|
"num_tokens": 7337045.0,
|
|
"step": 3270
|
|
},
|
|
{
|
|
"entropy": 5.335825634002686,
|
|
"epoch": 3.5408328826392643,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004206032879917667,
|
|
"loss": 5.1816,
|
|
"mean_token_accuracy": 0.20437230318784713,
|
|
"num_tokens": 7348184.0,
|
|
"step": 3275
|
|
},
|
|
{
|
|
"entropy": 5.362151193618774,
|
|
"epoch": 3.546241211465657,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004202760230219348,
|
|
"loss": 5.0544,
|
|
"mean_token_accuracy": 0.2070061221718788,
|
|
"num_tokens": 7359571.0,
|
|
"step": 3280
|
|
},
|
|
{
|
|
"entropy": 5.304046630859375,
|
|
"epoch": 3.5516495402920496,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00041994823011869955,
|
|
"loss": 5.0829,
|
|
"mean_token_accuracy": 0.20057242065668107,
|
|
"num_tokens": 7370429.0,
|
|
"step": 3285
|
|
},
|
|
{
|
|
"entropy": 5.385798215866089,
|
|
"epoch": 3.5570578691184425,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00041961991047326116,
|
|
"loss": 5.0513,
|
|
"mean_token_accuracy": 0.204226578772068,
|
|
"num_tokens": 7381569.0,
|
|
"step": 3290
|
|
},
|
|
{
|
|
"entropy": 5.347041654586792,
|
|
"epoch": 3.5624661979448353,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004192910652787338,
|
|
"loss": 5.0584,
|
|
"mean_token_accuracy": 0.20807305872440338,
|
|
"num_tokens": 7392746.0,
|
|
"step": 3295
|
|
},
|
|
{
|
|
"entropy": 5.335638523101807,
|
|
"epoch": 3.5678745267712277,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004189616957301419,
|
|
"loss": 5.1571,
|
|
"mean_token_accuracy": 0.18885393887758256,
|
|
"num_tokens": 7405520.0,
|
|
"step": 3300
|
|
},
|
|
{
|
|
"entropy": 5.328299713134766,
|
|
"epoch": 3.57328285559762,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004186318030244149,
|
|
"loss": 5.1186,
|
|
"mean_token_accuracy": 0.1970698893070221,
|
|
"num_tokens": 7417788.0,
|
|
"step": 3305
|
|
},
|
|
{
|
|
"entropy": 5.396932554244995,
|
|
"epoch": 3.578691184424013,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00041830138836038383,
|
|
"loss": 5.1949,
|
|
"mean_token_accuracy": 0.19803751409053802,
|
|
"num_tokens": 7428912.0,
|
|
"step": 3310
|
|
},
|
|
{
|
|
"entropy": 5.3418145179748535,
|
|
"epoch": 3.584099513250406,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004179704529387763,
|
|
"loss": 5.073,
|
|
"mean_token_accuracy": 0.2060249462723732,
|
|
"num_tokens": 7439797.0,
|
|
"step": 3315
|
|
},
|
|
{
|
|
"entropy": 5.3200156688690186,
|
|
"epoch": 3.5895078420767983,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004176389979622125,
|
|
"loss": 5.063,
|
|
"mean_token_accuracy": 0.20317167937755584,
|
|
"num_tokens": 7451755.0,
|
|
"step": 3320
|
|
},
|
|
{
|
|
"entropy": 5.34779052734375,
|
|
"epoch": 3.5949161709031907,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004173070246352003,
|
|
"loss": 5.1525,
|
|
"mean_token_accuracy": 0.20006269216537476,
|
|
"num_tokens": 7463393.0,
|
|
"step": 3325
|
|
},
|
|
{
|
|
"entropy": 5.312729597091675,
|
|
"epoch": 3.6003244997295836,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00041697453416413184,
|
|
"loss": 5.0841,
|
|
"mean_token_accuracy": 0.20348519086837769,
|
|
"num_tokens": 7474404.0,
|
|
"step": 3330
|
|
},
|
|
{
|
|
"entropy": 5.367253303527832,
|
|
"epoch": 3.6057328285559764,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000416641527757278,
|
|
"loss": 5.0964,
|
|
"mean_token_accuracy": 0.20138738304376602,
|
|
"num_tokens": 7485495.0,
|
|
"step": 3335
|
|
},
|
|
{
|
|
"entropy": 5.270753240585327,
|
|
"epoch": 3.611141157382369,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00041630800662478514,
|
|
"loss": 5.0484,
|
|
"mean_token_accuracy": 0.20604611337184905,
|
|
"num_tokens": 7497305.0,
|
|
"step": 3340
|
|
},
|
|
{
|
|
"entropy": 5.293817567825317,
|
|
"epoch": 3.6165494862087613,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00041597397197866957,
|
|
"loss": 4.9944,
|
|
"mean_token_accuracy": 0.21176810711622238,
|
|
"num_tokens": 7509008.0,
|
|
"step": 3345
|
|
},
|
|
{
|
|
"entropy": 5.230751609802246,
|
|
"epoch": 3.621957815035154,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004156394250328141,
|
|
"loss": 5.0718,
|
|
"mean_token_accuracy": 0.20938670486211777,
|
|
"num_tokens": 7520079.0,
|
|
"step": 3350
|
|
},
|
|
{
|
|
"entropy": 5.3698993682861325,
|
|
"epoch": 3.627366143861547,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000415304367002963,
|
|
"loss": 5.1089,
|
|
"mean_token_accuracy": 0.20625361800193787,
|
|
"num_tokens": 7531294.0,
|
|
"step": 3355
|
|
},
|
|
{
|
|
"entropy": 5.389321279525757,
|
|
"epoch": 3.6327744726879394,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00041496879910671796,
|
|
"loss": 5.1044,
|
|
"mean_token_accuracy": 0.20573567897081374,
|
|
"num_tokens": 7542508.0,
|
|
"step": 3360
|
|
},
|
|
{
|
|
"entropy": 5.306264925003052,
|
|
"epoch": 3.638182801514332,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004146327225635336,
|
|
"loss": 5.0493,
|
|
"mean_token_accuracy": 0.206992444396019,
|
|
"num_tokens": 7553288.0,
|
|
"step": 3365
|
|
},
|
|
{
|
|
"entropy": 5.346945905685425,
|
|
"epoch": 3.6435911303407247,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004142961385947127,
|
|
"loss": 5.258,
|
|
"mean_token_accuracy": 0.20198436081409454,
|
|
"num_tokens": 7564532.0,
|
|
"step": 3370
|
|
},
|
|
{
|
|
"entropy": 5.340420055389404,
|
|
"epoch": 3.6489994591671175,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00041395904842340215,
|
|
"loss": 5.0832,
|
|
"mean_token_accuracy": 0.20658657252788543,
|
|
"num_tokens": 7575109.0,
|
|
"step": 3375
|
|
},
|
|
{
|
|
"entropy": 5.280505895614624,
|
|
"epoch": 3.65440778799351,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004136214532745887,
|
|
"loss": 5.0686,
|
|
"mean_token_accuracy": 0.21024899035692216,
|
|
"num_tokens": 7585871.0,
|
|
"step": 3380
|
|
},
|
|
{
|
|
"entropy": 5.38696026802063,
|
|
"epoch": 3.6598161168199024,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004132833543750936,
|
|
"loss": 5.1336,
|
|
"mean_token_accuracy": 0.21070279330015182,
|
|
"num_tokens": 7596821.0,
|
|
"step": 3385
|
|
},
|
|
{
|
|
"entropy": 5.316754484176636,
|
|
"epoch": 3.6652244456462952,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00041294475295356917,
|
|
"loss": 5.0817,
|
|
"mean_token_accuracy": 0.20639661848545074,
|
|
"num_tokens": 7608101.0,
|
|
"step": 3390
|
|
},
|
|
{
|
|
"entropy": 5.305299234390259,
|
|
"epoch": 3.670632774472688,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004126056502404937,
|
|
"loss": 5.0862,
|
|
"mean_token_accuracy": 0.20040349215269088,
|
|
"num_tokens": 7618722.0,
|
|
"step": 3395
|
|
},
|
|
{
|
|
"entropy": 5.320506763458252,
|
|
"epoch": 3.6760411032990805,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004122660474681672,
|
|
"loss": 5.0853,
|
|
"mean_token_accuracy": 0.20719005316495895,
|
|
"num_tokens": 7630061.0,
|
|
"step": 3400
|
|
},
|
|
{
|
|
"entropy": 5.310720777511596,
|
|
"epoch": 3.6814494321254734,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00041192594587070706,
|
|
"loss": 5.076,
|
|
"mean_token_accuracy": 0.2020289495587349,
|
|
"num_tokens": 7641830.0,
|
|
"step": 3405
|
|
},
|
|
{
|
|
"entropy": 5.334624242782593,
|
|
"epoch": 3.686857760951866,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00041158534668404325,
|
|
"loss": 5.1235,
|
|
"mean_token_accuracy": 0.2014719843864441,
|
|
"num_tokens": 7652750.0,
|
|
"step": 3410
|
|
},
|
|
{
|
|
"entropy": 5.282159090042114,
|
|
"epoch": 3.6922660897782587,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00041124425114591386,
|
|
"loss": 5.0329,
|
|
"mean_token_accuracy": 0.20977533012628555,
|
|
"num_tokens": 7663405.0,
|
|
"step": 3415
|
|
},
|
|
{
|
|
"entropy": 5.259342622756958,
|
|
"epoch": 3.697674418604651,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00041090266049586107,
|
|
"loss": 5.0426,
|
|
"mean_token_accuracy": 0.19942890852689743,
|
|
"num_tokens": 7673958.0,
|
|
"step": 3420
|
|
},
|
|
{
|
|
"entropy": 5.352772521972656,
|
|
"epoch": 3.703082747431044,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004105605759752258,
|
|
"loss": 5.053,
|
|
"mean_token_accuracy": 0.202606862783432,
|
|
"num_tokens": 7685475.0,
|
|
"step": 3425
|
|
},
|
|
{
|
|
"entropy": 5.227557897567749,
|
|
"epoch": 3.7084910762574363,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00041021799882714436,
|
|
"loss": 5.0337,
|
|
"mean_token_accuracy": 0.20897262394428254,
|
|
"num_tokens": 7697296.0,
|
|
"step": 3430
|
|
},
|
|
{
|
|
"entropy": 5.354386854171753,
|
|
"epoch": 3.713899405083829,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004098749302965426,
|
|
"loss": 5.1808,
|
|
"mean_token_accuracy": 0.2017278030514717,
|
|
"num_tokens": 7709414.0,
|
|
"step": 3435
|
|
},
|
|
{
|
|
"entropy": 5.387508726119995,
|
|
"epoch": 3.7193077339102216,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004095313716301324,
|
|
"loss": 5.1719,
|
|
"mean_token_accuracy": 0.18890308439731598,
|
|
"num_tokens": 7720965.0,
|
|
"step": 3440
|
|
},
|
|
{
|
|
"entropy": 5.315954351425171,
|
|
"epoch": 3.7247160627366145,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004091873240764069,
|
|
"loss": 5.1258,
|
|
"mean_token_accuracy": 0.19422145783901215,
|
|
"num_tokens": 7733044.0,
|
|
"step": 3445
|
|
},
|
|
{
|
|
"entropy": 5.381974506378174,
|
|
"epoch": 3.730124391563007,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00040884278888563566,
|
|
"loss": 5.0987,
|
|
"mean_token_accuracy": 0.20453341156244279,
|
|
"num_tokens": 7743878.0,
|
|
"step": 3450
|
|
},
|
|
{
|
|
"entropy": 5.269688606262207,
|
|
"epoch": 3.7355327203893998,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00040849776730986014,
|
|
"loss": 5.0808,
|
|
"mean_token_accuracy": 0.208247509598732,
|
|
"num_tokens": 7755445.0,
|
|
"step": 3455
|
|
},
|
|
{
|
|
"entropy": 5.371634912490845,
|
|
"epoch": 3.740941049215792,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004081522606028899,
|
|
"loss": 5.1705,
|
|
"mean_token_accuracy": 0.192655149102211,
|
|
"num_tokens": 7765990.0,
|
|
"step": 3460
|
|
},
|
|
{
|
|
"entropy": 5.255275011062622,
|
|
"epoch": 3.746349378042185,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00040780627002029674,
|
|
"loss": 4.974,
|
|
"mean_token_accuracy": 0.21095251441001892,
|
|
"num_tokens": 7777064.0,
|
|
"step": 3465
|
|
},
|
|
{
|
|
"entropy": 5.2273749828338625,
|
|
"epoch": 3.7517577068685775,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004074597968194116,
|
|
"loss": 5.0073,
|
|
"mean_token_accuracy": 0.20131850391626357,
|
|
"num_tokens": 7788618.0,
|
|
"step": 3470
|
|
},
|
|
{
|
|
"entropy": 5.371102666854858,
|
|
"epoch": 3.7571660356949703,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004071128422593186,
|
|
"loss": 5.1398,
|
|
"mean_token_accuracy": 0.1931956559419632,
|
|
"num_tokens": 7798484.0,
|
|
"step": 3475
|
|
},
|
|
{
|
|
"entropy": 5.357185411453247,
|
|
"epoch": 3.7625743645213627,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004067654076008515,
|
|
"loss": 5.1016,
|
|
"mean_token_accuracy": 0.1995840236544609,
|
|
"num_tokens": 7810270.0,
|
|
"step": 3480
|
|
},
|
|
{
|
|
"entropy": 5.307087135314942,
|
|
"epoch": 3.7679826933477556,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00040641749410658877,
|
|
"loss": 5.1922,
|
|
"mean_token_accuracy": 0.19541362076997756,
|
|
"num_tokens": 7822616.0,
|
|
"step": 3485
|
|
},
|
|
{
|
|
"entropy": 5.3712913513183596,
|
|
"epoch": 3.773391022174148,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004060691030408487,
|
|
"loss": 5.1012,
|
|
"mean_token_accuracy": 0.20688474476337432,
|
|
"num_tokens": 7834933.0,
|
|
"step": 3490
|
|
},
|
|
{
|
|
"entropy": 5.350574922561646,
|
|
"epoch": 3.778799351000541,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00040572023566968536,
|
|
"loss": 5.112,
|
|
"mean_token_accuracy": 0.20370023250579833,
|
|
"num_tokens": 7847686.0,
|
|
"step": 3495
|
|
},
|
|
{
|
|
"entropy": 5.24457802772522,
|
|
"epoch": 3.7842076798269337,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00040537089326088345,
|
|
"loss": 5.1069,
|
|
"mean_token_accuracy": 0.19926053285598755,
|
|
"num_tokens": 7859087.0,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"epoch": 3.7842076798269337,
|
|
"eval_entropy": 5.3144024036147375,
|
|
"eval_loss": 5.897715091705322,
|
|
"eval_mean_token_accuracy": 0.17559453838250855,
|
|
"eval_num_tokens": 7859087.0,
|
|
"eval_runtime": 2.5268,
|
|
"eval_samples_per_second": 1391.063,
|
|
"eval_steps_per_second": 174.13,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"entropy": 5.401379251480103,
|
|
"epoch": 3.789616008653326,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004050210770839541,
|
|
"loss": 5.1591,
|
|
"mean_token_accuracy": 0.20308154225349426,
|
|
"num_tokens": 7870684.0,
|
|
"step": 3505
|
|
},
|
|
{
|
|
"entropy": 5.343354606628418,
|
|
"epoch": 3.7950243374797186,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004046707884101303,
|
|
"loss": 5.1039,
|
|
"mean_token_accuracy": 0.20302809327840804,
|
|
"num_tokens": 7880017.0,
|
|
"step": 3510
|
|
},
|
|
{
|
|
"entropy": 5.22679762840271,
|
|
"epoch": 3.8004326663061114,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00040432002851236164,
|
|
"loss": 5.0182,
|
|
"mean_token_accuracy": 0.20248215198516845,
|
|
"num_tokens": 7891751.0,
|
|
"step": 3515
|
|
},
|
|
{
|
|
"entropy": 5.417792224884034,
|
|
"epoch": 3.8058409951325043,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004039687986653106,
|
|
"loss": 5.1464,
|
|
"mean_token_accuracy": 0.20485824048519136,
|
|
"num_tokens": 7902602.0,
|
|
"step": 3520
|
|
},
|
|
{
|
|
"entropy": 5.331655025482178,
|
|
"epoch": 3.8112493239588967,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00040361710014534714,
|
|
"loss": 5.0674,
|
|
"mean_token_accuracy": 0.20003001093864442,
|
|
"num_tokens": 7913548.0,
|
|
"step": 3525
|
|
},
|
|
{
|
|
"entropy": 5.3777083396911625,
|
|
"epoch": 3.816657652785289,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004032649342305445,
|
|
"loss": 5.0718,
|
|
"mean_token_accuracy": 0.203711199760437,
|
|
"num_tokens": 7925042.0,
|
|
"step": 3530
|
|
},
|
|
{
|
|
"entropy": 5.394011116027832,
|
|
"epoch": 3.822065981611682,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004029123022006745,
|
|
"loss": 5.1342,
|
|
"mean_token_accuracy": 0.19612833112478256,
|
|
"num_tokens": 7937113.0,
|
|
"step": 3535
|
|
},
|
|
{
|
|
"entropy": 5.287375164031983,
|
|
"epoch": 3.827474310438075,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004025592053372027,
|
|
"loss": 5.0401,
|
|
"mean_token_accuracy": 0.21191460490226746,
|
|
"num_tokens": 7947796.0,
|
|
"step": 3540
|
|
},
|
|
{
|
|
"entropy": 5.343275356292724,
|
|
"epoch": 3.8328826392644673,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004022056449232839,
|
|
"loss": 5.1072,
|
|
"mean_token_accuracy": 0.19504276365041734,
|
|
"num_tokens": 7959642.0,
|
|
"step": 3545
|
|
},
|
|
{
|
|
"entropy": 5.31176872253418,
|
|
"epoch": 3.8382909680908597,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004018516222437575,
|
|
"loss": 5.0648,
|
|
"mean_token_accuracy": 0.1991144150495529,
|
|
"num_tokens": 7969996.0,
|
|
"step": 3550
|
|
},
|
|
{
|
|
"entropy": 5.273511123657227,
|
|
"epoch": 3.8436992969172525,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004014971385851428,
|
|
"loss": 5.0476,
|
|
"mean_token_accuracy": 0.20589675456285478,
|
|
"num_tokens": 7980079.0,
|
|
"step": 3555
|
|
},
|
|
{
|
|
"entropy": 5.297759771347046,
|
|
"epoch": 3.8491076257436454,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000401142195235634,
|
|
"loss": 5.1045,
|
|
"mean_token_accuracy": 0.20717412382364273,
|
|
"num_tokens": 7990453.0,
|
|
"step": 3560
|
|
},
|
|
{
|
|
"entropy": 5.37216854095459,
|
|
"epoch": 3.854515954570038,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004007867934850963,
|
|
"loss": 5.1781,
|
|
"mean_token_accuracy": 0.20191441476345062,
|
|
"num_tokens": 8001810.0,
|
|
"step": 3565
|
|
},
|
|
{
|
|
"entropy": 5.4306305885314945,
|
|
"epoch": 3.8599242833964302,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004004309346250603,
|
|
"loss": 5.0867,
|
|
"mean_token_accuracy": 0.20729905366897583,
|
|
"num_tokens": 8011821.0,
|
|
"step": 3570
|
|
},
|
|
{
|
|
"entropy": 5.269532012939453,
|
|
"epoch": 3.865332612222823,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004000746199487181,
|
|
"loss": 5.0453,
|
|
"mean_token_accuracy": 0.2095436930656433,
|
|
"num_tokens": 8024106.0,
|
|
"step": 3575
|
|
},
|
|
{
|
|
"entropy": 5.26743893623352,
|
|
"epoch": 3.870740941049216,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000399717850750918,
|
|
"loss": 5.0708,
|
|
"mean_token_accuracy": 0.206826114654541,
|
|
"num_tokens": 8035630.0,
|
|
"step": 3580
|
|
},
|
|
{
|
|
"entropy": 5.348873901367187,
|
|
"epoch": 3.8761492698756084,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00039936062832816,
|
|
"loss": 5.0914,
|
|
"mean_token_accuracy": 0.2055457666516304,
|
|
"num_tokens": 8047130.0,
|
|
"step": 3585
|
|
},
|
|
{
|
|
"entropy": 5.260032892227173,
|
|
"epoch": 3.8815575987020012,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003990029539785913,
|
|
"loss": 5.0632,
|
|
"mean_token_accuracy": 0.20160201638936998,
|
|
"num_tokens": 8059484.0,
|
|
"step": 3590
|
|
},
|
|
{
|
|
"entropy": 5.426659059524536,
|
|
"epoch": 3.8869659275283936,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00039864482900200135,
|
|
"loss": 5.1665,
|
|
"mean_token_accuracy": 0.1997232109308243,
|
|
"num_tokens": 8069467.0,
|
|
"step": 3595
|
|
},
|
|
{
|
|
"entropy": 5.398085260391236,
|
|
"epoch": 3.8923742563547865,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003982862546998169,
|
|
"loss": 5.1561,
|
|
"mean_token_accuracy": 0.20572247505187988,
|
|
"num_tokens": 8081390.0,
|
|
"step": 3600
|
|
},
|
|
{
|
|
"entropy": 5.197758960723877,
|
|
"epoch": 3.897782585181179,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003979272323750981,
|
|
"loss": 4.9524,
|
|
"mean_token_accuracy": 0.21975678503513335,
|
|
"num_tokens": 8091202.0,
|
|
"step": 3605
|
|
},
|
|
{
|
|
"entropy": 5.323245239257813,
|
|
"epoch": 3.903190914007572,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003975677633325327,
|
|
"loss": 5.0772,
|
|
"mean_token_accuracy": 0.19954726099967957,
|
|
"num_tokens": 8101900.0,
|
|
"step": 3610
|
|
},
|
|
{
|
|
"entropy": 5.337061929702759,
|
|
"epoch": 3.908599242833964,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003972078488784321,
|
|
"loss": 5.0856,
|
|
"mean_token_accuracy": 0.1967016026377678,
|
|
"num_tokens": 8114500.0,
|
|
"step": 3615
|
|
},
|
|
{
|
|
"entropy": 5.250361061096191,
|
|
"epoch": 3.914007571660357,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003968474903207263,
|
|
"loss": 4.9673,
|
|
"mean_token_accuracy": 0.21671238988637925,
|
|
"num_tokens": 8125319.0,
|
|
"step": 3620
|
|
},
|
|
{
|
|
"entropy": 5.308226442337036,
|
|
"epoch": 3.9194159004867495,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003964866889689592,
|
|
"loss": 5.1574,
|
|
"mean_token_accuracy": 0.20284880995750426,
|
|
"num_tokens": 8137077.0,
|
|
"step": 3625
|
|
},
|
|
{
|
|
"entropy": 5.299165678024292,
|
|
"epoch": 3.9248242293131423,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003961254461342834,
|
|
"loss": 5.0496,
|
|
"mean_token_accuracy": 0.2030861809849739,
|
|
"num_tokens": 8148430.0,
|
|
"step": 3630
|
|
},
|
|
{
|
|
"entropy": 5.290206718444824,
|
|
"epoch": 3.9302325581395348,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003957637631294566,
|
|
"loss": 5.0707,
|
|
"mean_token_accuracy": 0.20863705426454543,
|
|
"num_tokens": 8159250.0,
|
|
"step": 3635
|
|
},
|
|
{
|
|
"entropy": 5.371263122558593,
|
|
"epoch": 3.9356408869659276,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00039540164126883554,
|
|
"loss": 5.1485,
|
|
"mean_token_accuracy": 0.1988421067595482,
|
|
"num_tokens": 8170177.0,
|
|
"step": 3640
|
|
},
|
|
{
|
|
"entropy": 5.341719102859497,
|
|
"epoch": 3.94104921579232,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003950390818683718,
|
|
"loss": 5.1076,
|
|
"mean_token_accuracy": 0.1998066321015358,
|
|
"num_tokens": 8181351.0,
|
|
"step": 3645
|
|
},
|
|
{
|
|
"entropy": 5.252122497558593,
|
|
"epoch": 3.946457544618713,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0003946760862456071,
|
|
"loss": 5.1134,
|
|
"mean_token_accuracy": 0.2028968945145607,
|
|
"num_tokens": 8193472.0,
|
|
"step": 3650
|
|
},
|
|
{
|
|
"entropy": 5.388597917556763,
|
|
"epoch": 3.9518658734451053,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003943126557196685,
|
|
"loss": 5.2036,
|
|
"mean_token_accuracy": 0.20796288251876832,
|
|
"num_tokens": 8205449.0,
|
|
"step": 3655
|
|
},
|
|
{
|
|
"entropy": 5.24415340423584,
|
|
"epoch": 3.957274202271498,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00039394879161126333,
|
|
"loss": 4.953,
|
|
"mean_token_accuracy": 0.2125275358557701,
|
|
"num_tokens": 8216520.0,
|
|
"step": 3660
|
|
},
|
|
{
|
|
"entropy": 5.264657258987427,
|
|
"epoch": 3.9626825310978906,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003935844952426745,
|
|
"loss": 5.17,
|
|
"mean_token_accuracy": 0.19591122716665268,
|
|
"num_tokens": 8228727.0,
|
|
"step": 3665
|
|
},
|
|
{
|
|
"entropy": 5.405154705047607,
|
|
"epoch": 3.9680908599242835,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00039321976793775586,
|
|
"loss": 5.0854,
|
|
"mean_token_accuracy": 0.20178651362657546,
|
|
"num_tokens": 8240423.0,
|
|
"step": 3670
|
|
},
|
|
{
|
|
"entropy": 5.246245574951172,
|
|
"epoch": 3.973499188750676,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003928546110219275,
|
|
"loss": 5.0137,
|
|
"mean_token_accuracy": 0.20742493718862534,
|
|
"num_tokens": 8252757.0,
|
|
"step": 3675
|
|
},
|
|
{
|
|
"entropy": 5.140897989273071,
|
|
"epoch": 3.9789075175770687,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00039248902582217024,
|
|
"loss": 5.0012,
|
|
"mean_token_accuracy": 0.20537508130073548,
|
|
"num_tokens": 8263462.0,
|
|
"step": 3680
|
|
},
|
|
{
|
|
"entropy": 5.318847036361694,
|
|
"epoch": 3.9843158464034616,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00039212301366702153,
|
|
"loss": 5.0584,
|
|
"mean_token_accuracy": 0.20191151648759842,
|
|
"num_tokens": 8274364.0,
|
|
"step": 3685
|
|
},
|
|
{
|
|
"entropy": 5.289162969589233,
|
|
"epoch": 3.989724175229854,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00039175657588657063,
|
|
"loss": 5.0824,
|
|
"mean_token_accuracy": 0.20226163864135743,
|
|
"num_tokens": 8286026.0,
|
|
"step": 3690
|
|
},
|
|
{
|
|
"entropy": 5.342779731750488,
|
|
"epoch": 3.9951325040562464,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003913897138124531,
|
|
"loss": 5.1119,
|
|
"mean_token_accuracy": 0.20729459822177887,
|
|
"num_tokens": 8297990.0,
|
|
"step": 3695
|
|
},
|
|
{
|
|
"entropy": 5.4067295922173395,
|
|
"epoch": 4.0,
|
|
"grad_norm": 2.421875,
|
|
"learning_rate": 0.0003910224287778466,
|
|
"loss": 5.1529,
|
|
"mean_token_accuracy": 0.19613086846139696,
|
|
"num_tokens": 8307556.0,
|
|
"step": 3700
|
|
},
|
|
{
|
|
"entropy": 5.28830919265747,
|
|
"epoch": 4.005408328826393,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00039065472211746586,
|
|
"loss": 4.8361,
|
|
"mean_token_accuracy": 0.21606979817152022,
|
|
"num_tokens": 8320805.0,
|
|
"step": 3705
|
|
},
|
|
{
|
|
"entropy": 5.304356241226197,
|
|
"epoch": 4.010816657652786,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003902865951675576,
|
|
"loss": 4.8288,
|
|
"mean_token_accuracy": 0.21535313576459886,
|
|
"num_tokens": 8332390.0,
|
|
"step": 3710
|
|
},
|
|
{
|
|
"entropy": 5.24171028137207,
|
|
"epoch": 4.016224986479178,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00038991804926589616,
|
|
"loss": 4.7318,
|
|
"mean_token_accuracy": 0.22037425637245178,
|
|
"num_tokens": 8342325.0,
|
|
"step": 3715
|
|
},
|
|
{
|
|
"entropy": 5.3355223655700685,
|
|
"epoch": 4.0216333153055706,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000389549085751778,
|
|
"loss": 4.7492,
|
|
"mean_token_accuracy": 0.2229212775826454,
|
|
"num_tokens": 8352115.0,
|
|
"step": 3720
|
|
},
|
|
{
|
|
"entropy": 5.263841152191162,
|
|
"epoch": 4.027041644131963,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00038917970596601765,
|
|
"loss": 4.7949,
|
|
"mean_token_accuracy": 0.22654144763946532,
|
|
"num_tokens": 8364265.0,
|
|
"step": 3725
|
|
},
|
|
{
|
|
"entropy": 5.344356918334961,
|
|
"epoch": 4.032449972958356,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003888099112509419,
|
|
"loss": 4.7734,
|
|
"mean_token_accuracy": 0.2224753811955452,
|
|
"num_tokens": 8375979.0,
|
|
"step": 3730
|
|
},
|
|
{
|
|
"entropy": 5.294282484054565,
|
|
"epoch": 4.037858301784748,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00038843970295038564,
|
|
"loss": 4.7625,
|
|
"mean_token_accuracy": 0.22473599314689635,
|
|
"num_tokens": 8387794.0,
|
|
"step": 3735
|
|
},
|
|
{
|
|
"entropy": 5.217307710647583,
|
|
"epoch": 4.043266630611141,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00038806908240968666,
|
|
"loss": 4.6931,
|
|
"mean_token_accuracy": 0.22251309156417848,
|
|
"num_tokens": 8398796.0,
|
|
"step": 3740
|
|
},
|
|
{
|
|
"entropy": 5.226292943954467,
|
|
"epoch": 4.048674959437534,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000387698050975681,
|
|
"loss": 4.7922,
|
|
"mean_token_accuracy": 0.228670634329319,
|
|
"num_tokens": 8408788.0,
|
|
"step": 3745
|
|
},
|
|
{
|
|
"entropy": 5.286603784561157,
|
|
"epoch": 4.054083288263927,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003873266099966976,
|
|
"loss": 4.8169,
|
|
"mean_token_accuracy": 0.21755992323160173,
|
|
"num_tokens": 8420373.0,
|
|
"step": 3750
|
|
},
|
|
{
|
|
"entropy": 5.199351453781128,
|
|
"epoch": 4.059491617090319,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00038695476082255395,
|
|
"loss": 4.7688,
|
|
"mean_token_accuracy": 0.23158731907606125,
|
|
"num_tokens": 8430399.0,
|
|
"step": 3755
|
|
},
|
|
{
|
|
"entropy": 5.328314304351807,
|
|
"epoch": 4.064899945916712,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003865825048045507,
|
|
"loss": 4.8625,
|
|
"mean_token_accuracy": 0.2164784699678421,
|
|
"num_tokens": 8442649.0,
|
|
"step": 3760
|
|
},
|
|
{
|
|
"entropy": 5.40818305015564,
|
|
"epoch": 4.0703082747431045,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003862098432954672,
|
|
"loss": 4.8818,
|
|
"mean_token_accuracy": 0.2166826009750366,
|
|
"num_tokens": 8453985.0,
|
|
"step": 3765
|
|
},
|
|
{
|
|
"entropy": 5.317957496643066,
|
|
"epoch": 4.075716603569497,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003858367776495561,
|
|
"loss": 4.9031,
|
|
"mean_token_accuracy": 0.21231857538223267,
|
|
"num_tokens": 8465566.0,
|
|
"step": 3770
|
|
},
|
|
{
|
|
"entropy": 5.215158557891845,
|
|
"epoch": 4.081124932395889,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00038546330922253886,
|
|
"loss": 4.6877,
|
|
"mean_token_accuracy": 0.23677795082330705,
|
|
"num_tokens": 8477426.0,
|
|
"step": 3775
|
|
},
|
|
{
|
|
"entropy": 5.200477647781372,
|
|
"epoch": 4.086533261222282,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003850894393716007,
|
|
"loss": 4.8043,
|
|
"mean_token_accuracy": 0.22476764619350434,
|
|
"num_tokens": 8489654.0,
|
|
"step": 3780
|
|
},
|
|
{
|
|
"entropy": 5.199593210220337,
|
|
"epoch": 4.091941590048675,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003847151694553855,
|
|
"loss": 4.7793,
|
|
"mean_token_accuracy": 0.22019775360822677,
|
|
"num_tokens": 8500398.0,
|
|
"step": 3785
|
|
},
|
|
{
|
|
"entropy": 5.182802152633667,
|
|
"epoch": 4.097349918875068,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003843405008339908,
|
|
"loss": 4.7149,
|
|
"mean_token_accuracy": 0.22064192295074464,
|
|
"num_tokens": 8511974.0,
|
|
"step": 3790
|
|
},
|
|
{
|
|
"entropy": 5.2203083515167235,
|
|
"epoch": 4.10275824770146,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00038396543486896354,
|
|
"loss": 4.7866,
|
|
"mean_token_accuracy": 0.22132647782564163,
|
|
"num_tokens": 8523140.0,
|
|
"step": 3795
|
|
},
|
|
{
|
|
"entropy": 5.228065633773804,
|
|
"epoch": 4.108166576527853,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00038358997292329406,
|
|
"loss": 4.7894,
|
|
"mean_token_accuracy": 0.22525623589754104,
|
|
"num_tokens": 8534261.0,
|
|
"step": 3800
|
|
},
|
|
{
|
|
"entropy": 5.325959253311157,
|
|
"epoch": 4.113574905354246,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00038321411636141214,
|
|
"loss": 4.8544,
|
|
"mean_token_accuracy": 0.22114271223545073,
|
|
"num_tokens": 8544620.0,
|
|
"step": 3805
|
|
},
|
|
{
|
|
"entropy": 5.186372089385986,
|
|
"epoch": 4.1189832341806385,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003828378665491811,
|
|
"loss": 4.8034,
|
|
"mean_token_accuracy": 0.222377946972847,
|
|
"num_tokens": 8555427.0,
|
|
"step": 3810
|
|
},
|
|
{
|
|
"entropy": 5.249618768692017,
|
|
"epoch": 4.1243915630070305,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000382461224853894,
|
|
"loss": 4.7654,
|
|
"mean_token_accuracy": 0.22093690186738968,
|
|
"num_tokens": 8566956.0,
|
|
"step": 3815
|
|
},
|
|
{
|
|
"entropy": 5.297764873504638,
|
|
"epoch": 4.129799891833423,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003820841926442674,
|
|
"loss": 4.9048,
|
|
"mean_token_accuracy": 0.2140451639890671,
|
|
"num_tokens": 8579373.0,
|
|
"step": 3820
|
|
},
|
|
{
|
|
"entropy": 5.210286283493042,
|
|
"epoch": 4.135208220659816,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003817067712904372,
|
|
"loss": 4.7396,
|
|
"mean_token_accuracy": 0.222443388402462,
|
|
"num_tokens": 8590979.0,
|
|
"step": 3825
|
|
},
|
|
{
|
|
"entropy": 5.170242643356323,
|
|
"epoch": 4.140616549486209,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00038132896216395355,
|
|
"loss": 4.7827,
|
|
"mean_token_accuracy": 0.22016517221927642,
|
|
"num_tokens": 8602514.0,
|
|
"step": 3830
|
|
},
|
|
{
|
|
"entropy": 5.168073034286499,
|
|
"epoch": 4.146024878312601,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003809507666377758,
|
|
"loss": 4.7419,
|
|
"mean_token_accuracy": 0.22809492647647858,
|
|
"num_tokens": 8613766.0,
|
|
"step": 3835
|
|
},
|
|
{
|
|
"entropy": 5.2878227710723875,
|
|
"epoch": 4.151433207138994,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00038057218608626714,
|
|
"loss": 4.8527,
|
|
"mean_token_accuracy": 0.20982229262590407,
|
|
"num_tokens": 8625245.0,
|
|
"step": 3840
|
|
},
|
|
{
|
|
"entropy": 5.257644701004028,
|
|
"epoch": 4.156841535965387,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003801932218851903,
|
|
"loss": 4.7874,
|
|
"mean_token_accuracy": 0.22776189297437668,
|
|
"num_tokens": 8635653.0,
|
|
"step": 3845
|
|
},
|
|
{
|
|
"entropy": 5.194669532775879,
|
|
"epoch": 4.16224986479178,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000379813875411702,
|
|
"loss": 4.7386,
|
|
"mean_token_accuracy": 0.22557467222213745,
|
|
"num_tokens": 8646671.0,
|
|
"step": 3850
|
|
},
|
|
{
|
|
"entropy": 5.1517833232879635,
|
|
"epoch": 4.167658193618172,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00037943414804434826,
|
|
"loss": 4.6968,
|
|
"mean_token_accuracy": 0.22484721392393112,
|
|
"num_tokens": 8656441.0,
|
|
"step": 3855
|
|
},
|
|
{
|
|
"entropy": 5.214590167999267,
|
|
"epoch": 4.173066522444564,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003790540411630592,
|
|
"loss": 4.7772,
|
|
"mean_token_accuracy": 0.22223643958568573,
|
|
"num_tokens": 8666632.0,
|
|
"step": 3860
|
|
},
|
|
{
|
|
"entropy": 5.195592212677002,
|
|
"epoch": 4.178474851270957,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00037867355614914404,
|
|
"loss": 4.8048,
|
|
"mean_token_accuracy": 0.2189110189676285,
|
|
"num_tokens": 8678269.0,
|
|
"step": 3865
|
|
},
|
|
{
|
|
"entropy": 5.2433844089508055,
|
|
"epoch": 4.18388318009735,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003782926943852864,
|
|
"loss": 4.8363,
|
|
"mean_token_accuracy": 0.21939649134874345,
|
|
"num_tokens": 8690004.0,
|
|
"step": 3870
|
|
},
|
|
{
|
|
"entropy": 5.295675325393677,
|
|
"epoch": 4.189291508923742,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00037791145725553853,
|
|
"loss": 4.8537,
|
|
"mean_token_accuracy": 0.2160766080021858,
|
|
"num_tokens": 8702306.0,
|
|
"step": 3875
|
|
},
|
|
{
|
|
"entropy": 5.251738739013672,
|
|
"epoch": 4.194699837750135,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003775298461453172,
|
|
"loss": 4.8746,
|
|
"mean_token_accuracy": 0.21798058450222016,
|
|
"num_tokens": 8714070.0,
|
|
"step": 3880
|
|
},
|
|
{
|
|
"entropy": 5.217594766616822,
|
|
"epoch": 4.200108166576528,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000377147862441398,
|
|
"loss": 4.8259,
|
|
"mean_token_accuracy": 0.22422219514846803,
|
|
"num_tokens": 8724433.0,
|
|
"step": 3885
|
|
},
|
|
{
|
|
"entropy": 5.200416803359985,
|
|
"epoch": 4.205516495402921,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00037676550753191075,
|
|
"loss": 4.7591,
|
|
"mean_token_accuracy": 0.23396038860082627,
|
|
"num_tokens": 8735485.0,
|
|
"step": 3890
|
|
},
|
|
{
|
|
"entropy": 5.221590518951416,
|
|
"epoch": 4.210924824229313,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003763827828063339,
|
|
"loss": 4.8072,
|
|
"mean_token_accuracy": 0.21032058596611022,
|
|
"num_tokens": 8746356.0,
|
|
"step": 3895
|
|
},
|
|
{
|
|
"entropy": 5.23051643371582,
|
|
"epoch": 4.2163331530557056,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00037599968965549,
|
|
"loss": 4.8148,
|
|
"mean_token_accuracy": 0.21877963542938234,
|
|
"num_tokens": 8757667.0,
|
|
"step": 3900
|
|
},
|
|
{
|
|
"entropy": 5.212504768371582,
|
|
"epoch": 4.221741481882098,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003756162294715405,
|
|
"loss": 4.7615,
|
|
"mean_token_accuracy": 0.2238036498427391,
|
|
"num_tokens": 8768565.0,
|
|
"step": 3905
|
|
},
|
|
{
|
|
"entropy": 5.170410299301148,
|
|
"epoch": 4.227149810708491,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00037523240364798063,
|
|
"loss": 4.7978,
|
|
"mean_token_accuracy": 0.22108596861362456,
|
|
"num_tokens": 8779608.0,
|
|
"step": 3910
|
|
},
|
|
{
|
|
"entropy": 5.253938817977906,
|
|
"epoch": 4.232558139534884,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00037484821357963423,
|
|
"loss": 4.8131,
|
|
"mean_token_accuracy": 0.22581763565540314,
|
|
"num_tokens": 8791492.0,
|
|
"step": 3915
|
|
},
|
|
{
|
|
"entropy": 5.212199449539185,
|
|
"epoch": 4.237966468361276,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000374463660662649,
|
|
"loss": 4.8383,
|
|
"mean_token_accuracy": 0.22084805369377136,
|
|
"num_tokens": 8803280.0,
|
|
"step": 3920
|
|
},
|
|
{
|
|
"entropy": 5.238509893417358,
|
|
"epoch": 4.243374797187669,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000374078746294491,
|
|
"loss": 4.8572,
|
|
"mean_token_accuracy": 0.21829076260328292,
|
|
"num_tokens": 8814626.0,
|
|
"step": 3925
|
|
},
|
|
{
|
|
"entropy": 5.167253684997559,
|
|
"epoch": 4.248783126014062,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00037369347187393996,
|
|
"loss": 4.7029,
|
|
"mean_token_accuracy": 0.23064414262771607,
|
|
"num_tokens": 8825520.0,
|
|
"step": 3930
|
|
},
|
|
{
|
|
"entropy": 5.16299958229065,
|
|
"epoch": 4.254191454840455,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003733078388010842,
|
|
"loss": 4.7199,
|
|
"mean_token_accuracy": 0.22837960422039033,
|
|
"num_tokens": 8835980.0,
|
|
"step": 3935
|
|
},
|
|
{
|
|
"entropy": 5.287345218658447,
|
|
"epoch": 4.259599783666847,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003729218484773149,
|
|
"loss": 4.8996,
|
|
"mean_token_accuracy": 0.2098175972700119,
|
|
"num_tokens": 8846714.0,
|
|
"step": 3940
|
|
},
|
|
{
|
|
"entropy": 5.2743062496185305,
|
|
"epoch": 4.2650081124932395,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003725355023053221,
|
|
"loss": 4.8315,
|
|
"mean_token_accuracy": 0.22190943360328674,
|
|
"num_tokens": 8858215.0,
|
|
"step": 3945
|
|
},
|
|
{
|
|
"entropy": 5.11496639251709,
|
|
"epoch": 4.270416441319632,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003721488016890884,
|
|
"loss": 4.7212,
|
|
"mean_token_accuracy": 0.22722370475530623,
|
|
"num_tokens": 8870587.0,
|
|
"step": 3950
|
|
},
|
|
{
|
|
"entropy": 5.125867462158203,
|
|
"epoch": 4.275824770146025,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00037176174803388496,
|
|
"loss": 4.7901,
|
|
"mean_token_accuracy": 0.22261050939559937,
|
|
"num_tokens": 8881335.0,
|
|
"step": 3955
|
|
},
|
|
{
|
|
"entropy": 5.170544624328613,
|
|
"epoch": 4.281233098972417,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00037137434274626545,
|
|
"loss": 4.7672,
|
|
"mean_token_accuracy": 0.21677591353654863,
|
|
"num_tokens": 8893294.0,
|
|
"step": 3960
|
|
},
|
|
{
|
|
"entropy": 5.23744649887085,
|
|
"epoch": 4.28664142779881,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003709865872340618,
|
|
"loss": 4.8542,
|
|
"mean_token_accuracy": 0.21343087702989577,
|
|
"num_tokens": 8903914.0,
|
|
"step": 3965
|
|
},
|
|
{
|
|
"entropy": 5.258846569061279,
|
|
"epoch": 4.292049756625203,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003705984829063782,
|
|
"loss": 4.8288,
|
|
"mean_token_accuracy": 0.2160489246249199,
|
|
"num_tokens": 8915591.0,
|
|
"step": 3970
|
|
},
|
|
{
|
|
"entropy": 5.187211036682129,
|
|
"epoch": 4.297458085451596,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003702100311735869,
|
|
"loss": 4.8164,
|
|
"mean_token_accuracy": 0.21521014869213104,
|
|
"num_tokens": 8927031.0,
|
|
"step": 3975
|
|
},
|
|
{
|
|
"entropy": 5.170180940628052,
|
|
"epoch": 4.302866414277988,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00036982123344732223,
|
|
"loss": 4.7492,
|
|
"mean_token_accuracy": 0.22640395760536194,
|
|
"num_tokens": 8937220.0,
|
|
"step": 3980
|
|
},
|
|
{
|
|
"entropy": 5.188808250427246,
|
|
"epoch": 4.308274743104381,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00036943209114047613,
|
|
"loss": 4.7588,
|
|
"mean_token_accuracy": 0.2211114525794983,
|
|
"num_tokens": 8948032.0,
|
|
"step": 3985
|
|
},
|
|
{
|
|
"entropy": 5.131969738006592,
|
|
"epoch": 4.3136830719307735,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0003690426056671926,
|
|
"loss": 4.858,
|
|
"mean_token_accuracy": 0.21378480195999144,
|
|
"num_tokens": 8960735.0,
|
|
"step": 3990
|
|
},
|
|
{
|
|
"entropy": 5.182211923599243,
|
|
"epoch": 4.319091400757166,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003686527784428627,
|
|
"loss": 4.7393,
|
|
"mean_token_accuracy": 0.22918514907360077,
|
|
"num_tokens": 8971801.0,
|
|
"step": 3995
|
|
},
|
|
{
|
|
"entropy": 5.206542015075684,
|
|
"epoch": 4.324499729583558,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00036826261088411955,
|
|
"loss": 4.835,
|
|
"mean_token_accuracy": 0.21574117839336396,
|
|
"num_tokens": 8983041.0,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"epoch": 4.324499729583558,
|
|
"eval_entropy": 5.124637711048126,
|
|
"eval_loss": 5.907151222229004,
|
|
"eval_mean_token_accuracy": 0.17702669254927472,
|
|
"eval_num_tokens": 8983041.0,
|
|
"eval_runtime": 2.5248,
|
|
"eval_samples_per_second": 1392.189,
|
|
"eval_steps_per_second": 174.271,
|
|
"step": 4000
|
|
}
|
|
],
|
|
"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": 1.4032664082432e+16,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|