1035 lines
27 KiB
JSON
1035 lines
27 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.032988058322887114,
|
|
"eval_steps": 500,
|
|
"global_step": 500,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.691536808013916,
|
|
"epoch": 0.00032988058322887113,
|
|
"grad_norm": 11.9375,
|
|
"learning_rate": 2e-06,
|
|
"loss": 10.8148,
|
|
"mean_token_accuracy": 8.525149314664304e-05,
|
|
"num_tokens": 11578.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.691540431976318,
|
|
"epoch": 0.0006597611664577423,
|
|
"grad_norm": 10.6875,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 10.7492,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 22232.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.691376876831054,
|
|
"epoch": 0.0009896417496866134,
|
|
"grad_norm": 8.625,
|
|
"learning_rate": 7e-06,
|
|
"loss": 10.5708,
|
|
"mean_token_accuracy": 0.01264126447495073,
|
|
"num_tokens": 35495.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.685009479522705,
|
|
"epoch": 0.0013195223329154845,
|
|
"grad_norm": 6.3125,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 10.2629,
|
|
"mean_token_accuracy": 0.03841282296925783,
|
|
"num_tokens": 47005.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.637744235992432,
|
|
"epoch": 0.0016494029161443558,
|
|
"grad_norm": 3.84375,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 9.9947,
|
|
"mean_token_accuracy": 0.04075138829648495,
|
|
"num_tokens": 57675.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.588305950164795,
|
|
"epoch": 0.001979283499373227,
|
|
"grad_norm": 3.21875,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 9.9135,
|
|
"mean_token_accuracy": 0.03900990225374699,
|
|
"num_tokens": 70455.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.60552625656128,
|
|
"epoch": 0.002309164082602098,
|
|
"grad_norm": 2.609375,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 9.8171,
|
|
"mean_token_accuracy": 0.03996471259742975,
|
|
"num_tokens": 82422.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.598775672912598,
|
|
"epoch": 0.002639044665830969,
|
|
"grad_norm": 2.484375,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 9.7593,
|
|
"mean_token_accuracy": 0.035367689467966555,
|
|
"num_tokens": 92109.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.583949089050293,
|
|
"epoch": 0.0029689252490598403,
|
|
"grad_norm": 2.40625,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 9.7105,
|
|
"mean_token_accuracy": 0.04035525936633348,
|
|
"num_tokens": 103440.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.578855609893798,
|
|
"epoch": 0.0032988058322887116,
|
|
"grad_norm": 2.3125,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 9.6799,
|
|
"mean_token_accuracy": 0.040092948265373705,
|
|
"num_tokens": 114742.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.571938228607177,
|
|
"epoch": 0.0036286864155175825,
|
|
"grad_norm": 2.203125,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.5897,
|
|
"mean_token_accuracy": 0.043487312830984594,
|
|
"num_tokens": 125626.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.556445789337157,
|
|
"epoch": 0.003958566998746454,
|
|
"grad_norm": 2.21875,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.5516,
|
|
"mean_token_accuracy": 0.043048874288797376,
|
|
"num_tokens": 137957.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.548345756530761,
|
|
"epoch": 0.004288447581975325,
|
|
"grad_norm": 2.828125,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.4677,
|
|
"mean_token_accuracy": 0.04039878956973553,
|
|
"num_tokens": 148277.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.532929801940918,
|
|
"epoch": 0.004618328165204196,
|
|
"grad_norm": 2.046875,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.424,
|
|
"mean_token_accuracy": 0.046445511281490326,
|
|
"num_tokens": 159878.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.445225620269776,
|
|
"epoch": 0.004948208748433067,
|
|
"grad_norm": 2.15625,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.2866,
|
|
"mean_token_accuracy": 0.04957378730177879,
|
|
"num_tokens": 172156.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.42960033416748,
|
|
"epoch": 0.005278089331661938,
|
|
"grad_norm": 2.109375,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.2163,
|
|
"mean_token_accuracy": 0.04596305675804615,
|
|
"num_tokens": 183913.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.408072757720948,
|
|
"epoch": 0.00560796991489081,
|
|
"grad_norm": 1.953125,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.1345,
|
|
"mean_token_accuracy": 0.05241272747516632,
|
|
"num_tokens": 195817.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.41554536819458,
|
|
"epoch": 0.005937850498119681,
|
|
"grad_norm": 2.09375,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 9.0372,
|
|
"mean_token_accuracy": 0.05409228093922138,
|
|
"num_tokens": 207962.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.360993576049804,
|
|
"epoch": 0.0062677310813485515,
|
|
"grad_norm": 2.265625,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 8.9542,
|
|
"mean_token_accuracy": 0.0519526032730937,
|
|
"num_tokens": 219781.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.258287715911866,
|
|
"epoch": 0.006597611664577423,
|
|
"grad_norm": 2.015625,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 8.7831,
|
|
"mean_token_accuracy": 0.05676806531846523,
|
|
"num_tokens": 231163.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.228107357025147,
|
|
"epoch": 0.006927492247806294,
|
|
"grad_norm": 1.90625,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 8.683,
|
|
"mean_token_accuracy": 0.05635010786354542,
|
|
"num_tokens": 242141.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.161042213439941,
|
|
"epoch": 0.007257372831035165,
|
|
"grad_norm": 1.9609375,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 8.564,
|
|
"mean_token_accuracy": 0.057049044594168666,
|
|
"num_tokens": 253879.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 10.023370933532714,
|
|
"epoch": 0.007587253414264037,
|
|
"grad_norm": 1.96875,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 8.3791,
|
|
"mean_token_accuracy": 0.07059949524700641,
|
|
"num_tokens": 264486.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 9.973721122741699,
|
|
"epoch": 0.007917133997492908,
|
|
"grad_norm": 2.15625,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.3102,
|
|
"mean_token_accuracy": 0.06145399771630764,
|
|
"num_tokens": 275750.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 9.877373790740966,
|
|
"epoch": 0.008247014580721778,
|
|
"grad_norm": 1.796875,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.2344,
|
|
"mean_token_accuracy": 0.05942905358970165,
|
|
"num_tokens": 287746.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 9.738166522979736,
|
|
"epoch": 0.00857689516395065,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 8.0654,
|
|
"mean_token_accuracy": 0.06289364024996758,
|
|
"num_tokens": 300383.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 9.535709667205811,
|
|
"epoch": 0.008906775747179522,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 7.9878,
|
|
"mean_token_accuracy": 0.06938982531428337,
|
|
"num_tokens": 313091.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 9.33922004699707,
|
|
"epoch": 0.009236656330408393,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 7.8281,
|
|
"mean_token_accuracy": 0.0680146910250187,
|
|
"num_tokens": 323251.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 9.173285961151123,
|
|
"epoch": 0.009566536913637264,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 7.7617,
|
|
"mean_token_accuracy": 0.06867263428866863,
|
|
"num_tokens": 333707.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 8.991851806640625,
|
|
"epoch": 0.009896417496866134,
|
|
"grad_norm": 1.9375,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 7.7034,
|
|
"mean_token_accuracy": 0.06648395135998726,
|
|
"num_tokens": 345426.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 8.759401321411133,
|
|
"epoch": 0.010226298080095005,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 7.6818,
|
|
"mean_token_accuracy": 0.06103574819862843,
|
|
"num_tokens": 358876.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 8.579740428924561,
|
|
"epoch": 0.010556178663323876,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 7.629,
|
|
"mean_token_accuracy": 0.06325341649353504,
|
|
"num_tokens": 370955.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 8.471101760864258,
|
|
"epoch": 0.010886059246552749,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.5337,
|
|
"mean_token_accuracy": 0.07028876058757305,
|
|
"num_tokens": 383667.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 8.323129177093506,
|
|
"epoch": 0.01121593982978162,
|
|
"grad_norm": 2.1875,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.5188,
|
|
"mean_token_accuracy": 0.061643701419234274,
|
|
"num_tokens": 394347.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 8.2438063621521,
|
|
"epoch": 0.01154582041301049,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.4512,
|
|
"mean_token_accuracy": 0.06954124197363853,
|
|
"num_tokens": 406589.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.109932041168213,
|
|
"epoch": 0.011875700996239361,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.4349,
|
|
"mean_token_accuracy": 0.07020395621657372,
|
|
"num_tokens": 417462.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.160281562805176,
|
|
"epoch": 0.012205581579468232,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.5008,
|
|
"mean_token_accuracy": 0.0684724546968937,
|
|
"num_tokens": 428076.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 8.130372238159179,
|
|
"epoch": 0.012535462162697103,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.4143,
|
|
"mean_token_accuracy": 0.06652588658034801,
|
|
"num_tokens": 438714.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 8.05902099609375,
|
|
"epoch": 0.012865342745925976,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.401,
|
|
"mean_token_accuracy": 0.07104812636971473,
|
|
"num_tokens": 450373.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 8.005074548721314,
|
|
"epoch": 0.013195223329154846,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.3499,
|
|
"mean_token_accuracy": 0.07001710906624795,
|
|
"num_tokens": 461201.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 7.956696891784668,
|
|
"epoch": 0.013525103912383717,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.3057,
|
|
"mean_token_accuracy": 0.07321103028953076,
|
|
"num_tokens": 471519.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 7.904478693008423,
|
|
"epoch": 0.013854984495612588,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.3685,
|
|
"mean_token_accuracy": 0.07146023064851761,
|
|
"num_tokens": 483245.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 7.853240013122559,
|
|
"epoch": 0.014184865078841459,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.2524,
|
|
"mean_token_accuracy": 0.07811010889708996,
|
|
"num_tokens": 494398.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 7.8232927322387695,
|
|
"epoch": 0.01451474566207033,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.2768,
|
|
"mean_token_accuracy": 0.08055114448070526,
|
|
"num_tokens": 505550.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 7.807443523406983,
|
|
"epoch": 0.014844626245299202,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.2205,
|
|
"mean_token_accuracy": 0.06916935592889786,
|
|
"num_tokens": 515824.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 7.825547695159912,
|
|
"epoch": 0.015174506828528073,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.267,
|
|
"mean_token_accuracy": 0.07258688844740391,
|
|
"num_tokens": 527272.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 7.786549615859985,
|
|
"epoch": 0.015504387411756944,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.2425,
|
|
"mean_token_accuracy": 0.07283572107553482,
|
|
"num_tokens": 537970.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 7.657060623168945,
|
|
"epoch": 0.015834267994985815,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.1359,
|
|
"mean_token_accuracy": 0.07426832728087902,
|
|
"num_tokens": 549026.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.728560590744019,
|
|
"epoch": 0.016164148578214688,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.2478,
|
|
"mean_token_accuracy": 0.07338454909622669,
|
|
"num_tokens": 561001.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.736995792388916,
|
|
"epoch": 0.016494029161443557,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.1899,
|
|
"mean_token_accuracy": 0.07819501236081124,
|
|
"num_tokens": 571910.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 7.743732023239136,
|
|
"epoch": 0.01682390974467243,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.1879,
|
|
"mean_token_accuracy": 0.06838022135198116,
|
|
"num_tokens": 583073.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.579653358459472,
|
|
"epoch": 0.0171537903279013,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.1182,
|
|
"mean_token_accuracy": 0.07753507420420647,
|
|
"num_tokens": 593909.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.594156074523926,
|
|
"epoch": 0.01748367091113017,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.1641,
|
|
"mean_token_accuracy": 0.07598368227481841,
|
|
"num_tokens": 605235.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.632419633865356,
|
|
"epoch": 0.017813551494359044,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 7.0944,
|
|
"mean_token_accuracy": 0.07657683044672012,
|
|
"num_tokens": 616491.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.52728009223938,
|
|
"epoch": 0.018143432077587913,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.1101,
|
|
"mean_token_accuracy": 0.08200965896248817,
|
|
"num_tokens": 627576.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.5737220287323,
|
|
"epoch": 0.018473312660816785,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 7.026,
|
|
"mean_token_accuracy": 0.08090509474277496,
|
|
"num_tokens": 638833.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.627187490463257,
|
|
"epoch": 0.018803193244045655,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.1247,
|
|
"mean_token_accuracy": 0.07016714699566365,
|
|
"num_tokens": 648954.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.527385234832764,
|
|
"epoch": 0.019133073827274527,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.0426,
|
|
"mean_token_accuracy": 0.07938065230846406,
|
|
"num_tokens": 659653.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.521616458892822,
|
|
"epoch": 0.019462954410503396,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.1504,
|
|
"mean_token_accuracy": 0.073289405554533,
|
|
"num_tokens": 671282.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.579449462890625,
|
|
"epoch": 0.01979283499373227,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 7.0938,
|
|
"mean_token_accuracy": 0.07353611513972283,
|
|
"num_tokens": 683096.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.519705104827881,
|
|
"epoch": 0.02012271557696114,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000152,
|
|
"loss": 7.0635,
|
|
"mean_token_accuracy": 0.08060178570449353,
|
|
"num_tokens": 694278.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.556258678436279,
|
|
"epoch": 0.02045259616019001,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 7.0468,
|
|
"mean_token_accuracy": 0.07418892979621887,
|
|
"num_tokens": 705878.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.441033935546875,
|
|
"epoch": 0.020782476743418883,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000157,
|
|
"loss": 7.0041,
|
|
"mean_token_accuracy": 0.07709568962454796,
|
|
"num_tokens": 717459.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.459104728698731,
|
|
"epoch": 0.021112357326647752,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 6.9901,
|
|
"mean_token_accuracy": 0.07908840253949165,
|
|
"num_tokens": 726708.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.5175848484039305,
|
|
"epoch": 0.021442237909876625,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000162,
|
|
"loss": 6.9931,
|
|
"mean_token_accuracy": 0.08139066398143768,
|
|
"num_tokens": 736799.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.408169412612915,
|
|
"epoch": 0.021772118493105497,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 6.9851,
|
|
"mean_token_accuracy": 0.07909020446240903,
|
|
"num_tokens": 747820.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.367802715301513,
|
|
"epoch": 0.022101999076334367,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 6.9862,
|
|
"mean_token_accuracy": 0.08260660097002984,
|
|
"num_tokens": 757938.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.41139817237854,
|
|
"epoch": 0.02243187965956324,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 6.9832,
|
|
"mean_token_accuracy": 0.08281685560941696,
|
|
"num_tokens": 768900.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.428646087646484,
|
|
"epoch": 0.02276176024279211,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 7.0474,
|
|
"mean_token_accuracy": 0.07662288472056389,
|
|
"num_tokens": 780839.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.402420568466186,
|
|
"epoch": 0.02309164082602098,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 6.9878,
|
|
"mean_token_accuracy": 0.08407922312617302,
|
|
"num_tokens": 793144.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.35455961227417,
|
|
"epoch": 0.02342152140924985,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000177,
|
|
"loss": 6.9616,
|
|
"mean_token_accuracy": 0.07600980773568153,
|
|
"num_tokens": 805293.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.352949094772339,
|
|
"epoch": 0.023751401992478723,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 6.9015,
|
|
"mean_token_accuracy": 0.08443449512124061,
|
|
"num_tokens": 817313.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.3078234672546385,
|
|
"epoch": 0.024081282575707595,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.000182,
|
|
"loss": 6.8506,
|
|
"mean_token_accuracy": 0.08551364466547966,
|
|
"num_tokens": 827823.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.300553846359253,
|
|
"epoch": 0.024411163158936464,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 6.8796,
|
|
"mean_token_accuracy": 0.08822695687413215,
|
|
"num_tokens": 839221.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.312422800064087,
|
|
"epoch": 0.024741043742165337,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000187,
|
|
"loss": 6.8651,
|
|
"mean_token_accuracy": 0.08203414008021355,
|
|
"num_tokens": 850616.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.242887163162232,
|
|
"epoch": 0.025070924325394206,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 6.8928,
|
|
"mean_token_accuracy": 0.082766592502594,
|
|
"num_tokens": 862467.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.330187463760376,
|
|
"epoch": 0.02540080490862308,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.000192,
|
|
"loss": 6.859,
|
|
"mean_token_accuracy": 0.08123024478554726,
|
|
"num_tokens": 872130.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.239017629623413,
|
|
"epoch": 0.02573068549185195,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 6.8431,
|
|
"mean_token_accuracy": 0.09006358608603478,
|
|
"num_tokens": 883598.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.242371702194214,
|
|
"epoch": 0.02606056607508082,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 6.8065,
|
|
"mean_token_accuracy": 0.08576579764485359,
|
|
"num_tokens": 896478.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.231552362442017,
|
|
"epoch": 0.026390446658309693,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 6.8925,
|
|
"mean_token_accuracy": 0.08360796198248863,
|
|
"num_tokens": 909351.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.210038900375366,
|
|
"epoch": 0.026720327241538562,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000202,
|
|
"loss": 6.8603,
|
|
"mean_token_accuracy": 0.08492039665579795,
|
|
"num_tokens": 920170.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.224590682983399,
|
|
"epoch": 0.027050207824767435,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 6.8033,
|
|
"mean_token_accuracy": 0.08973109722137451,
|
|
"num_tokens": 929498.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.183574247360229,
|
|
"epoch": 0.027380088407996304,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000207,
|
|
"loss": 6.8064,
|
|
"mean_token_accuracy": 0.07935241311788559,
|
|
"num_tokens": 939975.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.181256818771362,
|
|
"epoch": 0.027709968991225176,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 6.7794,
|
|
"mean_token_accuracy": 0.08700343444943429,
|
|
"num_tokens": 950593.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.190191602706909,
|
|
"epoch": 0.02803984957445405,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000212,
|
|
"loss": 6.7929,
|
|
"mean_token_accuracy": 0.087624491751194,
|
|
"num_tokens": 962281.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.161286211013794,
|
|
"epoch": 0.028369730157682918,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 6.715,
|
|
"mean_token_accuracy": 0.09188914224505425,
|
|
"num_tokens": 972717.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.127433395385742,
|
|
"epoch": 0.02869961074091179,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 6.7404,
|
|
"mean_token_accuracy": 0.09332360476255416,
|
|
"num_tokens": 983669.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.106704998016357,
|
|
"epoch": 0.02902949132414066,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 6.716,
|
|
"mean_token_accuracy": 0.08401828110218049,
|
|
"num_tokens": 994528.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.022363710403442,
|
|
"epoch": 0.029359371907369532,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000222,
|
|
"loss": 6.7058,
|
|
"mean_token_accuracy": 0.09564009681344032,
|
|
"num_tokens": 1005445.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.108473300933838,
|
|
"epoch": 0.029689252490598405,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 6.8657,
|
|
"mean_token_accuracy": 0.08538904041051865,
|
|
"num_tokens": 1018747.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.169947957992553,
|
|
"epoch": 0.030019133073827274,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 6.7319,
|
|
"mean_token_accuracy": 0.08755266591906548,
|
|
"num_tokens": 1028956.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.097705221176147,
|
|
"epoch": 0.030349013657056147,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 6.9006,
|
|
"mean_token_accuracy": 0.08186126053333283,
|
|
"num_tokens": 1041227.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.106822156906128,
|
|
"epoch": 0.030678894240285016,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 6.7682,
|
|
"mean_token_accuracy": 0.08448890596628189,
|
|
"num_tokens": 1053002.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.122260570526123,
|
|
"epoch": 0.03100877482351389,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.7776,
|
|
"mean_token_accuracy": 0.08819250017404556,
|
|
"num_tokens": 1063693.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.027195596694947,
|
|
"epoch": 0.03133865540674276,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.7407,
|
|
"mean_token_accuracy": 0.09435751885175706,
|
|
"num_tokens": 1074245.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.088400888442993,
|
|
"epoch": 0.03166853598997163,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.7443,
|
|
"mean_token_accuracy": 0.09077147096395492,
|
|
"num_tokens": 1084663.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 6.996756076812744,
|
|
"epoch": 0.0319984165732005,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.6793,
|
|
"mean_token_accuracy": 0.08694756701588631,
|
|
"num_tokens": 1094777.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.0390161037445065,
|
|
"epoch": 0.032328297156429375,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.695,
|
|
"mean_token_accuracy": 0.09284983798861504,
|
|
"num_tokens": 1106712.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 6.954577827453614,
|
|
"epoch": 0.03265817773965824,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.7084,
|
|
"mean_token_accuracy": 0.09444784894585609,
|
|
"num_tokens": 1118488.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.0221700191497805,
|
|
"epoch": 0.032988058322887114,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 6.6757,
|
|
"mean_token_accuracy": 0.08829797506332397,
|
|
"num_tokens": 1128803.0,
|
|
"step": 500
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 4000,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 1,
|
|
"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": 1872696324096000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|