3035 lines
81 KiB
JSON
3035 lines
81 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.09896417496866135,
|
|
"eval_steps": 500,
|
|
"global_step": 1500,
|
|
"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
|
|
},
|
|
{
|
|
"entropy": 7.043442630767823,
|
|
"epoch": 0.033317938906115986,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.6432,
|
|
"mean_token_accuracy": 0.09375235810875893,
|
|
"num_tokens": 1140081.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 6.9111474514007565,
|
|
"epoch": 0.03364781948934486,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.7038,
|
|
"mean_token_accuracy": 0.0864577941596508,
|
|
"num_tokens": 1151740.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.121557140350342,
|
|
"epoch": 0.03397770007257373,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.8204,
|
|
"mean_token_accuracy": 0.08824097402393818,
|
|
"num_tokens": 1163554.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.00194354057312,
|
|
"epoch": 0.0343075806558026,
|
|
"grad_norm": 2.1875,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.6966,
|
|
"mean_token_accuracy": 0.09334781691432,
|
|
"num_tokens": 1174634.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 6.917184782028198,
|
|
"epoch": 0.03463746123903147,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.6518,
|
|
"mean_token_accuracy": 0.0927360899746418,
|
|
"num_tokens": 1186640.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 6.9482104778289795,
|
|
"epoch": 0.03496734182226034,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.5909,
|
|
"mean_token_accuracy": 0.09249644204974175,
|
|
"num_tokens": 1197168.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 6.932004117965699,
|
|
"epoch": 0.035297222405489215,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.6394,
|
|
"mean_token_accuracy": 0.0941959299147129,
|
|
"num_tokens": 1208560.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 6.918610239028931,
|
|
"epoch": 0.03562710298871809,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.6413,
|
|
"mean_token_accuracy": 0.0964904323220253,
|
|
"num_tokens": 1220492.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 6.950107192993164,
|
|
"epoch": 0.03595698357194695,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.6029,
|
|
"mean_token_accuracy": 0.09408045411109925,
|
|
"num_tokens": 1232753.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 6.926637411117554,
|
|
"epoch": 0.036286864155175826,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.7273,
|
|
"mean_token_accuracy": 0.0908595785498619,
|
|
"num_tokens": 1244972.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 6.87931637763977,
|
|
"epoch": 0.0366167447384047,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.6267,
|
|
"mean_token_accuracy": 0.09370757564902306,
|
|
"num_tokens": 1256101.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 6.946982574462891,
|
|
"epoch": 0.03694662532163357,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.668,
|
|
"mean_token_accuracy": 0.09059486165642738,
|
|
"num_tokens": 1268932.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 6.811250972747803,
|
|
"epoch": 0.037276505904862436,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.5818,
|
|
"mean_token_accuracy": 0.0916578710079193,
|
|
"num_tokens": 1278799.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 6.917531156539917,
|
|
"epoch": 0.03760638648809131,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.6272,
|
|
"mean_token_accuracy": 0.10078731775283814,
|
|
"num_tokens": 1288703.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 6.827515745162964,
|
|
"epoch": 0.03793626707132018,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.5747,
|
|
"mean_token_accuracy": 0.09092537686228752,
|
|
"num_tokens": 1300718.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 6.857426309585572,
|
|
"epoch": 0.038266147654549054,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.5513,
|
|
"mean_token_accuracy": 0.09267340302467346,
|
|
"num_tokens": 1311735.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 6.8739509105682375,
|
|
"epoch": 0.03859602823777793,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.5817,
|
|
"mean_token_accuracy": 0.09856103584170342,
|
|
"num_tokens": 1322027.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 6.770535230636597,
|
|
"epoch": 0.03892590882100679,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.5826,
|
|
"mean_token_accuracy": 0.09321667924523354,
|
|
"num_tokens": 1331326.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 6.9302942752838135,
|
|
"epoch": 0.039255789404235665,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.6071,
|
|
"mean_token_accuracy": 0.09590315148234367,
|
|
"num_tokens": 1342064.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 6.839506006240844,
|
|
"epoch": 0.03958566998746454,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.6871,
|
|
"mean_token_accuracy": 0.09009663611650467,
|
|
"num_tokens": 1354257.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 6.88766655921936,
|
|
"epoch": 0.03991555057069341,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.5764,
|
|
"mean_token_accuracy": 0.0959751695394516,
|
|
"num_tokens": 1365757.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 6.829199886322021,
|
|
"epoch": 0.04024543115392228,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.6634,
|
|
"mean_token_accuracy": 0.09242034405469894,
|
|
"num_tokens": 1376655.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 6.909858131408692,
|
|
"epoch": 0.04057531173715115,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.667,
|
|
"mean_token_accuracy": 0.09274478852748871,
|
|
"num_tokens": 1390293.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 6.6624908447265625,
|
|
"epoch": 0.04090519232038002,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.5191,
|
|
"mean_token_accuracy": 0.09633062928915023,
|
|
"num_tokens": 1402595.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 6.9067307472229,
|
|
"epoch": 0.041235072903608894,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.5489,
|
|
"mean_token_accuracy": 0.09810575544834137,
|
|
"num_tokens": 1413574.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 6.7468586444854735,
|
|
"epoch": 0.041564953486837766,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.5601,
|
|
"mean_token_accuracy": 0.0917210377752781,
|
|
"num_tokens": 1425520.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 6.802919054031372,
|
|
"epoch": 0.04189483407006664,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.5286,
|
|
"mean_token_accuracy": 0.09199431240558624,
|
|
"num_tokens": 1436285.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 6.731242513656616,
|
|
"epoch": 0.042224714653295504,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.5278,
|
|
"mean_token_accuracy": 0.10204686671495437,
|
|
"num_tokens": 1447194.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 6.775055360794068,
|
|
"epoch": 0.04255459523652438,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.5432,
|
|
"mean_token_accuracy": 0.10164099037647248,
|
|
"num_tokens": 1459587.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 6.7013648509979244,
|
|
"epoch": 0.04288447581975325,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.6417,
|
|
"mean_token_accuracy": 0.09035239443182945,
|
|
"num_tokens": 1472677.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 6.839371299743652,
|
|
"epoch": 0.04321435640298212,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.5732,
|
|
"mean_token_accuracy": 0.09552592635154725,
|
|
"num_tokens": 1484307.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 6.710844945907593,
|
|
"epoch": 0.043544236986210995,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.5126,
|
|
"mean_token_accuracy": 0.1014582246541977,
|
|
"num_tokens": 1496198.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 6.741171598434448,
|
|
"epoch": 0.04387411756943986,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.4549,
|
|
"mean_token_accuracy": 0.0980884775519371,
|
|
"num_tokens": 1507439.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 6.672511243820191,
|
|
"epoch": 0.04420399815266873,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.4862,
|
|
"mean_token_accuracy": 0.10077084973454475,
|
|
"num_tokens": 1518176.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 6.651635980606079,
|
|
"epoch": 0.044533878735897606,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.4868,
|
|
"mean_token_accuracy": 0.09717255085706711,
|
|
"num_tokens": 1529095.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 6.719832468032837,
|
|
"epoch": 0.04486375931912648,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.476,
|
|
"mean_token_accuracy": 0.09706435576081276,
|
|
"num_tokens": 1539431.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 6.694143629074096,
|
|
"epoch": 0.045193639902355344,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.5542,
|
|
"mean_token_accuracy": 0.09109089970588684,
|
|
"num_tokens": 1550878.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 6.7343847274780275,
|
|
"epoch": 0.04552352048558422,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.4899,
|
|
"mean_token_accuracy": 0.10312912464141846,
|
|
"num_tokens": 1561416.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 6.671954870223999,
|
|
"epoch": 0.04585340106881309,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.5166,
|
|
"mean_token_accuracy": 0.10138269811868668,
|
|
"num_tokens": 1574470.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 6.657031297683716,
|
|
"epoch": 0.04618328165204196,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.4782,
|
|
"mean_token_accuracy": 0.09927325844764709,
|
|
"num_tokens": 1586021.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 6.581523895263672,
|
|
"epoch": 0.046513162235270834,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.4913,
|
|
"mean_token_accuracy": 0.09822574928402901,
|
|
"num_tokens": 1597500.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 6.728161811828613,
|
|
"epoch": 0.0468430428184997,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.5277,
|
|
"mean_token_accuracy": 0.09439488500356674,
|
|
"num_tokens": 1609797.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 6.621676588058472,
|
|
"epoch": 0.04717292340172857,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.4982,
|
|
"mean_token_accuracy": 0.10082960352301598,
|
|
"num_tokens": 1620613.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 6.701622056961059,
|
|
"epoch": 0.047502803984957445,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.4811,
|
|
"mean_token_accuracy": 0.09837961718440055,
|
|
"num_tokens": 1631542.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 6.608582401275635,
|
|
"epoch": 0.04783268456818632,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.393,
|
|
"mean_token_accuracy": 0.10257931649684907,
|
|
"num_tokens": 1641783.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 6.5241758823394775,
|
|
"epoch": 0.04816256515141519,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.5415,
|
|
"mean_token_accuracy": 0.09878621324896812,
|
|
"num_tokens": 1652938.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 6.653652143478394,
|
|
"epoch": 0.048492445734644056,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.3777,
|
|
"mean_token_accuracy": 0.10180097445845604,
|
|
"num_tokens": 1663214.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.677776527404785,
|
|
"epoch": 0.04882232631787293,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.5485,
|
|
"mean_token_accuracy": 0.09322622120380401,
|
|
"num_tokens": 1674845.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 6.615256214141846,
|
|
"epoch": 0.0491522069011018,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.4312,
|
|
"mean_token_accuracy": 0.10337476581335067,
|
|
"num_tokens": 1685222.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.605000305175781,
|
|
"epoch": 0.049482087484330674,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.4928,
|
|
"mean_token_accuracy": 0.09721169024705886,
|
|
"num_tokens": 1696520.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 6.667041349411011,
|
|
"epoch": 0.049811968067559546,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.4516,
|
|
"mean_token_accuracy": 0.1022101253271103,
|
|
"num_tokens": 1706888.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 6.542471027374267,
|
|
"epoch": 0.05014184865078841,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.4465,
|
|
"mean_token_accuracy": 0.09919137805700302,
|
|
"num_tokens": 1718023.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.51314606666565,
|
|
"epoch": 0.050471729234017285,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.3977,
|
|
"mean_token_accuracy": 0.09866592213511467,
|
|
"num_tokens": 1729291.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.608963441848755,
|
|
"epoch": 0.05080160981724616,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.455,
|
|
"mean_token_accuracy": 0.09709095656871795,
|
|
"num_tokens": 1740801.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.562515926361084,
|
|
"epoch": 0.05113149040047503,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.3984,
|
|
"mean_token_accuracy": 0.10630032643675805,
|
|
"num_tokens": 1750933.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.494872283935547,
|
|
"epoch": 0.0514613709837039,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.4371,
|
|
"mean_token_accuracy": 0.09514718353748322,
|
|
"num_tokens": 1761820.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.530175828933716,
|
|
"epoch": 0.05179125156693277,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.3253,
|
|
"mean_token_accuracy": 0.10538506507873535,
|
|
"num_tokens": 1771628.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.5437602519989015,
|
|
"epoch": 0.05212113215016164,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.3548,
|
|
"mean_token_accuracy": 0.10138039365410804,
|
|
"num_tokens": 1783360.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.582814025878906,
|
|
"epoch": 0.05245101273339051,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.4106,
|
|
"mean_token_accuracy": 0.10058085098862649,
|
|
"num_tokens": 1794831.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.483518934249878,
|
|
"epoch": 0.052780893316619386,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.3508,
|
|
"mean_token_accuracy": 0.10989377275109291,
|
|
"num_tokens": 1805056.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.441272878646851,
|
|
"epoch": 0.05311077389984825,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.3004,
|
|
"mean_token_accuracy": 0.10613722503185272,
|
|
"num_tokens": 1815233.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.5487429141998295,
|
|
"epoch": 0.053440654483077124,
|
|
"grad_norm": 1.8828125,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.5721,
|
|
"mean_token_accuracy": 0.09885863438248635,
|
|
"num_tokens": 1826376.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.690734195709228,
|
|
"epoch": 0.053770535066306,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.45,
|
|
"mean_token_accuracy": 0.10282279625535011,
|
|
"num_tokens": 1836326.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.442183876037598,
|
|
"epoch": 0.05410041564953487,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.4297,
|
|
"mean_token_accuracy": 0.10119647011160851,
|
|
"num_tokens": 1847796.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.436309146881103,
|
|
"epoch": 0.05443029623276374,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.3425,
|
|
"mean_token_accuracy": 0.10139305889606476,
|
|
"num_tokens": 1858470.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.510668087005615,
|
|
"epoch": 0.05476017681599261,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.4308,
|
|
"mean_token_accuracy": 0.10197672247886658,
|
|
"num_tokens": 1870368.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.466615819931031,
|
|
"epoch": 0.05509005739922148,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.375,
|
|
"mean_token_accuracy": 0.10953678488731385,
|
|
"num_tokens": 1881841.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.5373430252075195,
|
|
"epoch": 0.05541993798245035,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.3539,
|
|
"mean_token_accuracy": 0.10704940110445023,
|
|
"num_tokens": 1892589.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.391769552230835,
|
|
"epoch": 0.055749818565679225,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.335,
|
|
"mean_token_accuracy": 0.10510399267077446,
|
|
"num_tokens": 1903730.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.534530448913574,
|
|
"epoch": 0.0560796991489081,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.3858,
|
|
"mean_token_accuracy": 0.10953137055039405,
|
|
"num_tokens": 1914872.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.457536935806274,
|
|
"epoch": 0.056409579732136964,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.4245,
|
|
"mean_token_accuracy": 0.10228719264268875,
|
|
"num_tokens": 1926127.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.443365859985351,
|
|
"epoch": 0.056739460315365836,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.4014,
|
|
"mean_token_accuracy": 0.10318510755896568,
|
|
"num_tokens": 1938867.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.4161969184875485,
|
|
"epoch": 0.05706934089859471,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.3378,
|
|
"mean_token_accuracy": 0.10708499625325203,
|
|
"num_tokens": 1950379.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.472681713104248,
|
|
"epoch": 0.05739922148182358,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.3626,
|
|
"mean_token_accuracy": 0.10622433125972748,
|
|
"num_tokens": 1961628.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.411462926864624,
|
|
"epoch": 0.057729102065052454,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.3667,
|
|
"mean_token_accuracy": 0.09961799010634423,
|
|
"num_tokens": 1971580.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.473099327087402,
|
|
"epoch": 0.05805898264828132,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.3758,
|
|
"mean_token_accuracy": 0.09764636233448983,
|
|
"num_tokens": 1981789.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.443149137496948,
|
|
"epoch": 0.05838886323151019,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.2809,
|
|
"mean_token_accuracy": 0.10284367948770523,
|
|
"num_tokens": 1992667.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.299668502807617,
|
|
"epoch": 0.058718743814739065,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.3055,
|
|
"mean_token_accuracy": 0.10764227584004402,
|
|
"num_tokens": 2003864.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.420055198669433,
|
|
"epoch": 0.05904862439796794,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.4087,
|
|
"mean_token_accuracy": 0.09908716157078742,
|
|
"num_tokens": 2017399.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.361506748199463,
|
|
"epoch": 0.05937850498119681,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.3476,
|
|
"mean_token_accuracy": 0.10406243354082108,
|
|
"num_tokens": 2027924.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.435241985321045,
|
|
"epoch": 0.059708385564425676,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.2482,
|
|
"mean_token_accuracy": 0.10816913768649102,
|
|
"num_tokens": 2040174.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.349743604660034,
|
|
"epoch": 0.06003826614765455,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.3419,
|
|
"mean_token_accuracy": 0.10285946726799011,
|
|
"num_tokens": 2052564.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.482759761810303,
|
|
"epoch": 0.06036814673088342,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.3695,
|
|
"mean_token_accuracy": 0.10420166552066804,
|
|
"num_tokens": 2063211.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.366058826446533,
|
|
"epoch": 0.06069802731411229,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.2988,
|
|
"mean_token_accuracy": 0.1029819369316101,
|
|
"num_tokens": 2075434.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.362752103805542,
|
|
"epoch": 0.061027907897341166,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.3144,
|
|
"mean_token_accuracy": 0.10445218682289123,
|
|
"num_tokens": 2087969.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.4367347240448,
|
|
"epoch": 0.06135778848057003,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.3586,
|
|
"mean_token_accuracy": 0.10209745094180107,
|
|
"num_tokens": 2098469.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.3036885261535645,
|
|
"epoch": 0.061687669063798904,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.2488,
|
|
"mean_token_accuracy": 0.11259722784161567,
|
|
"num_tokens": 2108739.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.382773780822754,
|
|
"epoch": 0.06201754964702778,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.3783,
|
|
"mean_token_accuracy": 0.10148499459028244,
|
|
"num_tokens": 2120234.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.356093311309815,
|
|
"epoch": 0.06234743023025665,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.2622,
|
|
"mean_token_accuracy": 0.10476080104708671,
|
|
"num_tokens": 2131585.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.4187659740448,
|
|
"epoch": 0.06267731081348552,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.4137,
|
|
"mean_token_accuracy": 0.10347631350159645,
|
|
"num_tokens": 2144537.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.382198667526245,
|
|
"epoch": 0.06300719139671439,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.371,
|
|
"mean_token_accuracy": 0.10194561555981636,
|
|
"num_tokens": 2157468.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.30818738937378,
|
|
"epoch": 0.06333707197994326,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.29,
|
|
"mean_token_accuracy": 0.10531722754240036,
|
|
"num_tokens": 2169099.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.39258918762207,
|
|
"epoch": 0.06366695256317213,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.2722,
|
|
"mean_token_accuracy": 0.11178096383810043,
|
|
"num_tokens": 2179970.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.194947099685669,
|
|
"epoch": 0.063996833146401,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.3211,
|
|
"mean_token_accuracy": 0.10740408301353455,
|
|
"num_tokens": 2191455.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.5183555603027346,
|
|
"epoch": 0.06432671372962988,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.4575,
|
|
"mean_token_accuracy": 0.09623705670237541,
|
|
"num_tokens": 2202711.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.344358825683594,
|
|
"epoch": 0.06465659431285875,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.2236,
|
|
"mean_token_accuracy": 0.11171902641654015,
|
|
"num_tokens": 2214035.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.262235069274903,
|
|
"epoch": 0.06498647489608762,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.255,
|
|
"mean_token_accuracy": 0.10907828509807586,
|
|
"num_tokens": 2225437.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.36060094833374,
|
|
"epoch": 0.06531635547931648,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.3263,
|
|
"mean_token_accuracy": 0.10713605284690857,
|
|
"num_tokens": 2235693.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.350505733489991,
|
|
"epoch": 0.06564623606254535,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.2613,
|
|
"mean_token_accuracy": 0.1159873902797699,
|
|
"num_tokens": 2245850.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.283651113510132,
|
|
"epoch": 0.06597611664577423,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.3439,
|
|
"mean_token_accuracy": 0.1068663902580738,
|
|
"num_tokens": 2257056.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.351960706710815,
|
|
"epoch": 0.0663059972290031,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000499998026082006,
|
|
"loss": 6.4227,
|
|
"mean_token_accuracy": 0.10501170232892036,
|
|
"num_tokens": 2270421.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.302370309829712,
|
|
"epoch": 0.06663587781223197,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004999900070995136,
|
|
"loss": 6.2511,
|
|
"mean_token_accuracy": 0.10850528180599213,
|
|
"num_tokens": 2282404.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.259668397903442,
|
|
"epoch": 0.06696575839546084,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004999758199023239,
|
|
"loss": 6.1848,
|
|
"mean_token_accuracy": 0.11252183243632316,
|
|
"num_tokens": 2292610.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.295137357711792,
|
|
"epoch": 0.06729563897868972,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999554648793858,
|
|
"loss": 6.2437,
|
|
"mean_token_accuracy": 0.1066038005053997,
|
|
"num_tokens": 2303230.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.2185522556304935,
|
|
"epoch": 0.06762551956191859,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999289425887425,
|
|
"loss": 6.3199,
|
|
"mean_token_accuracy": 0.10505600646138191,
|
|
"num_tokens": 2316522.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.315206575393677,
|
|
"epoch": 0.06795540014514746,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004998962537575161,
|
|
"loss": 6.2276,
|
|
"mean_token_accuracy": 0.10701064616441727,
|
|
"num_tokens": 2327136.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.283873224258423,
|
|
"epoch": 0.06828528072837632,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004998573992818874,
|
|
"loss": 6.2003,
|
|
"mean_token_accuracy": 0.11497593000531196,
|
|
"num_tokens": 2337787.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.218757677078247,
|
|
"epoch": 0.0686151613116052,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004998123802270715,
|
|
"loss": 6.212,
|
|
"mean_token_accuracy": 0.10830829069018363,
|
|
"num_tokens": 2349983.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.272822475433349,
|
|
"epoch": 0.06894504189483407,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004997611978272886,
|
|
"loss": 6.3239,
|
|
"mean_token_accuracy": 0.10436702370643616,
|
|
"num_tokens": 2361673.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.395520210266113,
|
|
"epoch": 0.06927492247806294,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004997038534857298,
|
|
"loss": 6.3214,
|
|
"mean_token_accuracy": 0.11030431911349296,
|
|
"num_tokens": 2373877.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.259383583068848,
|
|
"epoch": 0.06960480306129181,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004996403487745194,
|
|
"loss": 6.1779,
|
|
"mean_token_accuracy": 0.11633929535746575,
|
|
"num_tokens": 2384558.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.288488149642944,
|
|
"epoch": 0.06993468364452068,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000499570685434671,
|
|
"loss": 6.2914,
|
|
"mean_token_accuracy": 0.10511432439088822,
|
|
"num_tokens": 2395333.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.177704286575318,
|
|
"epoch": 0.07026456422774956,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004994948653760405,
|
|
"loss": 6.1901,
|
|
"mean_token_accuracy": 0.10982267633080482,
|
|
"num_tokens": 2407100.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.204628562927246,
|
|
"epoch": 0.07059444481097843,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004994128906772729,
|
|
"loss": 6.14,
|
|
"mean_token_accuracy": 0.11659134700894355,
|
|
"num_tokens": 2418137.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.177124738693237,
|
|
"epoch": 0.0709243253942073,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000499324763585746,
|
|
"loss": 6.2414,
|
|
"mean_token_accuracy": 0.11113048419356346,
|
|
"num_tokens": 2429058.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.23461880683899,
|
|
"epoch": 0.07125420597743617,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004992304865175085,
|
|
"loss": 6.2164,
|
|
"mean_token_accuracy": 0.11510425284504891,
|
|
"num_tokens": 2440716.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.341502380371094,
|
|
"epoch": 0.07158408656066503,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004991300620572138,
|
|
"loss": 6.2522,
|
|
"mean_token_accuracy": 0.1132346160709858,
|
|
"num_tokens": 2451189.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.104463815689087,
|
|
"epoch": 0.0719139671438939,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004990234929580494,
|
|
"loss": 6.2069,
|
|
"mean_token_accuracy": 0.10938044488430024,
|
|
"num_tokens": 2462860.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.285386323928833,
|
|
"epoch": 0.07224384772712278,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004989107821416609,
|
|
"loss": 6.2262,
|
|
"mean_token_accuracy": 0.10970520377159118,
|
|
"num_tokens": 2475555.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.252845001220703,
|
|
"epoch": 0.07257372831035165,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004987919326980723,
|
|
"loss": 6.2574,
|
|
"mean_token_accuracy": 0.10892420709133148,
|
|
"num_tokens": 2485599.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.239787149429321,
|
|
"epoch": 0.07290360889358052,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004986669478856011,
|
|
"loss": 6.1284,
|
|
"mean_token_accuracy": 0.10959701985120773,
|
|
"num_tokens": 2495857.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.162373018264771,
|
|
"epoch": 0.0732334894768094,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004985358311307688,
|
|
"loss": 6.2176,
|
|
"mean_token_accuracy": 0.1139893390238285,
|
|
"num_tokens": 2506330.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.2621618747711185,
|
|
"epoch": 0.07356337006003827,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004983985860282081,
|
|
"loss": 6.1959,
|
|
"mean_token_accuracy": 0.1081675685942173,
|
|
"num_tokens": 2518181.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.195673274993896,
|
|
"epoch": 0.07389325064326714,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004982552163405623,
|
|
"loss": 6.1263,
|
|
"mean_token_accuracy": 0.11639031916856765,
|
|
"num_tokens": 2528346.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.133630275726318,
|
|
"epoch": 0.07422313122649601,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.0004981057259983839,
|
|
"loss": 6.2094,
|
|
"mean_token_accuracy": 0.10735104158520699,
|
|
"num_tokens": 2539922.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.145591926574707,
|
|
"epoch": 0.07455301180972487,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004979501191000262,
|
|
"loss": 6.171,
|
|
"mean_token_accuracy": 0.11293767392635345,
|
|
"num_tokens": 2551145.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.293505001068115,
|
|
"epoch": 0.07488289239295375,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004977883999115311,
|
|
"loss": 6.2636,
|
|
"mean_token_accuracy": 0.10810731053352356,
|
|
"num_tokens": 2562178.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.312147808074951,
|
|
"epoch": 0.07521277297618262,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004976205728665113,
|
|
"loss": 6.3101,
|
|
"mean_token_accuracy": 0.10970097109675407,
|
|
"num_tokens": 2574002.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.1434488773345945,
|
|
"epoch": 0.07554265355941149,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004974466425660307,
|
|
"loss": 6.1977,
|
|
"mean_token_accuracy": 0.11293198987841606,
|
|
"num_tokens": 2584049.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.206194067001343,
|
|
"epoch": 0.07587253414264036,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0004972666137784759,
|
|
"loss": 6.2072,
|
|
"mean_token_accuracy": 0.10776998326182366,
|
|
"num_tokens": 2597415.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.174425745010376,
|
|
"epoch": 0.07620241472586924,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0004970804914394271,
|
|
"loss": 6.151,
|
|
"mean_token_accuracy": 0.1115573450922966,
|
|
"num_tokens": 2609381.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.203009986877442,
|
|
"epoch": 0.07653229530909811,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004968882806515225,
|
|
"loss": 6.2435,
|
|
"mean_token_accuracy": 0.11465907767415047,
|
|
"num_tokens": 2620525.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.169750690460205,
|
|
"epoch": 0.07686217589232698,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004966899866843177,
|
|
"loss": 6.1733,
|
|
"mean_token_accuracy": 0.11209767460823059,
|
|
"num_tokens": 2631978.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.191735363006591,
|
|
"epoch": 0.07719205647555585,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000496485614974142,
|
|
"loss": 6.1439,
|
|
"mean_token_accuracy": 0.11061731800436973,
|
|
"num_tokens": 2643083.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.113736867904663,
|
|
"epoch": 0.07752193705878473,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004962751711239492,
|
|
"loss": 6.1356,
|
|
"mean_token_accuracy": 0.11404857262969018,
|
|
"num_tokens": 2654175.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.246322917938232,
|
|
"epoch": 0.07785181764201358,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004960586609031636,
|
|
"loss": 6.1139,
|
|
"mean_token_accuracy": 0.11253655329346657,
|
|
"num_tokens": 2665513.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.023867750167847,
|
|
"epoch": 0.07818169822524246,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004958360902475224,
|
|
"loss": 6.1177,
|
|
"mean_token_accuracy": 0.11768655627965927,
|
|
"num_tokens": 2675842.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.269159746170044,
|
|
"epoch": 0.07851157880847133,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004956074652589125,
|
|
"loss": 6.2737,
|
|
"mean_token_accuracy": 0.1111219845712185,
|
|
"num_tokens": 2687289.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.171760988235474,
|
|
"epoch": 0.0788414593917002,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004953727922052035,
|
|
"loss": 6.1664,
|
|
"mean_token_accuracy": 0.10972021371126175,
|
|
"num_tokens": 2698409.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.155753421783447,
|
|
"epoch": 0.07917133997492908,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004951320775200756,
|
|
"loss": 6.1081,
|
|
"mean_token_accuracy": 0.12315625622868538,
|
|
"num_tokens": 2708695.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.189047718048096,
|
|
"epoch": 0.07950122055815795,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004948853278028436,
|
|
"loss": 6.1013,
|
|
"mean_token_accuracy": 0.12293576672673226,
|
|
"num_tokens": 2719631.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.186757516860962,
|
|
"epoch": 0.07983110114138682,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004946325498182755,
|
|
"loss": 6.1987,
|
|
"mean_token_accuracy": 0.11486271470785141,
|
|
"num_tokens": 2730070.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.120721006393433,
|
|
"epoch": 0.08016098172461569,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004943737504964076,
|
|
"loss": 6.1881,
|
|
"mean_token_accuracy": 0.10870257839560508,
|
|
"num_tokens": 2741450.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.198238134384155,
|
|
"epoch": 0.08049086230784457,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000494108936932354,
|
|
"loss": 6.1063,
|
|
"mean_token_accuracy": 0.12308499589562416,
|
|
"num_tokens": 2752108.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.0927952289581295,
|
|
"epoch": 0.08082074289107344,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004938381163861124,
|
|
"loss": 6.2003,
|
|
"mean_token_accuracy": 0.11003261134028434,
|
|
"num_tokens": 2763041.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.195061349868775,
|
|
"epoch": 0.0811506234743023,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004935612962823645,
|
|
"loss": 6.1448,
|
|
"mean_token_accuracy": 0.11436575651168823,
|
|
"num_tokens": 2773140.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.0863447189331055,
|
|
"epoch": 0.08148050405753117,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004932784842102739,
|
|
"loss": 6.1032,
|
|
"mean_token_accuracy": 0.11739343777298927,
|
|
"num_tokens": 2783811.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.175919055938721,
|
|
"epoch": 0.08181038464076004,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004929896879232758,
|
|
"loss": 6.0792,
|
|
"mean_token_accuracy": 0.11807719096541405,
|
|
"num_tokens": 2793957.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.133573293685913,
|
|
"epoch": 0.08214026522398891,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004926949153388668,
|
|
"loss": 6.1854,
|
|
"mean_token_accuracy": 0.11286567747592927,
|
|
"num_tokens": 2804750.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.153765344619751,
|
|
"epoch": 0.08247014580721779,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004923941745383859,
|
|
"loss": 6.2111,
|
|
"mean_token_accuracy": 0.11382120177149772,
|
|
"num_tokens": 2815559.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.113592338562012,
|
|
"epoch": 0.08280002639044666,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000492087473766794,
|
|
"loss": 6.0241,
|
|
"mean_token_accuracy": 0.12170703038573265,
|
|
"num_tokens": 2826450.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.032780885696411,
|
|
"epoch": 0.08312990697367553,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.000491774821432448,
|
|
"loss": 6.2134,
|
|
"mean_token_accuracy": 0.10796439573168755,
|
|
"num_tokens": 2837471.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.300673246383667,
|
|
"epoch": 0.0834597875569044,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004914562261068693,
|
|
"loss": 6.2415,
|
|
"mean_token_accuracy": 0.1093483418226242,
|
|
"num_tokens": 2849110.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.121099233627319,
|
|
"epoch": 0.08378966814013328,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004911316965245098,
|
|
"loss": 6.0804,
|
|
"mean_token_accuracy": 0.11565342396497727,
|
|
"num_tokens": 2860675.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.113272762298584,
|
|
"epoch": 0.08411954872336214,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.000490801241582512,
|
|
"loss": 6.1674,
|
|
"mean_token_accuracy": 0.1166732706129551,
|
|
"num_tokens": 2872482.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.1046217918396,
|
|
"epoch": 0.08444942930659101,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000490464870340465,
|
|
"loss": 6.0644,
|
|
"mean_token_accuracy": 0.11849834322929383,
|
|
"num_tokens": 2882199.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.150085544586181,
|
|
"epoch": 0.08477930988981988,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004901225920201563,
|
|
"loss": 6.1013,
|
|
"mean_token_accuracy": 0.11324956640601158,
|
|
"num_tokens": 2892581.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.132313966751099,
|
|
"epoch": 0.08510919047304875,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.000489774416005319,
|
|
"loss": 6.1362,
|
|
"mean_token_accuracy": 0.11294598430395127,
|
|
"num_tokens": 2904105.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.080805683135987,
|
|
"epoch": 0.08543907105627763,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004894203518413742,
|
|
"loss": 6.0695,
|
|
"mean_token_accuracy": 0.11511229053139686,
|
|
"num_tokens": 2914818.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.121900367736816,
|
|
"epoch": 0.0857689516395065,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004890604092351701,
|
|
"loss": 6.0744,
|
|
"mean_token_accuracy": 0.11540318131446839,
|
|
"num_tokens": 2925904.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.137706279754639,
|
|
"epoch": 0.08609883222273537,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.000488694598054715,
|
|
"loss": 6.1075,
|
|
"mean_token_accuracy": 0.11962982565164566,
|
|
"num_tokens": 2937177.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.082860279083252,
|
|
"epoch": 0.08642871280596424,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004883229283289071,
|
|
"loss": 6.1242,
|
|
"mean_token_accuracy": 0.11809448972344398,
|
|
"num_tokens": 2948641.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.081214570999146,
|
|
"epoch": 0.08675859338919312,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00048794541024725993,
|
|
"loss": 6.1467,
|
|
"mean_token_accuracy": 0.11472822427749634,
|
|
"num_tokens": 2960308.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.1382523536682125,
|
|
"epoch": 0.08708847397242199,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.0004875620541596221,
|
|
"loss": 6.1653,
|
|
"mean_token_accuracy": 0.1209314726293087,
|
|
"num_tokens": 2972666.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.159471464157105,
|
|
"epoch": 0.08741835455565085,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00048717287057589454,
|
|
"loss": 6.1412,
|
|
"mean_token_accuracy": 0.11160789728164673,
|
|
"num_tokens": 2982894.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.114540338516235,
|
|
"epoch": 0.08774823513887972,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004867778701657417,
|
|
"loss": 6.0599,
|
|
"mean_token_accuracy": 0.1218624509871006,
|
|
"num_tokens": 2992617.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.065075063705445,
|
|
"epoch": 0.0880781157221086,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00048637706375829955,
|
|
"loss": 6.061,
|
|
"mean_token_accuracy": 0.11655306667089463,
|
|
"num_tokens": 3003439.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.163606309890747,
|
|
"epoch": 0.08840799630533747,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.000485970462341878,
|
|
"loss": 6.2409,
|
|
"mean_token_accuracy": 0.1089333713054657,
|
|
"num_tokens": 3015204.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.185383558273315,
|
|
"epoch": 0.08873787688856634,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00048555807706366044,
|
|
"loss": 6.1638,
|
|
"mean_token_accuracy": 0.12425324842333793,
|
|
"num_tokens": 3026624.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.09044451713562,
|
|
"epoch": 0.08906775747179521,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00048513991922939756,
|
|
"loss": 6.0287,
|
|
"mean_token_accuracy": 0.11683051139116288,
|
|
"num_tokens": 3037828.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.085692453384399,
|
|
"epoch": 0.08939763805502408,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.00048471600030309744,
|
|
"loss": 6.0342,
|
|
"mean_token_accuracy": 0.11773991361260414,
|
|
"num_tokens": 3049666.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.058322858810425,
|
|
"epoch": 0.08972751863825296,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00048428633190671186,
|
|
"loss": 6.0978,
|
|
"mean_token_accuracy": 0.12333397567272186,
|
|
"num_tokens": 3061090.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.077864074707032,
|
|
"epoch": 0.09005739922148183,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004838509258198167,
|
|
"loss": 6.1162,
|
|
"mean_token_accuracy": 0.11454065665602683,
|
|
"num_tokens": 3073031.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.029748201370239,
|
|
"epoch": 0.09038727980471069,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.00048340979397929,
|
|
"loss": 6.0576,
|
|
"mean_token_accuracy": 0.11675121933221817,
|
|
"num_tokens": 3084858.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.226750135421753,
|
|
"epoch": 0.09071716038793956,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00048296294847898386,
|
|
"loss": 6.1121,
|
|
"mean_token_accuracy": 0.11243180558085442,
|
|
"num_tokens": 3095554.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.041839742660523,
|
|
"epoch": 0.09104704097116843,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004825104015693934,
|
|
"loss": 6.057,
|
|
"mean_token_accuracy": 0.12022657096385955,
|
|
"num_tokens": 3107164.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.078349208831787,
|
|
"epoch": 0.0913769215543973,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004820521656573208,
|
|
"loss": 6.1492,
|
|
"mean_token_accuracy": 0.11309162974357605,
|
|
"num_tokens": 3118681.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.191424703598022,
|
|
"epoch": 0.09170680213762618,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00048158825330553505,
|
|
"loss": 6.1243,
|
|
"mean_token_accuracy": 0.11376700103282929,
|
|
"num_tokens": 3130604.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.090463495254516,
|
|
"epoch": 0.09203668272085505,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00048111867723242763,
|
|
"loss": 6.1028,
|
|
"mean_token_accuracy": 0.1133815087378025,
|
|
"num_tokens": 3141774.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.073332262039185,
|
|
"epoch": 0.09236656330408392,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004806434503116637,
|
|
"loss": 6.0008,
|
|
"mean_token_accuracy": 0.12441991344094276,
|
|
"num_tokens": 3151693.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.050995111465454,
|
|
"epoch": 0.0926964438873128,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004801625855718296,
|
|
"loss": 5.9659,
|
|
"mean_token_accuracy": 0.1263166770339012,
|
|
"num_tokens": 3162832.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 5.999355936050415,
|
|
"epoch": 0.09302632447054167,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.00047967609619607477,
|
|
"loss": 6.0237,
|
|
"mean_token_accuracy": 0.12891323044896125,
|
|
"num_tokens": 3174048.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 5.998584127426147,
|
|
"epoch": 0.09335620505377054,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004791839955217513,
|
|
"loss": 6.009,
|
|
"mean_token_accuracy": 0.12211860343813896,
|
|
"num_tokens": 3185451.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.027903842926025,
|
|
"epoch": 0.0936860856369994,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.00047868629704004786,
|
|
"loss": 6.0876,
|
|
"mean_token_accuracy": 0.11863903552293778,
|
|
"num_tokens": 3197405.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.116988611221314,
|
|
"epoch": 0.09401596622022827,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00047818301439561965,
|
|
"loss": 6.0667,
|
|
"mean_token_accuracy": 0.12125076875090599,
|
|
"num_tokens": 3208421.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.021233987808228,
|
|
"epoch": 0.09434584680345715,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00047767416138621454,
|
|
"loss": 5.9726,
|
|
"mean_token_accuracy": 0.1186994805932045,
|
|
"num_tokens": 3219745.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.0462109565734865,
|
|
"epoch": 0.09467572738668602,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.000477159751962295,
|
|
"loss": 6.0507,
|
|
"mean_token_accuracy": 0.12183702439069748,
|
|
"num_tokens": 3231347.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.012817668914795,
|
|
"epoch": 0.09500560796991489,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00047663980022665507,
|
|
"loss": 6.0415,
|
|
"mean_token_accuracy": 0.11305382400751114,
|
|
"num_tokens": 3242616.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 5.987714004516602,
|
|
"epoch": 0.09533548855314376,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.00047611432043403437,
|
|
"loss": 5.928,
|
|
"mean_token_accuracy": 0.12776491940021514,
|
|
"num_tokens": 3253697.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.081219625473023,
|
|
"epoch": 0.09566536913637264,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004755833269907267,
|
|
"loss": 6.0543,
|
|
"mean_token_accuracy": 0.11684809550642968,
|
|
"num_tokens": 3263800.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.064343976974487,
|
|
"epoch": 0.09599524971960151,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004750468344541857,
|
|
"loss": 6.0013,
|
|
"mean_token_accuracy": 0.12487756982445716,
|
|
"num_tokens": 3274538.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.032814645767212,
|
|
"epoch": 0.09632513030283038,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00047450485753262525,
|
|
"loss": 5.9984,
|
|
"mean_token_accuracy": 0.12270476445555686,
|
|
"num_tokens": 3285719.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.063799333572388,
|
|
"epoch": 0.09665501088605925,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00047395741108461633,
|
|
"loss": 6.1269,
|
|
"mean_token_accuracy": 0.11172792389988899,
|
|
"num_tokens": 3296715.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.063994312286377,
|
|
"epoch": 0.09698489146928811,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.00047340451011867985,
|
|
"loss": 6.0587,
|
|
"mean_token_accuracy": 0.12208845466375351,
|
|
"num_tokens": 3308800.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.093899726867676,
|
|
"epoch": 0.09731477205251698,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00047284616979287515,
|
|
"loss": 6.1951,
|
|
"mean_token_accuracy": 0.11882652938365937,
|
|
"num_tokens": 3322180.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.066127347946167,
|
|
"epoch": 0.09764465263574586,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00047228240541438433,
|
|
"loss": 5.9834,
|
|
"mean_token_accuracy": 0.11412314474582672,
|
|
"num_tokens": 3333363.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.071869516372681,
|
|
"epoch": 0.09797453321897473,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00047171323243909257,
|
|
"loss": 6.0559,
|
|
"mean_token_accuracy": 0.12139012813568115,
|
|
"num_tokens": 3345040.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.101380777359009,
|
|
"epoch": 0.0983044138022036,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00047113866647116457,
|
|
"loss": 6.1356,
|
|
"mean_token_accuracy": 0.12105778455734253,
|
|
"num_tokens": 3356445.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.0262469291687015,
|
|
"epoch": 0.09863429438543247,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004705587232626164,
|
|
"loss": 5.9731,
|
|
"mean_token_accuracy": 0.12452970519661903,
|
|
"num_tokens": 3366948.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.009040498733521,
|
|
"epoch": 0.09896417496866135,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.00046997341871288424,
|
|
"loss": 6.0541,
|
|
"mean_token_accuracy": 0.12063171342015266,
|
|
"num_tokens": 3378616.0,
|
|
"step": 1500
|
|
}
|
|
],
|
|
"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": 5632370214912000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|