6035 lines
164 KiB
JSON
6035 lines
164 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.1979283499373227,
|
|
"eval_steps": 500,
|
|
"global_step": 3000,
|
|
"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
|
|
},
|
|
{
|
|
"entropy": 5.933080196380615,
|
|
"epoch": 0.09929405555189022,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004693827688683879,
|
|
"loss": 5.9274,
|
|
"mean_token_accuracy": 0.12509941309690475,
|
|
"num_tokens": 3390384.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 6.068911933898926,
|
|
"epoch": 0.09962393613511909,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004687867899220914,
|
|
"loss": 6.0006,
|
|
"mean_token_accuracy": 0.12345171794295311,
|
|
"num_tokens": 3401904.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 5.974150466918945,
|
|
"epoch": 0.09995381671834795,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.00046818549821305846,
|
|
"loss": 6.0096,
|
|
"mean_token_accuracy": 0.12344863340258598,
|
|
"num_tokens": 3413691.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 5.966073513031006,
|
|
"epoch": 0.10028369730157682,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00046757891022600494,
|
|
"loss": 5.9222,
|
|
"mean_token_accuracy": 0.12320040836930275,
|
|
"num_tokens": 3423751.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 6.082713508605957,
|
|
"epoch": 0.1006135778848057,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004669670425908471,
|
|
"loss": 6.1381,
|
|
"mean_token_accuracy": 0.11803044751286507,
|
|
"num_tokens": 3436597.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 6.035796499252319,
|
|
"epoch": 0.10094345846803457,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004663499120822451,
|
|
"loss": 6.0489,
|
|
"mean_token_accuracy": 0.12692047506570817,
|
|
"num_tokens": 3447705.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 6.11446590423584,
|
|
"epoch": 0.10127333905126344,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004657275356191437,
|
|
"loss": 6.0229,
|
|
"mean_token_accuracy": 0.12199754565954209,
|
|
"num_tokens": 3458643.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 5.959643793106079,
|
|
"epoch": 0.10160321963449231,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00046509993026430804,
|
|
"loss": 6.0698,
|
|
"mean_token_accuracy": 0.12218033298850059,
|
|
"num_tokens": 3469050.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 6.055213165283203,
|
|
"epoch": 0.10193310021772119,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004644671132238558,
|
|
"loss": 6.0222,
|
|
"mean_token_accuracy": 0.12194224745035172,
|
|
"num_tokens": 3480379.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 6.051732397079467,
|
|
"epoch": 0.10226298080095006,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.00046382910184678585,
|
|
"loss": 6.0365,
|
|
"mean_token_accuracy": 0.12287314757704734,
|
|
"num_tokens": 3492678.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 6.024369382858277,
|
|
"epoch": 0.10259286138417893,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004631859136245025,
|
|
"loss": 6.0129,
|
|
"mean_token_accuracy": 0.11959246546030045,
|
|
"num_tokens": 3503932.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 6.090898323059082,
|
|
"epoch": 0.1029227419674078,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004625375661903357,
|
|
"loss": 6.0408,
|
|
"mean_token_accuracy": 0.12016954272985458,
|
|
"num_tokens": 3517102.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 6.045927381515503,
|
|
"epoch": 0.10325262255063666,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00046188407731905787,
|
|
"loss": 5.9874,
|
|
"mean_token_accuracy": 0.122080397605896,
|
|
"num_tokens": 3530101.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 5.960421991348267,
|
|
"epoch": 0.10358250313386554,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00046122546492639643,
|
|
"loss": 5.9449,
|
|
"mean_token_accuracy": 0.12861818596720695,
|
|
"num_tokens": 3540631.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 6.0487302303314205,
|
|
"epoch": 0.10391238371709441,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.000460561747068543,
|
|
"loss": 6.0737,
|
|
"mean_token_accuracy": 0.12660795226693153,
|
|
"num_tokens": 3552730.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 6.045223665237427,
|
|
"epoch": 0.10424226430032328,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004598929419416578,
|
|
"loss": 5.9982,
|
|
"mean_token_accuracy": 0.12181697189807891,
|
|
"num_tokens": 3562997.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 5.97337703704834,
|
|
"epoch": 0.10457214488355215,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00045921906788137123,
|
|
"loss": 5.8971,
|
|
"mean_token_accuracy": 0.1267779976129532,
|
|
"num_tokens": 3574946.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 6.0198619842529295,
|
|
"epoch": 0.10490202546678103,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00045854014336228115,
|
|
"loss": 6.0155,
|
|
"mean_token_accuracy": 0.12620965838432313,
|
|
"num_tokens": 3586922.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 6.0225035667419435,
|
|
"epoch": 0.1052319060500099,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00045785618699744615,
|
|
"loss": 6.0882,
|
|
"mean_token_accuracy": 0.11795857101678849,
|
|
"num_tokens": 3598824.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 6.050491809844971,
|
|
"epoch": 0.10556178663323877,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.00045716721753787543,
|
|
"loss": 5.9788,
|
|
"mean_token_accuracy": 0.12132218405604363,
|
|
"num_tokens": 3610814.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 6.003845500946045,
|
|
"epoch": 0.10589166721646764,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004564732538720148,
|
|
"loss": 5.9889,
|
|
"mean_token_accuracy": 0.12396189272403717,
|
|
"num_tokens": 3622284.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 5.9358744621276855,
|
|
"epoch": 0.1062215477996965,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00045577431502522877,
|
|
"loss": 5.9027,
|
|
"mean_token_accuracy": 0.1252452127635479,
|
|
"num_tokens": 3633760.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 6.027674341201783,
|
|
"epoch": 0.10655142838292538,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004550704201592787,
|
|
"loss": 6.004,
|
|
"mean_token_accuracy": 0.12208143025636672,
|
|
"num_tokens": 3644667.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 5.986632013320923,
|
|
"epoch": 0.10688130896615425,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004543615885717981,
|
|
"loss": 5.9566,
|
|
"mean_token_accuracy": 0.12380741238594055,
|
|
"num_tokens": 3655375.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 5.993251085281372,
|
|
"epoch": 0.10721118954938312,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00045364783969576296,
|
|
"loss": 5.9549,
|
|
"mean_token_accuracy": 0.12169160172343255,
|
|
"num_tokens": 3666911.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 5.967620801925659,
|
|
"epoch": 0.107541070132612,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004529291930989592,
|
|
"loss": 5.9813,
|
|
"mean_token_accuracy": 0.12813740000128745,
|
|
"num_tokens": 3678251.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 5.956898212432861,
|
|
"epoch": 0.10787095071584087,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004522056684834464,
|
|
"loss": 5.9591,
|
|
"mean_token_accuracy": 0.12339179217815399,
|
|
"num_tokens": 3690099.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 6.012165117263794,
|
|
"epoch": 0.10820083129906974,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004514772856850173,
|
|
"loss": 6.0109,
|
|
"mean_token_accuracy": 0.12273171097040177,
|
|
"num_tokens": 3701856.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 6.066247272491455,
|
|
"epoch": 0.10853071188229861,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004507440646726542,
|
|
"loss": 6.1265,
|
|
"mean_token_accuracy": 0.11398563086986542,
|
|
"num_tokens": 3713985.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 5.980256414413452,
|
|
"epoch": 0.10886059246552748,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004500060255479818,
|
|
"loss": 6.0429,
|
|
"mean_token_accuracy": 0.11961570456624031,
|
|
"num_tokens": 3726176.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 5.99865756034851,
|
|
"epoch": 0.10919047304875636,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004492631885447151,
|
|
"loss": 5.8358,
|
|
"mean_token_accuracy": 0.13007402122020723,
|
|
"num_tokens": 3737305.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 5.956647253036499,
|
|
"epoch": 0.10952035363198522,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00044851557402810616,
|
|
"loss": 6.0085,
|
|
"mean_token_accuracy": 0.12054088041186332,
|
|
"num_tokens": 3748282.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 5.983574104309082,
|
|
"epoch": 0.10985023421521409,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00044776320249438444,
|
|
"loss": 5.8728,
|
|
"mean_token_accuracy": 0.13192812204360962,
|
|
"num_tokens": 3759035.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 5.955325889587402,
|
|
"epoch": 0.11018011479844296,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00044700609457019565,
|
|
"loss": 5.8909,
|
|
"mean_token_accuracy": 0.12752507999539375,
|
|
"num_tokens": 3769833.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 5.95144214630127,
|
|
"epoch": 0.11050999538167183,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004462442710120359,
|
|
"loss": 6.0295,
|
|
"mean_token_accuracy": 0.12280331328511238,
|
|
"num_tokens": 3781424.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 6.02058777809143,
|
|
"epoch": 0.1108398759649007,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000445477752705683,
|
|
"loss": 5.9871,
|
|
"mean_token_accuracy": 0.12472627833485603,
|
|
"num_tokens": 3792697.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 5.9440632343292235,
|
|
"epoch": 0.11116975654812958,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00044470656066562336,
|
|
"loss": 5.8724,
|
|
"mean_token_accuracy": 0.12543088048696518,
|
|
"num_tokens": 3804250.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 5.912532615661621,
|
|
"epoch": 0.11149963713135845,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004439307160344765,
|
|
"loss": 5.9649,
|
|
"mean_token_accuracy": 0.11689893901348114,
|
|
"num_tokens": 3816723.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 5.96519250869751,
|
|
"epoch": 0.11182951771458732,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00044315024008241473,
|
|
"loss": 5.8722,
|
|
"mean_token_accuracy": 0.1270854040980339,
|
|
"num_tokens": 3827765.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 5.908421611785888,
|
|
"epoch": 0.1121593982978162,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004423651542065806,
|
|
"loss": 5.9252,
|
|
"mean_token_accuracy": 0.11848557814955711,
|
|
"num_tokens": 3839261.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 5.968927717208862,
|
|
"epoch": 0.11248927888104507,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00044157547993050006,
|
|
"loss": 5.98,
|
|
"mean_token_accuracy": 0.12133546769618989,
|
|
"num_tokens": 3851532.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 5.967506790161133,
|
|
"epoch": 0.11281915946427393,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00044078123890349227,
|
|
"loss": 5.9523,
|
|
"mean_token_accuracy": 0.12052992954850197,
|
|
"num_tokens": 3863412.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 5.95634241104126,
|
|
"epoch": 0.1131490400475028,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.00043998245290007606,
|
|
"loss": 5.8327,
|
|
"mean_token_accuracy": 0.125997843593359,
|
|
"num_tokens": 3873660.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 5.957879590988159,
|
|
"epoch": 0.11347892063073167,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00043917914381937323,
|
|
"loss": 5.9492,
|
|
"mean_token_accuracy": 0.12157313376665116,
|
|
"num_tokens": 3884998.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 5.936851501464844,
|
|
"epoch": 0.11380880121396054,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00043837133368450815,
|
|
"loss": 5.9262,
|
|
"mean_token_accuracy": 0.1275871478021145,
|
|
"num_tokens": 3896449.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 5.95711498260498,
|
|
"epoch": 0.11413868179718942,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004375590446420037,
|
|
"loss": 5.9251,
|
|
"mean_token_accuracy": 0.1278728500008583,
|
|
"num_tokens": 3907446.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 5.946895170211792,
|
|
"epoch": 0.11446856238041829,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004367422989611743,
|
|
"loss": 5.8626,
|
|
"mean_token_accuracy": 0.1307248942553997,
|
|
"num_tokens": 3919132.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 5.9647520065307615,
|
|
"epoch": 0.11479844296364716,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004359211190335153,
|
|
"loss": 6.0379,
|
|
"mean_token_accuracy": 0.12036628425121307,
|
|
"num_tokens": 3930929.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 5.957888221740722,
|
|
"epoch": 0.11512832354687604,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00043509552737208923,
|
|
"loss": 6.0208,
|
|
"mean_token_accuracy": 0.12099091112613677,
|
|
"num_tokens": 3943358.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 6.042073392868042,
|
|
"epoch": 0.11545820413010491,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00043426554661090853,
|
|
"loss": 6.0229,
|
|
"mean_token_accuracy": 0.12397746220231057,
|
|
"num_tokens": 3953642.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 6.116046810150147,
|
|
"epoch": 0.11578808471333377,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00043343119950431516,
|
|
"loss": 6.0315,
|
|
"mean_token_accuracy": 0.1192188322544098,
|
|
"num_tokens": 3965905.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 5.889814186096191,
|
|
"epoch": 0.11611796529656264,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.00043259250892635644,
|
|
"loss": 5.9519,
|
|
"mean_token_accuracy": 0.12521821185946463,
|
|
"num_tokens": 3977729.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 6.058498859405518,
|
|
"epoch": 0.11644784587979151,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004317494978701582,
|
|
"loss": 5.9488,
|
|
"mean_token_accuracy": 0.13034996911883354,
|
|
"num_tokens": 3989346.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 5.918118715286255,
|
|
"epoch": 0.11677772646302038,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004309021894472943,
|
|
"loss": 6.0058,
|
|
"mean_token_accuracy": 0.11568161398172379,
|
|
"num_tokens": 4000997.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 6.006221055984497,
|
|
"epoch": 0.11710760704624926,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004300506068871534,
|
|
"loss": 5.8804,
|
|
"mean_token_accuracy": 0.12805683612823487,
|
|
"num_tokens": 4012087.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 5.952947616577148,
|
|
"epoch": 0.11743748762947813,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00042919477353630135,
|
|
"loss": 5.8777,
|
|
"mean_token_accuracy": 0.13223015442490577,
|
|
"num_tokens": 4021877.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 5.846690368652344,
|
|
"epoch": 0.117767368212707,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000428334712857842,
|
|
"loss": 5.8573,
|
|
"mean_token_accuracy": 0.1268261454999447,
|
|
"num_tokens": 4032590.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 5.971124649047852,
|
|
"epoch": 0.11809724879593587,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00042747044843077304,
|
|
"loss": 5.8876,
|
|
"mean_token_accuracy": 0.12811336368322374,
|
|
"num_tokens": 4042888.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 5.887255096435547,
|
|
"epoch": 0.11842712937916475,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00042660200394934047,
|
|
"loss": 5.8451,
|
|
"mean_token_accuracy": 0.13152144849300385,
|
|
"num_tokens": 4053081.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 5.909045171737671,
|
|
"epoch": 0.11875700996239362,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00042572940322238844,
|
|
"loss": 5.9288,
|
|
"mean_token_accuracy": 0.12463091686367989,
|
|
"num_tokens": 4063683.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 5.900584173202515,
|
|
"epoch": 0.11908689054562248,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00042485267017270664,
|
|
"loss": 5.8388,
|
|
"mean_token_accuracy": 0.13147015124559402,
|
|
"num_tokens": 4074502.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 5.893744421005249,
|
|
"epoch": 0.11941677112885135,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004239718288363745,
|
|
"loss": 5.8296,
|
|
"mean_token_accuracy": 0.12867452651262284,
|
|
"num_tokens": 4085866.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 5.848910236358643,
|
|
"epoch": 0.11974665171208022,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004230869033621023,
|
|
"loss": 5.915,
|
|
"mean_token_accuracy": 0.13088555932044982,
|
|
"num_tokens": 4098157.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 5.933797788619995,
|
|
"epoch": 0.1200765322953091,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004221979180105688,
|
|
"loss": 5.8331,
|
|
"mean_token_accuracy": 0.13074176236987115,
|
|
"num_tokens": 4108922.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 5.856889867782593,
|
|
"epoch": 0.12040641287853797,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00042130489715375645,
|
|
"loss": 5.8507,
|
|
"mean_token_accuracy": 0.12500329464673995,
|
|
"num_tokens": 4120240.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 5.87729320526123,
|
|
"epoch": 0.12073629346176684,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00042040786527428335,
|
|
"loss": 5.7215,
|
|
"mean_token_accuracy": 0.14235641956329345,
|
|
"num_tokens": 4130766.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 5.880376148223877,
|
|
"epoch": 0.12106617404499571,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004195068469647315,
|
|
"loss": 5.8629,
|
|
"mean_token_accuracy": 0.13628575429320336,
|
|
"num_tokens": 4140371.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 5.835314035415649,
|
|
"epoch": 0.12139605462822459,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.00041860186692697297,
|
|
"loss": 5.7926,
|
|
"mean_token_accuracy": 0.13235599920153618,
|
|
"num_tokens": 4151875.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 5.893461799621582,
|
|
"epoch": 0.12172593521145346,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00041769294997149264,
|
|
"loss": 5.8513,
|
|
"mean_token_accuracy": 0.13282495588064194,
|
|
"num_tokens": 4162941.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 5.942033529281616,
|
|
"epoch": 0.12205581579468233,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004167801210167081,
|
|
"loss": 5.8718,
|
|
"mean_token_accuracy": 0.13093628585338593,
|
|
"num_tokens": 4174770.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 5.902879095077514,
|
|
"epoch": 0.12238569637791119,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004158634050882861,
|
|
"loss": 5.9094,
|
|
"mean_token_accuracy": 0.1262867718935013,
|
|
"num_tokens": 4185169.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 5.889041566848755,
|
|
"epoch": 0.12271557696114006,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004149428273184569,
|
|
"loss": 5.8873,
|
|
"mean_token_accuracy": 0.1260581724345684,
|
|
"num_tokens": 4196840.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 5.957019853591919,
|
|
"epoch": 0.12304545754436894,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004140184129453253,
|
|
"loss": 5.8583,
|
|
"mean_token_accuracy": 0.1328165903687477,
|
|
"num_tokens": 4208304.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 5.89555082321167,
|
|
"epoch": 0.12337533812759781,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000413090187312178,
|
|
"loss": 5.8939,
|
|
"mean_token_accuracy": 0.1306886725127697,
|
|
"num_tokens": 4219591.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 5.928309679031372,
|
|
"epoch": 0.12370521871082668,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004121581758667898,
|
|
"loss": 5.9343,
|
|
"mean_token_accuracy": 0.12753283083438874,
|
|
"num_tokens": 4232018.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 5.951842832565307,
|
|
"epoch": 0.12403509929405555,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.00041122240416072533,
|
|
"loss": 5.8556,
|
|
"mean_token_accuracy": 0.1301908016204834,
|
|
"num_tokens": 4243933.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 5.8850939750671385,
|
|
"epoch": 0.12436497987728443,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004102828978486385,
|
|
"loss": 5.8534,
|
|
"mean_token_accuracy": 0.12924208343029023,
|
|
"num_tokens": 4254471.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 5.929201555252075,
|
|
"epoch": 0.1246948604605133,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004093396826875695,
|
|
"loss": 5.9326,
|
|
"mean_token_accuracy": 0.12578677833080293,
|
|
"num_tokens": 4266377.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 5.926929473876953,
|
|
"epoch": 0.12502474104374217,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.00040839278453623837,
|
|
"loss": 5.8102,
|
|
"mean_token_accuracy": 0.12675564959645272,
|
|
"num_tokens": 4277972.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 5.861911725997925,
|
|
"epoch": 0.12535462162697103,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004074422293543363,
|
|
"loss": 5.9103,
|
|
"mean_token_accuracy": 0.1278995081782341,
|
|
"num_tokens": 4289262.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 5.957873392105102,
|
|
"epoch": 0.12568450221019992,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004064880432018137,
|
|
"loss": 5.8715,
|
|
"mean_token_accuracy": 0.12517192736268043,
|
|
"num_tokens": 4299416.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 5.871857595443726,
|
|
"epoch": 0.12601438279342878,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.00040553025223816615,
|
|
"loss": 5.8679,
|
|
"mean_token_accuracy": 0.1243274100124836,
|
|
"num_tokens": 4309990.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 5.914406776428223,
|
|
"epoch": 0.12634426337665766,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00040456888272171653,
|
|
"loss": 5.7742,
|
|
"mean_token_accuracy": 0.13342205286026002,
|
|
"num_tokens": 4320825.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 5.877487564086914,
|
|
"epoch": 0.12667414395988652,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00040360396100889577,
|
|
"loss": 5.8491,
|
|
"mean_token_accuracy": 0.1308812364935875,
|
|
"num_tokens": 4332113.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 5.886609268188477,
|
|
"epoch": 0.12700402454311538,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004026355135535202,
|
|
"loss": 5.9013,
|
|
"mean_token_accuracy": 0.12625984475016594,
|
|
"num_tokens": 4343436.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 5.8597118854522705,
|
|
"epoch": 0.12733390512634427,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.000401663566906066,
|
|
"loss": 5.8672,
|
|
"mean_token_accuracy": 0.12835360690951347,
|
|
"num_tokens": 4354582.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 5.913357257843018,
|
|
"epoch": 0.12766378570957312,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00040068814771294134,
|
|
"loss": 5.8283,
|
|
"mean_token_accuracy": 0.13204079493880272,
|
|
"num_tokens": 4365175.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 5.860588836669922,
|
|
"epoch": 0.127993666292802,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0003997092827157562,
|
|
"loss": 5.8087,
|
|
"mean_token_accuracy": 0.13259225264191626,
|
|
"num_tokens": 4377613.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 5.883626556396484,
|
|
"epoch": 0.12832354687603087,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.000398726998750589,
|
|
"loss": 5.8827,
|
|
"mean_token_accuracy": 0.12795776650309562,
|
|
"num_tokens": 4389829.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 5.90832486152649,
|
|
"epoch": 0.12865342745925976,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00039774132274725076,
|
|
"loss": 5.8538,
|
|
"mean_token_accuracy": 0.12682538330554963,
|
|
"num_tokens": 4401437.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 5.79905948638916,
|
|
"epoch": 0.12898330804248861,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00039675228172854707,
|
|
"loss": 5.7671,
|
|
"mean_token_accuracy": 0.13425397872924805,
|
|
"num_tokens": 4413235.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 5.799765872955322,
|
|
"epoch": 0.1293131886257175,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0003957599028095371,
|
|
"loss": 5.8115,
|
|
"mean_token_accuracy": 0.13904800564050673,
|
|
"num_tokens": 4424686.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 5.876850414276123,
|
|
"epoch": 0.12964306920894636,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00039476421319679017,
|
|
"loss": 5.7738,
|
|
"mean_token_accuracy": 0.12886618971824645,
|
|
"num_tokens": 4435176.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 5.857044219970703,
|
|
"epoch": 0.12997294979217525,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00039376524018764,
|
|
"loss": 5.8187,
|
|
"mean_token_accuracy": 0.13269937857985498,
|
|
"num_tokens": 4446923.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 5.928748655319214,
|
|
"epoch": 0.1303028303754041,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00039276301116943616,
|
|
"loss": 5.8608,
|
|
"mean_token_accuracy": 0.13216537684202195,
|
|
"num_tokens": 4457820.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 5.875562429428101,
|
|
"epoch": 0.13063271095863296,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0003917575536187936,
|
|
"loss": 5.7639,
|
|
"mean_token_accuracy": 0.13873712122440338,
|
|
"num_tokens": 4468761.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 5.861223459243774,
|
|
"epoch": 0.13096259154186185,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00039074889510083894,
|
|
"loss": 5.8102,
|
|
"mean_token_accuracy": 0.13628667071461678,
|
|
"num_tokens": 4480352.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 5.767582607269287,
|
|
"epoch": 0.1312924721250907,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.00038973706326845495,
|
|
"loss": 5.8016,
|
|
"mean_token_accuracy": 0.1324378788471222,
|
|
"num_tokens": 4490792.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 5.943720769882202,
|
|
"epoch": 0.1316223527083196,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0003887220858615225,
|
|
"loss": 5.8987,
|
|
"mean_token_accuracy": 0.1264105223119259,
|
|
"num_tokens": 4502293.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 5.9420631408691404,
|
|
"epoch": 0.13195223329154845,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0003877039907061597,
|
|
"loss": 5.8665,
|
|
"mean_token_accuracy": 0.12843255698680878,
|
|
"num_tokens": 4512476.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 5.89266881942749,
|
|
"epoch": 0.13228211387477734,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0003866828057139598,
|
|
"loss": 5.7613,
|
|
"mean_token_accuracy": 0.13228247985243796,
|
|
"num_tokens": 4523874.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 5.884331703186035,
|
|
"epoch": 0.1326119944580062,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00038565855888122503,
|
|
"loss": 5.7406,
|
|
"mean_token_accuracy": 0.13756398260593414,
|
|
"num_tokens": 4534845.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 5.8414264678955075,
|
|
"epoch": 0.13294187504123509,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00038463127828819975,
|
|
"loss": 5.8308,
|
|
"mean_token_accuracy": 0.1307374931871891,
|
|
"num_tokens": 4545616.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 5.836323213577271,
|
|
"epoch": 0.13327175562446394,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00038360099209830043,
|
|
"loss": 5.7976,
|
|
"mean_token_accuracy": 0.13549839034676553,
|
|
"num_tokens": 4555834.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 5.914280414581299,
|
|
"epoch": 0.1336016362076928,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0003825677285573433,
|
|
"loss": 5.8058,
|
|
"mean_token_accuracy": 0.13284233286976815,
|
|
"num_tokens": 4568671.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 5.798496532440185,
|
|
"epoch": 0.1339315167909217,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00038153151599277027,
|
|
"loss": 5.7493,
|
|
"mean_token_accuracy": 0.1281888708472252,
|
|
"num_tokens": 4579114.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 5.852269458770752,
|
|
"epoch": 0.13426139737415055,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003804923828128723,
|
|
"loss": 5.7992,
|
|
"mean_token_accuracy": 0.13389373943209648,
|
|
"num_tokens": 4589587.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 5.821191501617432,
|
|
"epoch": 0.13459127795737943,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0003794503575060104,
|
|
"loss": 5.719,
|
|
"mean_token_accuracy": 0.133248271048069,
|
|
"num_tokens": 4600287.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 5.925690984725952,
|
|
"epoch": 0.1349211585406083,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00037840546863983484,
|
|
"loss": 5.8168,
|
|
"mean_token_accuracy": 0.13238527849316598,
|
|
"num_tokens": 4611279.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 5.844077157974243,
|
|
"epoch": 0.13525103912383718,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003773577448605015,
|
|
"loss": 5.8705,
|
|
"mean_token_accuracy": 0.13018327206373215,
|
|
"num_tokens": 4622109.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 5.905648326873779,
|
|
"epoch": 0.13558091970706604,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0003763072148918872,
|
|
"loss": 5.7999,
|
|
"mean_token_accuracy": 0.12815057262778282,
|
|
"num_tokens": 4633857.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 5.848460721969604,
|
|
"epoch": 0.13591080029029493,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003752539075348017,
|
|
"loss": 5.7582,
|
|
"mean_token_accuracy": 0.13111123740673064,
|
|
"num_tokens": 4644528.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 5.88867769241333,
|
|
"epoch": 0.13624068087352378,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00037419785166619817,
|
|
"loss": 5.8412,
|
|
"mean_token_accuracy": 0.133074089884758,
|
|
"num_tokens": 4655950.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 5.808388662338257,
|
|
"epoch": 0.13657056145675264,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0003731390762383818,
|
|
"loss": 5.7652,
|
|
"mean_token_accuracy": 0.13077523782849312,
|
|
"num_tokens": 4668083.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 5.787566375732422,
|
|
"epoch": 0.13690044203998153,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0003720776102782158,
|
|
"loss": 5.6961,
|
|
"mean_token_accuracy": 0.13819554150104524,
|
|
"num_tokens": 4679430.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 5.8818971633911135,
|
|
"epoch": 0.1372303226232104,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00037101348288632555,
|
|
"loss": 5.7589,
|
|
"mean_token_accuracy": 0.12942191883921622,
|
|
"num_tokens": 4690331.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 5.842368841171265,
|
|
"epoch": 0.13756020320643927,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.0003699467232363012,
|
|
"loss": 5.8953,
|
|
"mean_token_accuracy": 0.13223558962345122,
|
|
"num_tokens": 4701012.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 5.841617250442505,
|
|
"epoch": 0.13789008378966813,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0003688773605738973,
|
|
"loss": 5.7594,
|
|
"mean_token_accuracy": 0.12926423400640488,
|
|
"num_tokens": 4711043.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 5.872353887557983,
|
|
"epoch": 0.13821996437289702,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.00036780542421623134,
|
|
"loss": 5.7648,
|
|
"mean_token_accuracy": 0.13941656351089476,
|
|
"num_tokens": 4723596.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 5.792061138153076,
|
|
"epoch": 0.13854984495612588,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003667309435509802,
|
|
"loss": 5.7957,
|
|
"mean_token_accuracy": 0.13276198208332063,
|
|
"num_tokens": 4733911.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 5.8638628959655765,
|
|
"epoch": 0.13887972553935476,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0003656539480355741,
|
|
"loss": 5.7984,
|
|
"mean_token_accuracy": 0.13170343935489653,
|
|
"num_tokens": 4745958.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 5.854230356216431,
|
|
"epoch": 0.13920960612258362,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0003645744671963891,
|
|
"loss": 5.7899,
|
|
"mean_token_accuracy": 0.12614999040961267,
|
|
"num_tokens": 4757060.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 5.786357641220093,
|
|
"epoch": 0.13953948670581248,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0003634925306279376,
|
|
"loss": 5.7463,
|
|
"mean_token_accuracy": 0.13023732900619506,
|
|
"num_tokens": 4769029.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 5.843477249145508,
|
|
"epoch": 0.13986936728904137,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003624081679920574,
|
|
"loss": 5.8061,
|
|
"mean_token_accuracy": 0.13537542074918746,
|
|
"num_tokens": 4779451.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 5.782986164093018,
|
|
"epoch": 0.14019924787227023,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0003613214090170977,
|
|
"loss": 5.7467,
|
|
"mean_token_accuracy": 0.13251159638166427,
|
|
"num_tokens": 4791350.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 5.805937433242798,
|
|
"epoch": 0.1405291284554991,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0003602322834971048,
|
|
"loss": 5.7458,
|
|
"mean_token_accuracy": 0.13328628540039061,
|
|
"num_tokens": 4802117.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 5.856236219406128,
|
|
"epoch": 0.14085900903872797,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0003591408212910051,
|
|
"loss": 5.7541,
|
|
"mean_token_accuracy": 0.13448208421468735,
|
|
"num_tokens": 4812966.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 5.910129356384277,
|
|
"epoch": 0.14118888962195686,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0003580470523217863,
|
|
"loss": 5.8151,
|
|
"mean_token_accuracy": 0.13002887442708017,
|
|
"num_tokens": 4825228.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 5.846267366409302,
|
|
"epoch": 0.14151877020518572,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0003569510065756771,
|
|
"loss": 5.8321,
|
|
"mean_token_accuracy": 0.13108750730752944,
|
|
"num_tokens": 4837282.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 5.794789743423462,
|
|
"epoch": 0.1418486507884146,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0003558527141013254,
|
|
"loss": 5.7032,
|
|
"mean_token_accuracy": 0.13718815073370932,
|
|
"num_tokens": 4848882.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 5.81343560218811,
|
|
"epoch": 0.14217853137164346,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0003547522050089742,
|
|
"loss": 5.7294,
|
|
"mean_token_accuracy": 0.13713167160749434,
|
|
"num_tokens": 4858963.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 5.796407222747803,
|
|
"epoch": 0.14250841195487235,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00035364950946963606,
|
|
"loss": 5.7845,
|
|
"mean_token_accuracy": 0.12842966318130494,
|
|
"num_tokens": 4870055.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 5.85999026298523,
|
|
"epoch": 0.1428382925381012,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0003525446577142663,
|
|
"loss": 5.7916,
|
|
"mean_token_accuracy": 0.13417006507515908,
|
|
"num_tokens": 4881840.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 5.872052717208862,
|
|
"epoch": 0.14316817312133007,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.00035143768003293395,
|
|
"loss": 5.7796,
|
|
"mean_token_accuracy": 0.13351808339357377,
|
|
"num_tokens": 4894108.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 5.836113786697387,
|
|
"epoch": 0.14349805370455895,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.0003503286067739913,
|
|
"loss": 5.797,
|
|
"mean_token_accuracy": 0.13248383179306983,
|
|
"num_tokens": 4906495.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 5.8112053871154785,
|
|
"epoch": 0.1438279342877878,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00034921746834324193,
|
|
"loss": 5.7689,
|
|
"mean_token_accuracy": 0.13340991139411926,
|
|
"num_tokens": 4917701.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 5.794090127944946,
|
|
"epoch": 0.1441578148710167,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.0003481042952031072,
|
|
"loss": 5.7978,
|
|
"mean_token_accuracy": 0.1279979020357132,
|
|
"num_tokens": 4930697.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 5.83536958694458,
|
|
"epoch": 0.14448769545424556,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0003469891178717911,
|
|
"loss": 5.7565,
|
|
"mean_token_accuracy": 0.1364119306206703,
|
|
"num_tokens": 4942197.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 5.868331241607666,
|
|
"epoch": 0.14481757603747444,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0003458719669224436,
|
|
"loss": 5.7726,
|
|
"mean_token_accuracy": 0.13440410867333413,
|
|
"num_tokens": 4953963.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 5.782655429840088,
|
|
"epoch": 0.1451474566207033,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.0003447528729823221,
|
|
"loss": 5.6752,
|
|
"mean_token_accuracy": 0.139103439450264,
|
|
"num_tokens": 4965878.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 5.832769060134888,
|
|
"epoch": 0.1454773372039322,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0003436318667319525,
|
|
"loss": 5.7817,
|
|
"mean_token_accuracy": 0.13693246692419053,
|
|
"num_tokens": 4977421.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 5.797252035140991,
|
|
"epoch": 0.14580721778716105,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00034250897890428716,
|
|
"loss": 5.7301,
|
|
"mean_token_accuracy": 0.13630487620830536,
|
|
"num_tokens": 4989018.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 5.862127351760864,
|
|
"epoch": 0.1461370983703899,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0003413842402838633,
|
|
"loss": 5.8043,
|
|
"mean_token_accuracy": 0.13428834751248359,
|
|
"num_tokens": 5000800.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 5.833758974075318,
|
|
"epoch": 0.1464669789536188,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00034025768170595834,
|
|
"loss": 5.8013,
|
|
"mean_token_accuracy": 0.1311846248805523,
|
|
"num_tokens": 5012229.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 5.870166254043579,
|
|
"epoch": 0.14679685953684765,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0003391293340557446,
|
|
"loss": 5.8086,
|
|
"mean_token_accuracy": 0.13377721905708312,
|
|
"num_tokens": 5022606.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 5.822594356536865,
|
|
"epoch": 0.14712674012007654,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003379992282674431,
|
|
"loss": 5.6235,
|
|
"mean_token_accuracy": 0.14179335311055183,
|
|
"num_tokens": 5033238.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 5.8109277248382565,
|
|
"epoch": 0.1474566207033054,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0003368673953234749,
|
|
"loss": 5.7216,
|
|
"mean_token_accuracy": 0.13757843896746635,
|
|
"num_tokens": 5043847.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 5.802297592163086,
|
|
"epoch": 0.14778650128653428,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00033573386625361176,
|
|
"loss": 5.7494,
|
|
"mean_token_accuracy": 0.14292570650577546,
|
|
"num_tokens": 5055709.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 5.8141319274902346,
|
|
"epoch": 0.14811638186976314,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00033459867213412567,
|
|
"loss": 5.6903,
|
|
"mean_token_accuracy": 0.13453157171607016,
|
|
"num_tokens": 5068465.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 5.831532955169678,
|
|
"epoch": 0.14844626245299203,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000333461844086937,
|
|
"loss": 5.8049,
|
|
"mean_token_accuracy": 0.12702211886644363,
|
|
"num_tokens": 5078044.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 5.765716981887818,
|
|
"epoch": 0.1487761430362209,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00033232341327876097,
|
|
"loss": 5.669,
|
|
"mean_token_accuracy": 0.13550253808498383,
|
|
"num_tokens": 5089419.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 5.768707227706909,
|
|
"epoch": 0.14910602361944975,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0003311834109202531,
|
|
"loss": 5.7468,
|
|
"mean_token_accuracy": 0.1355135530233383,
|
|
"num_tokens": 5100949.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 5.842748498916626,
|
|
"epoch": 0.14943590420267863,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00033004186826515416,
|
|
"loss": 5.7695,
|
|
"mean_token_accuracy": 0.1309277407824993,
|
|
"num_tokens": 5111339.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 5.787210941314697,
|
|
"epoch": 0.1497657847859075,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003288988166094324,
|
|
"loss": 5.7581,
|
|
"mean_token_accuracy": 0.13953553810715674,
|
|
"num_tokens": 5121090.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 5.795881509780884,
|
|
"epoch": 0.15009566536913638,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00032775428729042656,
|
|
"loss": 5.783,
|
|
"mean_token_accuracy": 0.13089647740125657,
|
|
"num_tokens": 5133413.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 5.840445137023925,
|
|
"epoch": 0.15042554595236524,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000326608311685986,
|
|
"loss": 5.7027,
|
|
"mean_token_accuracy": 0.13714413791894914,
|
|
"num_tokens": 5144011.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 5.8135875225067135,
|
|
"epoch": 0.15075542653559412,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0003254609212136108,
|
|
"loss": 5.7275,
|
|
"mean_token_accuracy": 0.1346348188817501,
|
|
"num_tokens": 5154766.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 5.832667636871338,
|
|
"epoch": 0.15108530711882298,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00032431214732959036,
|
|
"loss": 5.7579,
|
|
"mean_token_accuracy": 0.1312684342265129,
|
|
"num_tokens": 5166092.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 5.7650025367736815,
|
|
"epoch": 0.15141518770205187,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.000323162021528141,
|
|
"loss": 5.6621,
|
|
"mean_token_accuracy": 0.14111037403345109,
|
|
"num_tokens": 5176533.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 5.815569305419922,
|
|
"epoch": 0.15174506828528073,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00032201057534054264,
|
|
"loss": 5.7769,
|
|
"mean_token_accuracy": 0.13670074865221976,
|
|
"num_tokens": 5186647.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 5.836896228790283,
|
|
"epoch": 0.1520749488685096,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00032085784033427414,
|
|
"loss": 5.7191,
|
|
"mean_token_accuracy": 0.14382374212145804,
|
|
"num_tokens": 5197785.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 5.706895446777343,
|
|
"epoch": 0.15240482945173847,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003197038481121478,
|
|
"loss": 5.7602,
|
|
"mean_token_accuracy": 0.1329088881611824,
|
|
"num_tokens": 5208557.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 5.917945003509521,
|
|
"epoch": 0.15273471003496733,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0003185486303114436,
|
|
"loss": 5.7984,
|
|
"mean_token_accuracy": 0.1355454221367836,
|
|
"num_tokens": 5220243.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 5.897228193283081,
|
|
"epoch": 0.15306459061819622,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003173922186030409,
|
|
"loss": 5.8083,
|
|
"mean_token_accuracy": 0.1302622489631176,
|
|
"num_tokens": 5230426.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 5.734656143188476,
|
|
"epoch": 0.15339447120142508,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000316234644690551,
|
|
"loss": 5.7354,
|
|
"mean_token_accuracy": 0.1334875725209713,
|
|
"num_tokens": 5241425.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 5.937242698669434,
|
|
"epoch": 0.15372435178465396,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003150759403094473,
|
|
"loss": 5.8043,
|
|
"mean_token_accuracy": 0.13442886993288994,
|
|
"num_tokens": 5252620.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 5.807015657424927,
|
|
"epoch": 0.15405423236788282,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00031391613722619587,
|
|
"loss": 5.7632,
|
|
"mean_token_accuracy": 0.13534248173236846,
|
|
"num_tokens": 5264284.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 5.806159925460816,
|
|
"epoch": 0.1543841129511117,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000312755267237384,
|
|
"loss": 5.7463,
|
|
"mean_token_accuracy": 0.14027821868658066,
|
|
"num_tokens": 5274382.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 5.847917938232422,
|
|
"epoch": 0.15471399353434057,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0003115933621688488,
|
|
"loss": 5.7779,
|
|
"mean_token_accuracy": 0.13340894803404807,
|
|
"num_tokens": 5286116.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 5.796377325057984,
|
|
"epoch": 0.15504387411756945,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00031043045387480487,
|
|
"loss": 5.6861,
|
|
"mean_token_accuracy": 0.1381146304309368,
|
|
"num_tokens": 5295918.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 5.794605922698975,
|
|
"epoch": 0.1553737547007983,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0003092665742369703,
|
|
"loss": 5.7169,
|
|
"mean_token_accuracy": 0.1330300159752369,
|
|
"num_tokens": 5306529.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 5.878596830368042,
|
|
"epoch": 0.15570363528402717,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00030810175516369343,
|
|
"loss": 5.7398,
|
|
"mean_token_accuracy": 0.1362038716673851,
|
|
"num_tokens": 5318089.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 5.823804092407227,
|
|
"epoch": 0.15603351586725606,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0003069360285890775,
|
|
"loss": 5.7447,
|
|
"mean_token_accuracy": 0.1360376223921776,
|
|
"num_tokens": 5329563.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 5.803675889968872,
|
|
"epoch": 0.15636339645048491,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00030576942647210547,
|
|
"loss": 5.6521,
|
|
"mean_token_accuracy": 0.14254518449306489,
|
|
"num_tokens": 5341100.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 5.76872181892395,
|
|
"epoch": 0.1566932770337138,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00030460198079576355,
|
|
"loss": 5.7034,
|
|
"mean_token_accuracy": 0.1379267930984497,
|
|
"num_tokens": 5351150.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 5.77161865234375,
|
|
"epoch": 0.15702315761694266,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0003034337235661648,
|
|
"loss": 5.6928,
|
|
"mean_token_accuracy": 0.13706344813108445,
|
|
"num_tokens": 5362481.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 5.790349912643433,
|
|
"epoch": 0.15735303820017155,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0003022646868116714,
|
|
"loss": 5.6785,
|
|
"mean_token_accuracy": 0.14309663847088813,
|
|
"num_tokens": 5372869.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 5.783022880554199,
|
|
"epoch": 0.1576829187834004,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003010949025820163,
|
|
"loss": 5.7008,
|
|
"mean_token_accuracy": 0.13485823348164558,
|
|
"num_tokens": 5382804.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 5.780787754058838,
|
|
"epoch": 0.1580127993666293,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0002999244029474252,
|
|
"loss": 5.6463,
|
|
"mean_token_accuracy": 0.14524321407079696,
|
|
"num_tokens": 5393682.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 5.788739490509033,
|
|
"epoch": 0.15834267994985815,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.00029875321999773684,
|
|
"loss": 5.7422,
|
|
"mean_token_accuracy": 0.13701190501451493,
|
|
"num_tokens": 5406709.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 5.78806438446045,
|
|
"epoch": 0.158672560533087,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00029758138584152333,
|
|
"loss": 5.6803,
|
|
"mean_token_accuracy": 0.13673285618424416,
|
|
"num_tokens": 5418103.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 5.830794763565064,
|
|
"epoch": 0.1590024411163159,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0002964089326052102,
|
|
"loss": 5.7244,
|
|
"mean_token_accuracy": 0.13409437164664267,
|
|
"num_tokens": 5428210.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 5.760257291793823,
|
|
"epoch": 0.15933232169954475,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0002952358924321949,
|
|
"loss": 5.695,
|
|
"mean_token_accuracy": 0.13741234466433525,
|
|
"num_tokens": 5440270.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 5.79687180519104,
|
|
"epoch": 0.15966220228277364,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00029406229748196657,
|
|
"loss": 5.6628,
|
|
"mean_token_accuracy": 0.14232583045959474,
|
|
"num_tokens": 5450757.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 5.795097875595093,
|
|
"epoch": 0.1599920828660025,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0002928881799292235,
|
|
"loss": 5.6951,
|
|
"mean_token_accuracy": 0.14370114505290985,
|
|
"num_tokens": 5461022.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 5.804535341262818,
|
|
"epoch": 0.16032196344923139,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00029171357196299154,
|
|
"loss": 5.6645,
|
|
"mean_token_accuracy": 0.1391290545463562,
|
|
"num_tokens": 5472135.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 5.787743282318115,
|
|
"epoch": 0.16065184403246024,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0002905385057857414,
|
|
"loss": 5.7385,
|
|
"mean_token_accuracy": 0.13327718228101731,
|
|
"num_tokens": 5483948.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 5.820004415512085,
|
|
"epoch": 0.16098172461568913,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0002893630136125058,
|
|
"loss": 5.7216,
|
|
"mean_token_accuracy": 0.13662510216236115,
|
|
"num_tokens": 5495668.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 5.7640032291412355,
|
|
"epoch": 0.161311605198918,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0002881871276699967,
|
|
"loss": 5.7678,
|
|
"mean_token_accuracy": 0.13400556817650794,
|
|
"num_tokens": 5506798.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 5.862719631195068,
|
|
"epoch": 0.16164148578214688,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00028701088019572114,
|
|
"loss": 5.6464,
|
|
"mean_token_accuracy": 0.13870992213487626,
|
|
"num_tokens": 5516519.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 5.714185571670532,
|
|
"epoch": 0.16197136636537574,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0002858343034370977,
|
|
"loss": 5.5558,
|
|
"mean_token_accuracy": 0.14398761540651323,
|
|
"num_tokens": 5526177.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 5.759743976593017,
|
|
"epoch": 0.1623012469486046,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00028465742965057267,
|
|
"loss": 5.7682,
|
|
"mean_token_accuracy": 0.13722870796918868,
|
|
"num_tokens": 5538006.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 5.832506370544434,
|
|
"epoch": 0.16263112753183348,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00028348029110073533,
|
|
"loss": 5.6628,
|
|
"mean_token_accuracy": 0.13731009736657143,
|
|
"num_tokens": 5548270.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 5.75463376045227,
|
|
"epoch": 0.16296100811506234,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00028230292005943365,
|
|
"loss": 5.729,
|
|
"mean_token_accuracy": 0.1391895443201065,
|
|
"num_tokens": 5560101.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 5.881022167205811,
|
|
"epoch": 0.16329088869829123,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00028112534880488945,
|
|
"loss": 5.6926,
|
|
"mean_token_accuracy": 0.13788802474737166,
|
|
"num_tokens": 5570572.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 5.882670259475708,
|
|
"epoch": 0.16362076928152008,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0002799476096208137,
|
|
"loss": 5.7341,
|
|
"mean_token_accuracy": 0.14039146453142165,
|
|
"num_tokens": 5581067.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 5.72330961227417,
|
|
"epoch": 0.16395064986474897,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00027876973479552087,
|
|
"loss": 5.5966,
|
|
"mean_token_accuracy": 0.13605255708098413,
|
|
"num_tokens": 5591818.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 5.673970746994018,
|
|
"epoch": 0.16428053044797783,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00027759175662104424,
|
|
"loss": 5.6756,
|
|
"mean_token_accuracy": 0.14364527985453607,
|
|
"num_tokens": 5604533.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 5.840668392181397,
|
|
"epoch": 0.16461041103120672,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0002764137073922508,
|
|
"loss": 5.7506,
|
|
"mean_token_accuracy": 0.130050141364336,
|
|
"num_tokens": 5615379.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 5.822918128967285,
|
|
"epoch": 0.16494029161443557,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00027523561940595505,
|
|
"loss": 5.6826,
|
|
"mean_token_accuracy": 0.13614660277962684,
|
|
"num_tokens": 5627467.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 5.766177701950073,
|
|
"epoch": 0.16527017219766443,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0002740575249600342,
|
|
"loss": 5.7781,
|
|
"mean_token_accuracy": 0.13605818003416062,
|
|
"num_tokens": 5639177.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 5.827766990661621,
|
|
"epoch": 0.16560005278089332,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.00027287945635254263,
|
|
"loss": 5.6809,
|
|
"mean_token_accuracy": 0.1369494304060936,
|
|
"num_tokens": 5651465.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 5.765095806121826,
|
|
"epoch": 0.16592993336412218,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00027170144588082635,
|
|
"loss": 5.6722,
|
|
"mean_token_accuracy": 0.14486928880214692,
|
|
"num_tokens": 5662940.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 5.847509860992432,
|
|
"epoch": 0.16625981394735106,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00027052352584063763,
|
|
"loss": 5.699,
|
|
"mean_token_accuracy": 0.13465003669261932,
|
|
"num_tokens": 5674774.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 5.765509414672851,
|
|
"epoch": 0.16658969453057992,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00026934572852524907,
|
|
"loss": 5.6142,
|
|
"mean_token_accuracy": 0.13957108929753304,
|
|
"num_tokens": 5686838.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 5.670578861236573,
|
|
"epoch": 0.1669195751138088,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00026816808622456937,
|
|
"loss": 5.6528,
|
|
"mean_token_accuracy": 0.14574100375175475,
|
|
"num_tokens": 5697711.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 5.7856707096099855,
|
|
"epoch": 0.16724945569703767,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0002669906312242569,
|
|
"loss": 5.6618,
|
|
"mean_token_accuracy": 0.14712682217359543,
|
|
"num_tokens": 5708627.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 5.765988302230835,
|
|
"epoch": 0.16757933628026656,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.00026581339580483525,
|
|
"loss": 5.7426,
|
|
"mean_token_accuracy": 0.14225278347730635,
|
|
"num_tokens": 5719822.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 5.7975242137908936,
|
|
"epoch": 0.16790921686349541,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0002646364122408082,
|
|
"loss": 5.6313,
|
|
"mean_token_accuracy": 0.14257853478193283,
|
|
"num_tokens": 5730392.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 5.841425275802612,
|
|
"epoch": 0.16823909744672427,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0002634597127997749,
|
|
"loss": 5.6875,
|
|
"mean_token_accuracy": 0.14245052486658097,
|
|
"num_tokens": 5742923.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 5.811284399032592,
|
|
"epoch": 0.16856897802995316,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0002622833297415445,
|
|
"loss": 5.6875,
|
|
"mean_token_accuracy": 0.13664121627807618,
|
|
"num_tokens": 5753787.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 5.768994903564453,
|
|
"epoch": 0.16889885861318202,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0002611072953172531,
|
|
"loss": 5.655,
|
|
"mean_token_accuracy": 0.13747793212532997,
|
|
"num_tokens": 5766449.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 5.7201892852783205,
|
|
"epoch": 0.1692287391964109,
|
|
"grad_norm": 0.79296875,
|
|
"learning_rate": 0.00025993164176847845,
|
|
"loss": 5.6409,
|
|
"mean_token_accuracy": 0.14449931532144547,
|
|
"num_tokens": 5779102.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 5.835258150100708,
|
|
"epoch": 0.16955861977963976,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0002587564013263564,
|
|
"loss": 5.7508,
|
|
"mean_token_accuracy": 0.1352461151778698,
|
|
"num_tokens": 5790072.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 5.757137203216553,
|
|
"epoch": 0.16988850036286865,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0002575816062106974,
|
|
"loss": 5.6297,
|
|
"mean_token_accuracy": 0.14009721279144288,
|
|
"num_tokens": 5800896.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 5.765016365051269,
|
|
"epoch": 0.1702183809460975,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00025640728862910293,
|
|
"loss": 5.7207,
|
|
"mean_token_accuracy": 0.1391066126525402,
|
|
"num_tokens": 5811897.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 5.822316408157349,
|
|
"epoch": 0.1705482615293264,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.00025523348077608285,
|
|
"loss": 5.6741,
|
|
"mean_token_accuracy": 0.13919610753655434,
|
|
"num_tokens": 5823999.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 5.759761571884155,
|
|
"epoch": 0.17087814211255525,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00025406021483217225,
|
|
"loss": 5.704,
|
|
"mean_token_accuracy": 0.14073672890663147,
|
|
"num_tokens": 5836490.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 5.7949645042419435,
|
|
"epoch": 0.17120802269578414,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00025288752296304963,
|
|
"loss": 5.6664,
|
|
"mean_token_accuracy": 0.142671100795269,
|
|
"num_tokens": 5847674.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 5.774451684951782,
|
|
"epoch": 0.171537903279013,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.000251715437318655,
|
|
"loss": 5.606,
|
|
"mean_token_accuracy": 0.14566963613033296,
|
|
"num_tokens": 5858585.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 5.7460608959198,
|
|
"epoch": 0.17186778386224186,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0002505439900323084,
|
|
"loss": 5.5903,
|
|
"mean_token_accuracy": 0.1375995397567749,
|
|
"num_tokens": 5868679.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 5.828930902481079,
|
|
"epoch": 0.17219766444547074,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00024937321321982894,
|
|
"loss": 5.7342,
|
|
"mean_token_accuracy": 0.131529800593853,
|
|
"num_tokens": 5879662.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 5.767243528366089,
|
|
"epoch": 0.1725275450286996,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00024820313897865433,
|
|
"loss": 5.6746,
|
|
"mean_token_accuracy": 0.13928751945495604,
|
|
"num_tokens": 5891948.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 5.817196035385132,
|
|
"epoch": 0.1728574256119285,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00024703379938696105,
|
|
"loss": 5.5946,
|
|
"mean_token_accuracy": 0.1431368075311184,
|
|
"num_tokens": 5902994.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 5.787667989730835,
|
|
"epoch": 0.17318730619515735,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00024586522650278447,
|
|
"loss": 5.6576,
|
|
"mean_token_accuracy": 0.1369992323219776,
|
|
"num_tokens": 5913708.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 5.7192298412323,
|
|
"epoch": 0.17351718677838623,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.00024469745236314064,
|
|
"loss": 5.6958,
|
|
"mean_token_accuracy": 0.13990109115839006,
|
|
"num_tokens": 5926093.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 5.768823194503784,
|
|
"epoch": 0.1738470673616151,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.00024353050898314767,
|
|
"loss": 5.6552,
|
|
"mean_token_accuracy": 0.14087175726890563,
|
|
"num_tokens": 5936664.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 5.7733049392700195,
|
|
"epoch": 0.17417694794484398,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00024236442835514743,
|
|
"loss": 5.6185,
|
|
"mean_token_accuracy": 0.13843318596482276,
|
|
"num_tokens": 5946713.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 5.863949537277222,
|
|
"epoch": 0.17450682852807284,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00024119924244782965,
|
|
"loss": 5.7199,
|
|
"mean_token_accuracy": 0.13668633103370667,
|
|
"num_tokens": 5957671.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 5.77341661453247,
|
|
"epoch": 0.1748367091113017,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00024003498320535462,
|
|
"loss": 5.5963,
|
|
"mean_token_accuracy": 0.1414831206202507,
|
|
"num_tokens": 5967555.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 5.762756490707398,
|
|
"epoch": 0.17516658969453058,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.00023887168254647727,
|
|
"loss": 5.7215,
|
|
"mean_token_accuracy": 0.13925121277570723,
|
|
"num_tokens": 5979210.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 5.809481143951416,
|
|
"epoch": 0.17549647027775944,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00023770937236367308,
|
|
"loss": 5.6622,
|
|
"mean_token_accuracy": 0.1392768770456314,
|
|
"num_tokens": 5990501.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 5.764357662200927,
|
|
"epoch": 0.17582635086098833,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00023654808452226278,
|
|
"loss": 5.7297,
|
|
"mean_token_accuracy": 0.1385626032948494,
|
|
"num_tokens": 6003520.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 5.805681943893433,
|
|
"epoch": 0.1761562314442172,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00023538785085953912,
|
|
"loss": 5.7085,
|
|
"mean_token_accuracy": 0.14266811311244965,
|
|
"num_tokens": 6013404.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 5.730310773849487,
|
|
"epoch": 0.17648611202744607,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00023422870318389404,
|
|
"loss": 5.5395,
|
|
"mean_token_accuracy": 0.1547106996178627,
|
|
"num_tokens": 6024275.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 5.7997983455657955,
|
|
"epoch": 0.17681599261067493,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0002330706732739468,
|
|
"loss": 5.6634,
|
|
"mean_token_accuracy": 0.1378786638379097,
|
|
"num_tokens": 6036693.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 5.800734996795654,
|
|
"epoch": 0.17714587319390382,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00023191379287767211,
|
|
"loss": 5.662,
|
|
"mean_token_accuracy": 0.13974825590848922,
|
|
"num_tokens": 6046984.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 5.897027158737183,
|
|
"epoch": 0.17747575377713268,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0002307580937115305,
|
|
"loss": 5.8217,
|
|
"mean_token_accuracy": 0.13140921518206597,
|
|
"num_tokens": 6059435.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 5.827291679382324,
|
|
"epoch": 0.17780563436036154,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00022960360745959846,
|
|
"loss": 5.5757,
|
|
"mean_token_accuracy": 0.1396414391696453,
|
|
"num_tokens": 6069800.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 5.834706830978393,
|
|
"epoch": 0.17813551494359042,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.00022845036577269972,
|
|
"loss": 5.7347,
|
|
"mean_token_accuracy": 0.1358237124979496,
|
|
"num_tokens": 6081438.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 5.794511699676514,
|
|
"epoch": 0.17846539552681928,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00022729840026753777,
|
|
"loss": 5.7403,
|
|
"mean_token_accuracy": 0.1409032590687275,
|
|
"num_tokens": 6094507.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 5.815364742279053,
|
|
"epoch": 0.17879527611004817,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0002261477425258287,
|
|
"loss": 5.6542,
|
|
"mean_token_accuracy": 0.1419000506401062,
|
|
"num_tokens": 6104406.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 5.7367048263549805,
|
|
"epoch": 0.17912515669327703,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0002249984240934358,
|
|
"loss": 5.7132,
|
|
"mean_token_accuracy": 0.13407342061400412,
|
|
"num_tokens": 6116970.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 5.805830335617065,
|
|
"epoch": 0.1794550372765059,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00022385047647950464,
|
|
"loss": 5.6859,
|
|
"mean_token_accuracy": 0.13379801586270332,
|
|
"num_tokens": 6127418.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 5.772081756591797,
|
|
"epoch": 0.17978491785973477,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0002227039311555986,
|
|
"loss": 5.6179,
|
|
"mean_token_accuracy": 0.14391135573387145,
|
|
"num_tokens": 6138575.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 5.8133808135986325,
|
|
"epoch": 0.18011479844296366,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0002215588195548372,
|
|
"loss": 5.7117,
|
|
"mean_token_accuracy": 0.13847993835806846,
|
|
"num_tokens": 6148888.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 5.7434896469116214,
|
|
"epoch": 0.18044467902619252,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00022041517307103337,
|
|
"loss": 5.6138,
|
|
"mean_token_accuracy": 0.14814298450946808,
|
|
"num_tokens": 6160482.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 5.754063558578491,
|
|
"epoch": 0.18077455960942138,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0002192730230578331,
|
|
"loss": 5.6022,
|
|
"mean_token_accuracy": 0.1389575108885765,
|
|
"num_tokens": 6172290.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 5.769651412963867,
|
|
"epoch": 0.18110444019265026,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002181324008278559,
|
|
"loss": 5.6793,
|
|
"mean_token_accuracy": 0.14809389263391495,
|
|
"num_tokens": 6183144.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 5.78060998916626,
|
|
"epoch": 0.18143432077587912,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00021699333765183655,
|
|
"loss": 5.6614,
|
|
"mean_token_accuracy": 0.1429641515016556,
|
|
"num_tokens": 6193976.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 5.79771580696106,
|
|
"epoch": 0.181764201359108,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0002158558647577673,
|
|
"loss": 5.6008,
|
|
"mean_token_accuracy": 0.14147469848394395,
|
|
"num_tokens": 6203953.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 5.704411458969116,
|
|
"epoch": 0.18209408194233687,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00021472001333004215,
|
|
"loss": 5.5736,
|
|
"mean_token_accuracy": 0.14785194993019105,
|
|
"num_tokens": 6214858.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 5.808193874359131,
|
|
"epoch": 0.18242396252556575,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00021358581450860186,
|
|
"loss": 5.6841,
|
|
"mean_token_accuracy": 0.14824768602848054,
|
|
"num_tokens": 6225934.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 5.768705892562866,
|
|
"epoch": 0.1827538431087946,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0002124532993880799,
|
|
"loss": 5.6534,
|
|
"mean_token_accuracy": 0.13919488862156867,
|
|
"num_tokens": 6237381.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 5.82682580947876,
|
|
"epoch": 0.1830837236920235,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00021132249901695044,
|
|
"loss": 5.7387,
|
|
"mean_token_accuracy": 0.13470394536852837,
|
|
"num_tokens": 6250303.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 5.799285888671875,
|
|
"epoch": 0.18341360427525236,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00021019344439667705,
|
|
"loss": 5.5909,
|
|
"mean_token_accuracy": 0.1465739220380783,
|
|
"num_tokens": 6261840.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 5.849857807159424,
|
|
"epoch": 0.18374348485848124,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00020906616648086213,
|
|
"loss": 5.7173,
|
|
"mean_token_accuracy": 0.135524670034647,
|
|
"num_tokens": 6273598.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 5.816228437423706,
|
|
"epoch": 0.1840733654417101,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00020794069617439942,
|
|
"loss": 5.6239,
|
|
"mean_token_accuracy": 0.1463257133960724,
|
|
"num_tokens": 6284449.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 5.808806133270264,
|
|
"epoch": 0.18440324602493896,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00020681706433262593,
|
|
"loss": 5.7394,
|
|
"mean_token_accuracy": 0.14109240621328353,
|
|
"num_tokens": 6297659.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 5.724093627929688,
|
|
"epoch": 0.18473312660816785,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00020569530176047602,
|
|
"loss": 5.6525,
|
|
"mean_token_accuracy": 0.14370780140161515,
|
|
"num_tokens": 6309173.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 5.742221260070801,
|
|
"epoch": 0.1850630071913967,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0002045754392116374,
|
|
"loss": 5.5771,
|
|
"mean_token_accuracy": 0.14463583007454872,
|
|
"num_tokens": 6320819.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 5.7456131935119625,
|
|
"epoch": 0.1853928877746256,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.00020345750738770757,
|
|
"loss": 5.5604,
|
|
"mean_token_accuracy": 0.14705618768930434,
|
|
"num_tokens": 6333224.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 5.737521696090698,
|
|
"epoch": 0.18572276835785445,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00020234153693735214,
|
|
"loss": 5.5795,
|
|
"mean_token_accuracy": 0.1456141196191311,
|
|
"num_tokens": 6343603.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 5.684789943695068,
|
|
"epoch": 0.18605264894108334,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0002012275584554647,
|
|
"loss": 5.5248,
|
|
"mean_token_accuracy": 0.1523616284132004,
|
|
"num_tokens": 6354478.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 5.718015909194946,
|
|
"epoch": 0.1863825295243122,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00020011560248232803,
|
|
"loss": 5.5919,
|
|
"mean_token_accuracy": 0.1455427460372448,
|
|
"num_tokens": 6364947.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 5.741596937179565,
|
|
"epoch": 0.18671241010754108,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00019900569950277692,
|
|
"loss": 5.5878,
|
|
"mean_token_accuracy": 0.14628631472587586,
|
|
"num_tokens": 6377153.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 5.788537979125977,
|
|
"epoch": 0.18704229069076994,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00019789787994536228,
|
|
"loss": 5.7735,
|
|
"mean_token_accuracy": 0.13262998983263968,
|
|
"num_tokens": 6389476.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 5.847185564041138,
|
|
"epoch": 0.1873721712739988,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00019679217418151667,
|
|
"loss": 5.7176,
|
|
"mean_token_accuracy": 0.13764084354043007,
|
|
"num_tokens": 6401723.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 5.813411617279053,
|
|
"epoch": 0.1877020518572277,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00019568861252472236,
|
|
"loss": 5.56,
|
|
"mean_token_accuracy": 0.14620804265141488,
|
|
"num_tokens": 6411224.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 5.800295782089234,
|
|
"epoch": 0.18803193244045655,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00019458722522967952,
|
|
"loss": 5.6814,
|
|
"mean_token_accuracy": 0.13616736084222794,
|
|
"num_tokens": 6422281.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 5.78706431388855,
|
|
"epoch": 0.18836181302368543,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00019348804249147723,
|
|
"loss": 5.6204,
|
|
"mean_token_accuracy": 0.14028710573911668,
|
|
"num_tokens": 6432698.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 5.769730186462402,
|
|
"epoch": 0.1886916936069143,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0001923910944447655,
|
|
"loss": 5.5375,
|
|
"mean_token_accuracy": 0.1447432279586792,
|
|
"num_tokens": 6442626.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 5.723214769363404,
|
|
"epoch": 0.18902157419014318,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00019129641116292928,
|
|
"loss": 5.596,
|
|
"mean_token_accuracy": 0.1439092352986336,
|
|
"num_tokens": 6454868.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 5.738249063491821,
|
|
"epoch": 0.18935145477337204,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.00019020402265726343,
|
|
"loss": 5.6403,
|
|
"mean_token_accuracy": 0.14119425863027574,
|
|
"num_tokens": 6467088.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 5.730573320388794,
|
|
"epoch": 0.18968133535660092,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0001891139588761509,
|
|
"loss": 5.5781,
|
|
"mean_token_accuracy": 0.14112994074821472,
|
|
"num_tokens": 6477767.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 5.7887030124664305,
|
|
"epoch": 0.19001121593982978,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00018802624970424076,
|
|
"loss": 5.6427,
|
|
"mean_token_accuracy": 0.14168466180562972,
|
|
"num_tokens": 6488815.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 5.70197377204895,
|
|
"epoch": 0.19034109652305864,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00018694092496162945,
|
|
"loss": 5.4975,
|
|
"mean_token_accuracy": 0.1480999082326889,
|
|
"num_tokens": 6499928.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 5.757118654251099,
|
|
"epoch": 0.19067097710628753,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00018585801440304306,
|
|
"loss": 5.6301,
|
|
"mean_token_accuracy": 0.13674165904521943,
|
|
"num_tokens": 6510164.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 5.8058709621429445,
|
|
"epoch": 0.19100085768951638,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00018477754771702165,
|
|
"loss": 5.683,
|
|
"mean_token_accuracy": 0.1374169819056988,
|
|
"num_tokens": 6521088.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 5.7437645435333256,
|
|
"epoch": 0.19133073827274527,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00018369955452510506,
|
|
"loss": 5.5827,
|
|
"mean_token_accuracy": 0.14729346483945846,
|
|
"num_tokens": 6531481.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 5.791945505142212,
|
|
"epoch": 0.19166061885597413,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0001826240643810212,
|
|
"loss": 5.5851,
|
|
"mean_token_accuracy": 0.14583178758621215,
|
|
"num_tokens": 6541548.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 5.7771087169647215,
|
|
"epoch": 0.19199049943920302,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0001815511067698758,
|
|
"loss": 5.7414,
|
|
"mean_token_accuracy": 0.14041516780853272,
|
|
"num_tokens": 6553587.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 5.769548606872559,
|
|
"epoch": 0.19232038002243187,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0001804807111073436,
|
|
"loss": 5.5819,
|
|
"mean_token_accuracy": 0.1399087689816952,
|
|
"num_tokens": 6565321.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 5.737796449661255,
|
|
"epoch": 0.19265026060566076,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0001794129067388625,
|
|
"loss": 5.5299,
|
|
"mean_token_accuracy": 0.14897729009389876,
|
|
"num_tokens": 6576844.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 5.794737768173218,
|
|
"epoch": 0.19298014118888962,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00017834772293882868,
|
|
"loss": 5.6701,
|
|
"mean_token_accuracy": 0.13966731429100038,
|
|
"num_tokens": 6588455.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 5.822253656387329,
|
|
"epoch": 0.1933100217721185,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000177285188909794,
|
|
"loss": 5.6892,
|
|
"mean_token_accuracy": 0.13955119699239732,
|
|
"num_tokens": 6599408.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 5.741854572296143,
|
|
"epoch": 0.19363990235534737,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0001762253337816656,
|
|
"loss": 5.582,
|
|
"mean_token_accuracy": 0.14925238639116287,
|
|
"num_tokens": 6610208.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 5.700358200073242,
|
|
"epoch": 0.19396978293857622,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00017516818661090738,
|
|
"loss": 5.5253,
|
|
"mean_token_accuracy": 0.1476726844906807,
|
|
"num_tokens": 6620762.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 5.723061800003052,
|
|
"epoch": 0.1942996635218051,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.0001741137763797428,
|
|
"loss": 5.546,
|
|
"mean_token_accuracy": 0.14653745144605637,
|
|
"num_tokens": 6633051.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 5.744390249252319,
|
|
"epoch": 0.19462954410503397,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00017306213199536115,
|
|
"loss": 5.6028,
|
|
"mean_token_accuracy": 0.14124337434768677,
|
|
"num_tokens": 6643665.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 5.732020044326783,
|
|
"epoch": 0.19495942468826286,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0001720132822891243,
|
|
"loss": 5.5268,
|
|
"mean_token_accuracy": 0.1446928083896637,
|
|
"num_tokens": 6654814.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 5.764629125595093,
|
|
"epoch": 0.19528930527149171,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0001709672560157769,
|
|
"loss": 5.5305,
|
|
"mean_token_accuracy": 0.15121119767427443,
|
|
"num_tokens": 6665865.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 5.808870458602906,
|
|
"epoch": 0.1956191858547206,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00016992408185265758,
|
|
"loss": 5.6064,
|
|
"mean_token_accuracy": 0.1421129435300827,
|
|
"num_tokens": 6676980.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 5.7312071323394775,
|
|
"epoch": 0.19594906643794946,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00016888378839891298,
|
|
"loss": 5.544,
|
|
"mean_token_accuracy": 0.14642035216093063,
|
|
"num_tokens": 6688301.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 5.707869338989258,
|
|
"epoch": 0.19627894702117835,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0001678464041747137,
|
|
"loss": 5.5622,
|
|
"mean_token_accuracy": 0.14264266714453697,
|
|
"num_tokens": 6699076.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 5.794270944595337,
|
|
"epoch": 0.1966088276044072,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00016681195762047223,
|
|
"loss": 5.6811,
|
|
"mean_token_accuracy": 0.14203905463218688,
|
|
"num_tokens": 6711229.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 5.749297475814819,
|
|
"epoch": 0.19693870818763606,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00016578047709606337,
|
|
"loss": 5.5874,
|
|
"mean_token_accuracy": 0.1446269765496254,
|
|
"num_tokens": 6723384.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 5.755163717269897,
|
|
"epoch": 0.19726858877086495,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00016475199088004678,
|
|
"loss": 5.6229,
|
|
"mean_token_accuracy": 0.14119048193097114,
|
|
"num_tokens": 6735310.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 5.708594226837159,
|
|
"epoch": 0.1975984693540938,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00016372652716889163,
|
|
"loss": 5.5733,
|
|
"mean_token_accuracy": 0.14752901643514632,
|
|
"num_tokens": 6745268.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 5.74776701927185,
|
|
"epoch": 0.1979283499373227,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0001627041140762035,
|
|
"loss": 5.6335,
|
|
"mean_token_accuracy": 0.1416826516389847,
|
|
"num_tokens": 6755062.0,
|
|
"step": 3000
|
|
}
|
|
],
|
|
"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": 1.1308033253376e+16,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|