2035 lines
54 KiB
JSON
2035 lines
54 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.05904233335301411,
|
|
"eval_steps": 500,
|
|
"global_step": 1000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.742573165893555,
|
|
"epoch": 0.00029521166676507054,
|
|
"grad_norm": 4.96875,
|
|
"learning_rate": 2e-06,
|
|
"loss": 10.7764,
|
|
"mean_token_accuracy": 0.00011173184029757976,
|
|
"num_tokens": 10275.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.74255199432373,
|
|
"epoch": 0.0005904233335301411,
|
|
"grad_norm": 5.4375,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 10.7784,
|
|
"mean_token_accuracy": 0.00011086474405601621,
|
|
"num_tokens": 22020.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.742577171325683,
|
|
"epoch": 0.0008856350002952116,
|
|
"grad_norm": 5.09375,
|
|
"learning_rate": 7e-06,
|
|
"loss": 10.7479,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 32612.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.742585563659668,
|
|
"epoch": 0.0011808466670602822,
|
|
"grad_norm": 5.125,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 10.692,
|
|
"mean_token_accuracy": 0.0002895977755542845,
|
|
"num_tokens": 43270.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.74253044128418,
|
|
"epoch": 0.0014760583338253527,
|
|
"grad_norm": 4.34375,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 10.5945,
|
|
"mean_token_accuracy": 0.0037050600280053914,
|
|
"num_tokens": 54178.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.742242050170898,
|
|
"epoch": 0.0017712700005904232,
|
|
"grad_norm": 4.15625,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 10.4602,
|
|
"mean_token_accuracy": 0.02422279641032219,
|
|
"num_tokens": 64390.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.741262912750244,
|
|
"epoch": 0.002066481667355494,
|
|
"grad_norm": 3.3125,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 10.2806,
|
|
"mean_token_accuracy": 0.041601650416851044,
|
|
"num_tokens": 73703.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.738601016998292,
|
|
"epoch": 0.0023616933341205643,
|
|
"grad_norm": 2.625,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 10.1485,
|
|
"mean_token_accuracy": 0.04163007400929928,
|
|
"num_tokens": 83463.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.734784793853759,
|
|
"epoch": 0.002656905000885635,
|
|
"grad_norm": 2.265625,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 10.0675,
|
|
"mean_token_accuracy": 0.04389659091830254,
|
|
"num_tokens": 94129.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.73138256072998,
|
|
"epoch": 0.0029521166676507054,
|
|
"grad_norm": 2.109375,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 9.9969,
|
|
"mean_token_accuracy": 0.043133416399359706,
|
|
"num_tokens": 105482.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.729689979553223,
|
|
"epoch": 0.003247328334415776,
|
|
"grad_norm": 2.0625,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.9129,
|
|
"mean_token_accuracy": 0.04309733100235462,
|
|
"num_tokens": 117342.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.728583812713623,
|
|
"epoch": 0.0035425400011808465,
|
|
"grad_norm": 1.859375,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.8878,
|
|
"mean_token_accuracy": 0.04118645377457142,
|
|
"num_tokens": 127691.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.727480697631837,
|
|
"epoch": 0.0038377516679459172,
|
|
"grad_norm": 1.90625,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.7761,
|
|
"mean_token_accuracy": 0.04004005677998066,
|
|
"num_tokens": 138186.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.725414276123047,
|
|
"epoch": 0.004132963334710988,
|
|
"grad_norm": 1.9375,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.7581,
|
|
"mean_token_accuracy": 0.0433019045740366,
|
|
"num_tokens": 149578.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.720968532562257,
|
|
"epoch": 0.004428175001476058,
|
|
"grad_norm": 1.9140625,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.6737,
|
|
"mean_token_accuracy": 0.04341540932655334,
|
|
"num_tokens": 160139.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.71543197631836,
|
|
"epoch": 0.004723386668241129,
|
|
"grad_norm": 1.90625,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.6176,
|
|
"mean_token_accuracy": 0.042911310121417044,
|
|
"num_tokens": 170768.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.708538818359376,
|
|
"epoch": 0.0050185983350062,
|
|
"grad_norm": 1.984375,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.5004,
|
|
"mean_token_accuracy": 0.04607887454330921,
|
|
"num_tokens": 180702.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.699075031280518,
|
|
"epoch": 0.00531381000177127,
|
|
"grad_norm": 1.9921875,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 9.4474,
|
|
"mean_token_accuracy": 0.04201225638389587,
|
|
"num_tokens": 191188.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.691750621795654,
|
|
"epoch": 0.0056090216685363405,
|
|
"grad_norm": 1.90625,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 9.381,
|
|
"mean_token_accuracy": 0.03792413603514433,
|
|
"num_tokens": 201094.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.6837984085083,
|
|
"epoch": 0.005904233335301411,
|
|
"grad_norm": 1.8984375,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 9.2614,
|
|
"mean_token_accuracy": 0.04629642106592655,
|
|
"num_tokens": 211718.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.667647075653075,
|
|
"epoch": 0.006199445002066482,
|
|
"grad_norm": 1.8828125,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 9.1409,
|
|
"mean_token_accuracy": 0.05333527401089668,
|
|
"num_tokens": 221446.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.640782737731934,
|
|
"epoch": 0.006494656668831552,
|
|
"grad_norm": 1.7578125,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 9.1093,
|
|
"mean_token_accuracy": 0.05261831916868687,
|
|
"num_tokens": 232848.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 10.602476406097413,
|
|
"epoch": 0.006789868335596623,
|
|
"grad_norm": 1.75,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 9.0166,
|
|
"mean_token_accuracy": 0.05226000994443893,
|
|
"num_tokens": 244008.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 10.5585186958313,
|
|
"epoch": 0.007085080002361693,
|
|
"grad_norm": 1.7265625,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.8449,
|
|
"mean_token_accuracy": 0.05914722830057144,
|
|
"num_tokens": 256211.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 10.501475620269776,
|
|
"epoch": 0.007380291669126764,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.7664,
|
|
"mean_token_accuracy": 0.05910652950406074,
|
|
"num_tokens": 266688.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 10.426000499725342,
|
|
"epoch": 0.0076755033358918345,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 8.57,
|
|
"mean_token_accuracy": 0.0610476765781641,
|
|
"num_tokens": 276930.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 10.339440727233887,
|
|
"epoch": 0.007970715002656906,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 8.4277,
|
|
"mean_token_accuracy": 0.07068866156041623,
|
|
"num_tokens": 287508.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 10.240792846679687,
|
|
"epoch": 0.008265926669421976,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 8.2713,
|
|
"mean_token_accuracy": 0.07111315764486789,
|
|
"num_tokens": 298021.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 10.135569763183593,
|
|
"epoch": 0.008561138336187046,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 8.1726,
|
|
"mean_token_accuracy": 0.07181189507246018,
|
|
"num_tokens": 308301.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 9.984276580810548,
|
|
"epoch": 0.008856350002952117,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 8.0427,
|
|
"mean_token_accuracy": 0.06862774454057216,
|
|
"num_tokens": 319321.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 9.829378795623779,
|
|
"epoch": 0.009151561669717187,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 8.0055,
|
|
"mean_token_accuracy": 0.07413771599531174,
|
|
"num_tokens": 330926.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 9.623138332366944,
|
|
"epoch": 0.009446773336482257,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 7.7929,
|
|
"mean_token_accuracy": 0.07890797778964043,
|
|
"num_tokens": 341133.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 9.416749095916748,
|
|
"epoch": 0.009741985003247328,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.7407,
|
|
"mean_token_accuracy": 0.07391801774501801,
|
|
"num_tokens": 351897.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 9.196506309509278,
|
|
"epoch": 0.0100371966700124,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.6663,
|
|
"mean_token_accuracy": 0.07125836983323097,
|
|
"num_tokens": 362644.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 8.932462692260742,
|
|
"epoch": 0.01033240833677747,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.628,
|
|
"mean_token_accuracy": 0.07178853973746299,
|
|
"num_tokens": 373902.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.703715133666993,
|
|
"epoch": 0.01062762000354254,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.4958,
|
|
"mean_token_accuracy": 0.08398130461573601,
|
|
"num_tokens": 384828.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.474815559387206,
|
|
"epoch": 0.01092283167030761,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.4684,
|
|
"mean_token_accuracy": 0.07413811013102531,
|
|
"num_tokens": 395103.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 8.360427570343017,
|
|
"epoch": 0.011218043337072681,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.374,
|
|
"mean_token_accuracy": 0.08157729879021644,
|
|
"num_tokens": 405283.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 8.21293478012085,
|
|
"epoch": 0.011513255003837751,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.3468,
|
|
"mean_token_accuracy": 0.0755577202886343,
|
|
"num_tokens": 416005.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 8.12078037261963,
|
|
"epoch": 0.011808466670602822,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.3518,
|
|
"mean_token_accuracy": 0.08057705461978912,
|
|
"num_tokens": 426547.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 8.019514322280884,
|
|
"epoch": 0.012103678337367892,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.2823,
|
|
"mean_token_accuracy": 0.08034192845225334,
|
|
"num_tokens": 436262.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 7.997368812561035,
|
|
"epoch": 0.012398890004132964,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.3018,
|
|
"mean_token_accuracy": 0.07936789095401764,
|
|
"num_tokens": 446899.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 7.968120145797729,
|
|
"epoch": 0.012694101670898034,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.3322,
|
|
"mean_token_accuracy": 0.07839082404971123,
|
|
"num_tokens": 458224.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 8.01854157447815,
|
|
"epoch": 0.012989313337663105,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.2688,
|
|
"mean_token_accuracy": 0.0764270156621933,
|
|
"num_tokens": 469306.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 7.9349792957305905,
|
|
"epoch": 0.013284525004428175,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.2614,
|
|
"mean_token_accuracy": 0.08024230971932411,
|
|
"num_tokens": 478685.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 7.891508388519287,
|
|
"epoch": 0.013579736671193245,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.1822,
|
|
"mean_token_accuracy": 0.0861165277659893,
|
|
"num_tokens": 489376.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 7.872423553466797,
|
|
"epoch": 0.013874948337958316,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.1724,
|
|
"mean_token_accuracy": 0.08608285039663315,
|
|
"num_tokens": 500260.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 7.763920402526855,
|
|
"epoch": 0.014170160004723386,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.1567,
|
|
"mean_token_accuracy": 0.08700917065143585,
|
|
"num_tokens": 511953.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.8764701843261715,
|
|
"epoch": 0.014465371671488458,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.3427,
|
|
"mean_token_accuracy": 0.07973492071032524,
|
|
"num_tokens": 523055.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.788686466217041,
|
|
"epoch": 0.014760583338253528,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.1424,
|
|
"mean_token_accuracy": 0.0809820182621479,
|
|
"num_tokens": 534337.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 7.773524141311645,
|
|
"epoch": 0.015055795005018599,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.1488,
|
|
"mean_token_accuracy": 0.08801065608859063,
|
|
"num_tokens": 545189.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.762392187118531,
|
|
"epoch": 0.015351006671783669,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.1789,
|
|
"mean_token_accuracy": 0.08976453244686126,
|
|
"num_tokens": 555506.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.752877712249756,
|
|
"epoch": 0.01564621833854874,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.0892,
|
|
"mean_token_accuracy": 0.08789646476507187,
|
|
"num_tokens": 565936.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.730583906173706,
|
|
"epoch": 0.01594143000531381,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 7.1826,
|
|
"mean_token_accuracy": 0.08930487185716629,
|
|
"num_tokens": 575691.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.710203313827515,
|
|
"epoch": 0.01623664167207888,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.1496,
|
|
"mean_token_accuracy": 0.08626075237989425,
|
|
"num_tokens": 586170.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.728785562515259,
|
|
"epoch": 0.016531853338843952,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 7.1356,
|
|
"mean_token_accuracy": 0.08597519621253014,
|
|
"num_tokens": 596458.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.711720895767212,
|
|
"epoch": 0.016827065005609022,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.0549,
|
|
"mean_token_accuracy": 0.08382375612854957,
|
|
"num_tokens": 606214.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.6635658740997314,
|
|
"epoch": 0.017122276672374093,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.1085,
|
|
"mean_token_accuracy": 0.08702588006854058,
|
|
"num_tokens": 617074.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.613530492782592,
|
|
"epoch": 0.017417488339139163,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.1582,
|
|
"mean_token_accuracy": 0.0854655608534813,
|
|
"num_tokens": 628057.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.640975618362427,
|
|
"epoch": 0.017712700005904233,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 7.0762,
|
|
"mean_token_accuracy": 0.0919909045100212,
|
|
"num_tokens": 638150.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.644310426712036,
|
|
"epoch": 0.018007911672669304,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000152,
|
|
"loss": 6.9574,
|
|
"mean_token_accuracy": 0.09665790945291519,
|
|
"num_tokens": 648377.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.570417594909668,
|
|
"epoch": 0.018303123339434374,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 7.0219,
|
|
"mean_token_accuracy": 0.09368904307484627,
|
|
"num_tokens": 658777.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.5545618534088135,
|
|
"epoch": 0.018598335006199444,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000157,
|
|
"loss": 7.1283,
|
|
"mean_token_accuracy": 0.08728318810462951,
|
|
"num_tokens": 669105.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.581419992446899,
|
|
"epoch": 0.018893546672964515,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 6.9931,
|
|
"mean_token_accuracy": 0.08857197985053063,
|
|
"num_tokens": 679294.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.5300342559814455,
|
|
"epoch": 0.019188758339729585,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.000162,
|
|
"loss": 6.9576,
|
|
"mean_token_accuracy": 0.09436112716794014,
|
|
"num_tokens": 690495.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.517716407775879,
|
|
"epoch": 0.019483970006494655,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 6.979,
|
|
"mean_token_accuracy": 0.09718031734228134,
|
|
"num_tokens": 701002.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.48517951965332,
|
|
"epoch": 0.019779181673259726,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 6.8828,
|
|
"mean_token_accuracy": 0.09275271818041801,
|
|
"num_tokens": 711488.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.405734252929688,
|
|
"epoch": 0.0200743933400248,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 6.9022,
|
|
"mean_token_accuracy": 0.09012737646698951,
|
|
"num_tokens": 722637.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.489488410949707,
|
|
"epoch": 0.02036960500678987,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 6.9705,
|
|
"mean_token_accuracy": 0.09417287185788155,
|
|
"num_tokens": 733212.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.45473804473877,
|
|
"epoch": 0.02066481667355494,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 6.9008,
|
|
"mean_token_accuracy": 0.09309542924165726,
|
|
"num_tokens": 743602.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.441885089874267,
|
|
"epoch": 0.02096002834032001,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000177,
|
|
"loss": 6.9506,
|
|
"mean_token_accuracy": 0.09587474688887596,
|
|
"num_tokens": 754094.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.430218076705932,
|
|
"epoch": 0.02125524000708508,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 6.8964,
|
|
"mean_token_accuracy": 0.09253331273794174,
|
|
"num_tokens": 763823.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.417900943756104,
|
|
"epoch": 0.02155045167385015,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000182,
|
|
"loss": 6.8424,
|
|
"mean_token_accuracy": 0.09944578483700753,
|
|
"num_tokens": 774128.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.384387731552124,
|
|
"epoch": 0.02184566334061522,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 6.913,
|
|
"mean_token_accuracy": 0.09629636481404305,
|
|
"num_tokens": 784292.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.428071117401123,
|
|
"epoch": 0.02214087500738029,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.000187,
|
|
"loss": 6.896,
|
|
"mean_token_accuracy": 0.09847814217209816,
|
|
"num_tokens": 795437.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.368625783920288,
|
|
"epoch": 0.022436086674145362,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 6.8038,
|
|
"mean_token_accuracy": 0.09553659558296204,
|
|
"num_tokens": 806543.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.34774341583252,
|
|
"epoch": 0.022731298340910432,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000192,
|
|
"loss": 6.9575,
|
|
"mean_token_accuracy": 0.09456263706088067,
|
|
"num_tokens": 817179.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.4415544986724855,
|
|
"epoch": 0.023026510007675503,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 6.8667,
|
|
"mean_token_accuracy": 0.09976130351424217,
|
|
"num_tokens": 826749.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.363486337661743,
|
|
"epoch": 0.023321721674440573,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 6.8927,
|
|
"mean_token_accuracy": 0.0938490979373455,
|
|
"num_tokens": 838807.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.37176923751831,
|
|
"epoch": 0.023616933341205643,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 6.8827,
|
|
"mean_token_accuracy": 0.09748194366693497,
|
|
"num_tokens": 849651.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.367251443862915,
|
|
"epoch": 0.023912145007970714,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000202,
|
|
"loss": 6.934,
|
|
"mean_token_accuracy": 0.09160904809832573,
|
|
"num_tokens": 859753.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.345250177383423,
|
|
"epoch": 0.024207356674735784,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 6.8134,
|
|
"mean_token_accuracy": 0.10118522644042968,
|
|
"num_tokens": 869904.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.267154836654663,
|
|
"epoch": 0.024502568341500858,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.000207,
|
|
"loss": 6.8152,
|
|
"mean_token_accuracy": 0.09964827746152878,
|
|
"num_tokens": 880948.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.350914382934571,
|
|
"epoch": 0.024797780008265928,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 6.8961,
|
|
"mean_token_accuracy": 0.09352625384926797,
|
|
"num_tokens": 892331.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.300309419631958,
|
|
"epoch": 0.025092991675031,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000212,
|
|
"loss": 6.731,
|
|
"mean_token_accuracy": 0.10073179304599762,
|
|
"num_tokens": 901691.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.247861719131469,
|
|
"epoch": 0.02538820334179607,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 6.7662,
|
|
"mean_token_accuracy": 0.0965045303106308,
|
|
"num_tokens": 912495.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.235934734344482,
|
|
"epoch": 0.02568341500856114,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 6.7406,
|
|
"mean_token_accuracy": 0.10061677768826485,
|
|
"num_tokens": 923420.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.153478145599365,
|
|
"epoch": 0.02597862667532621,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 6.7327,
|
|
"mean_token_accuracy": 0.10134619474411011,
|
|
"num_tokens": 933598.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.279621028900147,
|
|
"epoch": 0.02627383834209128,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000222,
|
|
"loss": 6.8611,
|
|
"mean_token_accuracy": 0.09994317889213562,
|
|
"num_tokens": 943229.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.256894063949585,
|
|
"epoch": 0.02656905000885635,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 6.7288,
|
|
"mean_token_accuracy": 0.09972334206104279,
|
|
"num_tokens": 954425.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.333712911605835,
|
|
"epoch": 0.02686426167562142,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 6.8322,
|
|
"mean_token_accuracy": 0.10051686838269233,
|
|
"num_tokens": 965516.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.109683799743652,
|
|
"epoch": 0.02715947334238649,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 6.7124,
|
|
"mean_token_accuracy": 0.10072791278362274,
|
|
"num_tokens": 976469.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.249271726608276,
|
|
"epoch": 0.02745468500915156,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 6.8029,
|
|
"mean_token_accuracy": 0.09443021714687347,
|
|
"num_tokens": 986946.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.233931303024292,
|
|
"epoch": 0.02774989667591663,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.6988,
|
|
"mean_token_accuracy": 0.10011776387691498,
|
|
"num_tokens": 997101.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.123860883712768,
|
|
"epoch": 0.0280451083426817,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.8063,
|
|
"mean_token_accuracy": 0.09995891153812408,
|
|
"num_tokens": 1007866.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.302121591567993,
|
|
"epoch": 0.028340320009446772,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.8209,
|
|
"mean_token_accuracy": 0.09563429430127143,
|
|
"num_tokens": 1019377.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.214538860321045,
|
|
"epoch": 0.028635531676211842,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.7985,
|
|
"mean_token_accuracy": 0.10173870101571084,
|
|
"num_tokens": 1031278.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.053379869461059,
|
|
"epoch": 0.028930743342976916,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.609,
|
|
"mean_token_accuracy": 0.10857169032096863,
|
|
"num_tokens": 1041663.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.141558599472046,
|
|
"epoch": 0.029225955009741986,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.68,
|
|
"mean_token_accuracy": 0.10325370877981185,
|
|
"num_tokens": 1052286.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.082495021820068,
|
|
"epoch": 0.029521166676507057,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 6.7424,
|
|
"mean_token_accuracy": 0.10430480614304542,
|
|
"num_tokens": 1061945.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.1918810367584225,
|
|
"epoch": 0.029816378343272127,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.7286,
|
|
"mean_token_accuracy": 0.09890020415186881,
|
|
"num_tokens": 1074057.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.1423930644989015,
|
|
"epoch": 0.030111590010037197,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.7168,
|
|
"mean_token_accuracy": 0.10386160910129547,
|
|
"num_tokens": 1084830.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.120156860351562,
|
|
"epoch": 0.030406801676802268,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.6608,
|
|
"mean_token_accuracy": 0.10988141670823097,
|
|
"num_tokens": 1095540.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.1051233291625975,
|
|
"epoch": 0.030702013343567338,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.6721,
|
|
"mean_token_accuracy": 0.10224077701568604,
|
|
"num_tokens": 1105395.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.0473850727081295,
|
|
"epoch": 0.03099722501033241,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.6362,
|
|
"mean_token_accuracy": 0.101447893679142,
|
|
"num_tokens": 1117007.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 7.1096114158630375,
|
|
"epoch": 0.03129243667709748,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.6533,
|
|
"mean_token_accuracy": 0.10817406699061394,
|
|
"num_tokens": 1128531.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.0557671070098875,
|
|
"epoch": 0.03158764834386255,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.6677,
|
|
"mean_token_accuracy": 0.10507503002882004,
|
|
"num_tokens": 1139194.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 7.096202421188354,
|
|
"epoch": 0.03188286001062762,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.7203,
|
|
"mean_token_accuracy": 0.09265467226505279,
|
|
"num_tokens": 1150371.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 7.058306312561035,
|
|
"epoch": 0.03217807167739269,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.6125,
|
|
"mean_token_accuracy": 0.10768099054694176,
|
|
"num_tokens": 1160840.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 7.038476371765137,
|
|
"epoch": 0.03247328334415776,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.6663,
|
|
"mean_token_accuracy": 0.10139054954051971,
|
|
"num_tokens": 1171438.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 7.088100433349609,
|
|
"epoch": 0.032768495010922834,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.6141,
|
|
"mean_token_accuracy": 0.10576796308159828,
|
|
"num_tokens": 1180889.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 6.985959911346436,
|
|
"epoch": 0.033063706677687904,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.6253,
|
|
"mean_token_accuracy": 0.1048286035656929,
|
|
"num_tokens": 1190755.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 7.112613868713379,
|
|
"epoch": 0.033358918344452974,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.696,
|
|
"mean_token_accuracy": 0.10307028964161873,
|
|
"num_tokens": 1202002.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 7.091805744171142,
|
|
"epoch": 0.033654130011218045,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.7127,
|
|
"mean_token_accuracy": 0.09907137230038643,
|
|
"num_tokens": 1213611.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 7.045700263977051,
|
|
"epoch": 0.033949341677983115,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.7282,
|
|
"mean_token_accuracy": 0.10283370316028595,
|
|
"num_tokens": 1224450.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 7.071837377548218,
|
|
"epoch": 0.034244553344748185,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.6784,
|
|
"mean_token_accuracy": 0.10268105193972588,
|
|
"num_tokens": 1235229.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 7.046443367004395,
|
|
"epoch": 0.034539765011513256,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.6526,
|
|
"mean_token_accuracy": 0.10178791806101799,
|
|
"num_tokens": 1245015.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 6.939273357391357,
|
|
"epoch": 0.034834976678278326,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.5344,
|
|
"mean_token_accuracy": 0.11150246858596802,
|
|
"num_tokens": 1256307.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 6.975376844406128,
|
|
"epoch": 0.035130188345043396,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.6105,
|
|
"mean_token_accuracy": 0.10918646082282066,
|
|
"num_tokens": 1267042.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 7.035576105117798,
|
|
"epoch": 0.03542540001180847,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.5985,
|
|
"mean_token_accuracy": 0.10955018848180771,
|
|
"num_tokens": 1277571.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 6.996139717102051,
|
|
"epoch": 0.03572061167857354,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.665,
|
|
"mean_token_accuracy": 0.10872082561254501,
|
|
"num_tokens": 1288203.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 6.967156648635864,
|
|
"epoch": 0.03601582334533861,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.6366,
|
|
"mean_token_accuracy": 0.10694377049803734,
|
|
"num_tokens": 1299510.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 6.97563099861145,
|
|
"epoch": 0.03631103501210368,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.5127,
|
|
"mean_token_accuracy": 0.11007486656308174,
|
|
"num_tokens": 1309573.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 6.906992053985595,
|
|
"epoch": 0.03660624667886875,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.5878,
|
|
"mean_token_accuracy": 0.10692002549767494,
|
|
"num_tokens": 1321104.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 6.917443513870239,
|
|
"epoch": 0.03690145834563382,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.5916,
|
|
"mean_token_accuracy": 0.10439730063080788,
|
|
"num_tokens": 1332671.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 6.861900806427002,
|
|
"epoch": 0.03719667001239889,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.4747,
|
|
"mean_token_accuracy": 0.11168068125844002,
|
|
"num_tokens": 1344416.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 7.0240528106689455,
|
|
"epoch": 0.03749188167916396,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.606,
|
|
"mean_token_accuracy": 0.10131711885333061,
|
|
"num_tokens": 1355740.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 6.917941236495972,
|
|
"epoch": 0.03778709334592903,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.5638,
|
|
"mean_token_accuracy": 0.1071008414030075,
|
|
"num_tokens": 1366465.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 6.809306955337524,
|
|
"epoch": 0.0380823050126941,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.4827,
|
|
"mean_token_accuracy": 0.10806326791644097,
|
|
"num_tokens": 1376476.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 7.007070827484131,
|
|
"epoch": 0.03837751667945917,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.4837,
|
|
"mean_token_accuracy": 0.1201641671359539,
|
|
"num_tokens": 1386172.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 6.835911703109741,
|
|
"epoch": 0.03867272834622424,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.4762,
|
|
"mean_token_accuracy": 0.1076883353292942,
|
|
"num_tokens": 1396655.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 6.812751722335816,
|
|
"epoch": 0.03896794001298931,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.4853,
|
|
"mean_token_accuracy": 0.11088723838329315,
|
|
"num_tokens": 1407816.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 6.715743827819824,
|
|
"epoch": 0.03926315167975438,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.4645,
|
|
"mean_token_accuracy": 0.11460546851158142,
|
|
"num_tokens": 1418265.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 6.841448259353638,
|
|
"epoch": 0.03955836334651945,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.449,
|
|
"mean_token_accuracy": 0.1139136053621769,
|
|
"num_tokens": 1429776.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 6.8621501445770265,
|
|
"epoch": 0.03985357501328453,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.475,
|
|
"mean_token_accuracy": 0.11356844678521157,
|
|
"num_tokens": 1440295.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 6.859308671951294,
|
|
"epoch": 0.0401487866800496,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.5194,
|
|
"mean_token_accuracy": 0.11261903643608093,
|
|
"num_tokens": 1450505.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 6.922615432739258,
|
|
"epoch": 0.04044399834681467,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.6088,
|
|
"mean_token_accuracy": 0.10697626769542694,
|
|
"num_tokens": 1461768.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 6.882122468948364,
|
|
"epoch": 0.04073921001357974,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.5527,
|
|
"mean_token_accuracy": 0.1090085543692112,
|
|
"num_tokens": 1473965.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 6.839772033691406,
|
|
"epoch": 0.04103442168034481,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.4935,
|
|
"mean_token_accuracy": 0.1170326255261898,
|
|
"num_tokens": 1484750.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 6.841724634170532,
|
|
"epoch": 0.04132963334710988,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.4818,
|
|
"mean_token_accuracy": 0.10687893703579902,
|
|
"num_tokens": 1496474.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 6.801630210876465,
|
|
"epoch": 0.04162484501387495,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.6095,
|
|
"mean_token_accuracy": 0.10371644273400307,
|
|
"num_tokens": 1507880.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 6.8716658592224125,
|
|
"epoch": 0.04192005668064002,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.5201,
|
|
"mean_token_accuracy": 0.11054393872618676,
|
|
"num_tokens": 1519115.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 6.830336380004883,
|
|
"epoch": 0.04221526834740509,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.4674,
|
|
"mean_token_accuracy": 0.11907401829957961,
|
|
"num_tokens": 1529031.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 6.835891342163086,
|
|
"epoch": 0.04251048001417016,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.4217,
|
|
"mean_token_accuracy": 0.10899701043963432,
|
|
"num_tokens": 1539954.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 6.782188510894775,
|
|
"epoch": 0.04280569168093523,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.5934,
|
|
"mean_token_accuracy": 0.10825273916125297,
|
|
"num_tokens": 1550713.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 6.784424448013306,
|
|
"epoch": 0.0431009033477003,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.4856,
|
|
"mean_token_accuracy": 0.1092301182448864,
|
|
"num_tokens": 1562569.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 6.857142400741577,
|
|
"epoch": 0.04339611501446537,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.5065,
|
|
"mean_token_accuracy": 0.12176962345838546,
|
|
"num_tokens": 1572999.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.821650075912475,
|
|
"epoch": 0.04369132668123044,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.5002,
|
|
"mean_token_accuracy": 0.10788078010082244,
|
|
"num_tokens": 1583560.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 6.770935106277466,
|
|
"epoch": 0.04398653834799551,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.4037,
|
|
"mean_token_accuracy": 0.1184255413711071,
|
|
"num_tokens": 1593931.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.738411855697632,
|
|
"epoch": 0.04428175001476058,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.4744,
|
|
"mean_token_accuracy": 0.11096556782722473,
|
|
"num_tokens": 1605182.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 6.793981218338013,
|
|
"epoch": 0.044576961681525654,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.4367,
|
|
"mean_token_accuracy": 0.11491051763296127,
|
|
"num_tokens": 1615912.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 6.664726591110229,
|
|
"epoch": 0.044872173348290724,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.4296,
|
|
"mean_token_accuracy": 0.11536637470126151,
|
|
"num_tokens": 1626322.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.80378909111023,
|
|
"epoch": 0.045167385015055794,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.4088,
|
|
"mean_token_accuracy": 0.11972564309835435,
|
|
"num_tokens": 1636859.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.7000233173370365,
|
|
"epoch": 0.045462596681820865,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.4417,
|
|
"mean_token_accuracy": 0.11232743039727211,
|
|
"num_tokens": 1648511.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.739704561233521,
|
|
"epoch": 0.045757808348585935,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.4156,
|
|
"mean_token_accuracy": 0.11628528609871865,
|
|
"num_tokens": 1659350.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.675900602340699,
|
|
"epoch": 0.046053020015351005,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.4359,
|
|
"mean_token_accuracy": 0.10967965424060822,
|
|
"num_tokens": 1670077.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.777168655395508,
|
|
"epoch": 0.046348231682116076,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.4389,
|
|
"mean_token_accuracy": 0.11821731701493263,
|
|
"num_tokens": 1681877.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.737879991531372,
|
|
"epoch": 0.046643443348881146,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.4192,
|
|
"mean_token_accuracy": 0.11788794547319412,
|
|
"num_tokens": 1692448.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.714100122451782,
|
|
"epoch": 0.046938655015646216,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.4915,
|
|
"mean_token_accuracy": 0.11439365297555923,
|
|
"num_tokens": 1702437.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.764936017990112,
|
|
"epoch": 0.04723386668241129,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.3825,
|
|
"mean_token_accuracy": 0.1186959095299244,
|
|
"num_tokens": 1712587.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.780900764465332,
|
|
"epoch": 0.04752907834917636,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.4733,
|
|
"mean_token_accuracy": 0.1111880011856556,
|
|
"num_tokens": 1722782.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.784157133102417,
|
|
"epoch": 0.04782429001594143,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.4429,
|
|
"mean_token_accuracy": 0.11309729292988777,
|
|
"num_tokens": 1733993.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.661256790161133,
|
|
"epoch": 0.0481195016827065,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.4319,
|
|
"mean_token_accuracy": 0.12013896554708481,
|
|
"num_tokens": 1744095.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.627638578414917,
|
|
"epoch": 0.04841471334947157,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.3862,
|
|
"mean_token_accuracy": 0.11798101216554642,
|
|
"num_tokens": 1754356.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.718645000457764,
|
|
"epoch": 0.048709925016236645,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.4632,
|
|
"mean_token_accuracy": 0.11508424431085587,
|
|
"num_tokens": 1765628.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.697535753250122,
|
|
"epoch": 0.049005136683001715,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.3341,
|
|
"mean_token_accuracy": 0.11765867844223976,
|
|
"num_tokens": 1775697.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.703864526748657,
|
|
"epoch": 0.049300348349766786,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.4334,
|
|
"mean_token_accuracy": 0.11458561718463897,
|
|
"num_tokens": 1786066.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.613457870483399,
|
|
"epoch": 0.049595560016531856,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.2576,
|
|
"mean_token_accuracy": 0.12075778916478157,
|
|
"num_tokens": 1796735.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.7064070224761965,
|
|
"epoch": 0.049890771683296926,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.4148,
|
|
"mean_token_accuracy": 0.11420085206627846,
|
|
"num_tokens": 1807164.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.664913654327393,
|
|
"epoch": 0.050185983350062,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.4446,
|
|
"mean_token_accuracy": 0.11588526219129562,
|
|
"num_tokens": 1818788.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.706023120880127,
|
|
"epoch": 0.05048119501682707,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.4336,
|
|
"mean_token_accuracy": 0.11620423197746277,
|
|
"num_tokens": 1830488.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.5659849643707275,
|
|
"epoch": 0.05077640668359214,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.3689,
|
|
"mean_token_accuracy": 0.11279175132513046,
|
|
"num_tokens": 1841085.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.62478985786438,
|
|
"epoch": 0.05107161835035721,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.3439,
|
|
"mean_token_accuracy": 0.11958979964256286,
|
|
"num_tokens": 1852831.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.638272142410278,
|
|
"epoch": 0.05136683001712228,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.3074,
|
|
"mean_token_accuracy": 0.12059865668416023,
|
|
"num_tokens": 1863868.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.604048776626587,
|
|
"epoch": 0.05166204168388735,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.2958,
|
|
"mean_token_accuracy": 0.11414773985743523,
|
|
"num_tokens": 1874227.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.589703607559204,
|
|
"epoch": 0.05195725335065242,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.2993,
|
|
"mean_token_accuracy": 0.12205377146601677,
|
|
"num_tokens": 1883712.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.678598642349243,
|
|
"epoch": 0.05225246501741749,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.4011,
|
|
"mean_token_accuracy": 0.11998820379376411,
|
|
"num_tokens": 1894909.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.598682498931884,
|
|
"epoch": 0.05254767668418256,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.3208,
|
|
"mean_token_accuracy": 0.1276823714375496,
|
|
"num_tokens": 1904973.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.58713927268982,
|
|
"epoch": 0.05284288835094763,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.29,
|
|
"mean_token_accuracy": 0.11684366688132286,
|
|
"num_tokens": 1915131.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.627308464050293,
|
|
"epoch": 0.0531381000177127,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.3868,
|
|
"mean_token_accuracy": 0.11667721047997474,
|
|
"num_tokens": 1925893.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.6627851009368895,
|
|
"epoch": 0.05343331168447777,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.3972,
|
|
"mean_token_accuracy": 0.11407171338796615,
|
|
"num_tokens": 1936830.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.569732666015625,
|
|
"epoch": 0.05372852335124284,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.3268,
|
|
"mean_token_accuracy": 0.12334559336304665,
|
|
"num_tokens": 1948291.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.645078706741333,
|
|
"epoch": 0.05402373501800791,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.3853,
|
|
"mean_token_accuracy": 0.11478180214762687,
|
|
"num_tokens": 1959168.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.565263366699218,
|
|
"epoch": 0.05431894668477298,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.3516,
|
|
"mean_token_accuracy": 0.12495195642113685,
|
|
"num_tokens": 1970046.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.566036319732666,
|
|
"epoch": 0.05461415835153805,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.3371,
|
|
"mean_token_accuracy": 0.12042524293065071,
|
|
"num_tokens": 1980430.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.663130235671997,
|
|
"epoch": 0.05490937001830312,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.3409,
|
|
"mean_token_accuracy": 0.12344788387417793,
|
|
"num_tokens": 1991035.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.647854423522949,
|
|
"epoch": 0.05520458168506819,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.3706,
|
|
"mean_token_accuracy": 0.1162357397377491,
|
|
"num_tokens": 2002122.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.557170581817627,
|
|
"epoch": 0.05549979335183326,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.3101,
|
|
"mean_token_accuracy": 0.11861225590109825,
|
|
"num_tokens": 2012676.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.556992673873902,
|
|
"epoch": 0.05579500501859833,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.328,
|
|
"mean_token_accuracy": 0.11912066489458084,
|
|
"num_tokens": 2023124.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.683170127868652,
|
|
"epoch": 0.0560902166853634,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.3176,
|
|
"mean_token_accuracy": 0.11822493821382522,
|
|
"num_tokens": 2034030.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.448392868041992,
|
|
"epoch": 0.056385428352128474,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.3329,
|
|
"mean_token_accuracy": 0.11704078093171119,
|
|
"num_tokens": 2044499.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.59168119430542,
|
|
"epoch": 0.056680640018893544,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.3684,
|
|
"mean_token_accuracy": 0.11949558109045029,
|
|
"num_tokens": 2055242.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.52003698348999,
|
|
"epoch": 0.056975851685658614,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.3533,
|
|
"mean_token_accuracy": 0.11890814751386643,
|
|
"num_tokens": 2065987.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.531438779830933,
|
|
"epoch": 0.057271063352423685,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.2626,
|
|
"mean_token_accuracy": 0.12552469298243524,
|
|
"num_tokens": 2077015.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.661432361602783,
|
|
"epoch": 0.05756627501918876,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.3877,
|
|
"mean_token_accuracy": 0.12421264126896858,
|
|
"num_tokens": 2087411.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.554572677612304,
|
|
"epoch": 0.05786148668595383,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.2903,
|
|
"mean_token_accuracy": 0.12133480757474899,
|
|
"num_tokens": 2097612.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.477239942550659,
|
|
"epoch": 0.0581566983527189,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.2478,
|
|
"mean_token_accuracy": 0.12306424006819724,
|
|
"num_tokens": 2108118.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.513996744155884,
|
|
"epoch": 0.05845191001948397,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.3221,
|
|
"mean_token_accuracy": 0.12351070642471314,
|
|
"num_tokens": 2118262.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.5929190158844,
|
|
"epoch": 0.05874712168624904,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.3173,
|
|
"mean_token_accuracy": 0.12096539437770844,
|
|
"num_tokens": 2128819.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.52128791809082,
|
|
"epoch": 0.05904233335301411,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.2396,
|
|
"mean_token_accuracy": 0.12479885071516036,
|
|
"num_tokens": 2139852.0,
|
|
"step": 1000
|
|
}
|
|
],
|
|
"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": 464442284113920.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|