6101 lines
165 KiB
JSON
6101 lines
165 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 3.243374797187669,
|
|
"eval_steps": 500,
|
|
"global_step": 3000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 4.7960851192474365,
|
|
"epoch": 0.005408328826392645,
|
|
"grad_norm": 27.25,
|
|
"learning_rate": 2e-06,
|
|
"loss": 14.2304,
|
|
"mean_token_accuracy": 0.0003160792635753751,
|
|
"num_tokens": 10942.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 4.8001548767089846,
|
|
"epoch": 0.01081665765278529,
|
|
"grad_norm": 34.0,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 14.2056,
|
|
"mean_token_accuracy": 0.0001242236001417041,
|
|
"num_tokens": 21887.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 5.043719053268433,
|
|
"epoch": 0.016224986479177934,
|
|
"grad_norm": 36.5,
|
|
"learning_rate": 7e-06,
|
|
"loss": 14.0392,
|
|
"mean_token_accuracy": 6.711409660056233e-05,
|
|
"num_tokens": 34345.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 5.49084243774414,
|
|
"epoch": 0.02163331530557058,
|
|
"grad_norm": 34.75,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 13.6544,
|
|
"mean_token_accuracy": 9.285050909966231e-05,
|
|
"num_tokens": 45445.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 7.0108997344970705,
|
|
"epoch": 0.02704164413196322,
|
|
"grad_norm": 19.25,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 12.8774,
|
|
"mean_token_accuracy": 0.00023736039875075222,
|
|
"num_tokens": 58237.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 8.019237947463989,
|
|
"epoch": 0.03244997295835587,
|
|
"grad_norm": 9.3125,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 12.3405,
|
|
"mean_token_accuracy": 9.813542710617185e-05,
|
|
"num_tokens": 68098.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 9.054293632507324,
|
|
"epoch": 0.03785830178474851,
|
|
"grad_norm": 5.46875,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 11.9463,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 79867.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 9.511068725585938,
|
|
"epoch": 0.04326663061114116,
|
|
"grad_norm": 3.796875,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 11.6959,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 91257.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 9.867530632019044,
|
|
"epoch": 0.048674959437533805,
|
|
"grad_norm": 5.53125,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 11.45,
|
|
"mean_token_accuracy": 9.38967161346227e-05,
|
|
"num_tokens": 103474.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.309116744995118,
|
|
"epoch": 0.05408328826392644,
|
|
"grad_norm": 4.5,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 11.0292,
|
|
"mean_token_accuracy": 0.0003331616928335279,
|
|
"num_tokens": 114830.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.523942279815675,
|
|
"epoch": 0.05949161709031909,
|
|
"grad_norm": 3.921875,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 10.774,
|
|
"mean_token_accuracy": 0.011405656964052468,
|
|
"num_tokens": 126017.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.591793823242188,
|
|
"epoch": 0.06489994591671173,
|
|
"grad_norm": 2.734375,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 10.5283,
|
|
"mean_token_accuracy": 0.03312199003994465,
|
|
"num_tokens": 138509.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.653960514068604,
|
|
"epoch": 0.07030827474310439,
|
|
"grad_norm": 2.375,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 10.3296,
|
|
"mean_token_accuracy": 0.032539681531488895,
|
|
"num_tokens": 148353.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.66145372390747,
|
|
"epoch": 0.07571660356949703,
|
|
"grad_norm": 2.140625,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 10.2093,
|
|
"mean_token_accuracy": 0.031699617952108385,
|
|
"num_tokens": 160269.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.610020542144776,
|
|
"epoch": 0.08112493239588967,
|
|
"grad_norm": 2.0,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 10.0327,
|
|
"mean_token_accuracy": 0.03554317634552717,
|
|
"num_tokens": 171568.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.5712140083313,
|
|
"epoch": 0.08653326122228232,
|
|
"grad_norm": 1.984375,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.8889,
|
|
"mean_token_accuracy": 0.03888488262891769,
|
|
"num_tokens": 184243.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.565914058685303,
|
|
"epoch": 0.09194159004867496,
|
|
"grad_norm": 1.96875,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.7667,
|
|
"mean_token_accuracy": 0.03745253346860409,
|
|
"num_tokens": 194887.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.57132396697998,
|
|
"epoch": 0.09734991887506761,
|
|
"grad_norm": 1.9296875,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 9.6353,
|
|
"mean_token_accuracy": 0.03890889976173639,
|
|
"num_tokens": 206476.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.575814723968506,
|
|
"epoch": 0.10275824770146025,
|
|
"grad_norm": 1.8671875,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 9.4592,
|
|
"mean_token_accuracy": 0.03984561040997505,
|
|
"num_tokens": 216578.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.575993537902832,
|
|
"epoch": 0.10816657652785289,
|
|
"grad_norm": 1.8515625,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 9.3575,
|
|
"mean_token_accuracy": 0.04156542681157589,
|
|
"num_tokens": 226941.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.572124099731445,
|
|
"epoch": 0.11357490535424554,
|
|
"grad_norm": 1.828125,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 9.2249,
|
|
"mean_token_accuracy": 0.03909948579967022,
|
|
"num_tokens": 238060.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.561905097961425,
|
|
"epoch": 0.11898323418063818,
|
|
"grad_norm": 1.9453125,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 9.0694,
|
|
"mean_token_accuracy": 0.04209643751382828,
|
|
"num_tokens": 249137.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 10.530157947540284,
|
|
"epoch": 0.12439156300703083,
|
|
"grad_norm": 1.828125,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 8.9517,
|
|
"mean_token_accuracy": 0.042229291051626205,
|
|
"num_tokens": 260712.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 10.49021110534668,
|
|
"epoch": 0.12979989183342347,
|
|
"grad_norm": 1.84375,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.7898,
|
|
"mean_token_accuracy": 0.0394113190472126,
|
|
"num_tokens": 271039.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 10.455241107940674,
|
|
"epoch": 0.1352082206598161,
|
|
"grad_norm": 1.921875,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.6467,
|
|
"mean_token_accuracy": 0.040281732007861136,
|
|
"num_tokens": 281353.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 10.391654300689698,
|
|
"epoch": 0.14061654948620878,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 8.5617,
|
|
"mean_token_accuracy": 0.04575536660850048,
|
|
"num_tokens": 292221.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 10.302424240112305,
|
|
"epoch": 0.1460248783126014,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 8.4303,
|
|
"mean_token_accuracy": 0.0425492487847805,
|
|
"num_tokens": 303353.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 10.208961296081544,
|
|
"epoch": 0.15143320713899405,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 8.3137,
|
|
"mean_token_accuracy": 0.04581625089049339,
|
|
"num_tokens": 314165.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 10.080995559692383,
|
|
"epoch": 0.1568415359653867,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 8.1479,
|
|
"mean_token_accuracy": 0.049863246828317644,
|
|
"num_tokens": 326387.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 9.93353729248047,
|
|
"epoch": 0.16224986479177933,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 8.0989,
|
|
"mean_token_accuracy": 0.04789762534201145,
|
|
"num_tokens": 339155.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 9.778404998779298,
|
|
"epoch": 0.167658193618172,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 7.9369,
|
|
"mean_token_accuracy": 0.04837009124457836,
|
|
"num_tokens": 350643.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 9.568048095703125,
|
|
"epoch": 0.17306652244456464,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 7.8363,
|
|
"mean_token_accuracy": 0.04552422165870666,
|
|
"num_tokens": 361977.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 9.345962619781494,
|
|
"epoch": 0.17847485127095727,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.8056,
|
|
"mean_token_accuracy": 0.048871561884880066,
|
|
"num_tokens": 373604.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 9.073436737060547,
|
|
"epoch": 0.1838831800973499,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.7117,
|
|
"mean_token_accuracy": 0.04806899577379227,
|
|
"num_tokens": 384306.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 8.822536659240722,
|
|
"epoch": 0.18929150892374255,
|
|
"grad_norm": 0.77734375,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.5951,
|
|
"mean_token_accuracy": 0.04366670176386833,
|
|
"num_tokens": 396479.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.54340581893921,
|
|
"epoch": 0.19469983775013522,
|
|
"grad_norm": 0.74609375,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.5166,
|
|
"mean_token_accuracy": 0.05379514247179031,
|
|
"num_tokens": 408489.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.377008247375489,
|
|
"epoch": 0.20010816657652786,
|
|
"grad_norm": 0.671875,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.5106,
|
|
"mean_token_accuracy": 0.06423841267824174,
|
|
"num_tokens": 419339.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 8.208265113830567,
|
|
"epoch": 0.2055164954029205,
|
|
"grad_norm": 0.67578125,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.4712,
|
|
"mean_token_accuracy": 0.06963393315672875,
|
|
"num_tokens": 430486.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 8.0866042137146,
|
|
"epoch": 0.21092482422931313,
|
|
"grad_norm": 0.75,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.4643,
|
|
"mean_token_accuracy": 0.06951842345297336,
|
|
"num_tokens": 440414.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 7.999132299423218,
|
|
"epoch": 0.21633315305570577,
|
|
"grad_norm": 0.6640625,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.4064,
|
|
"mean_token_accuracy": 0.07127611935138703,
|
|
"num_tokens": 451475.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 7.92455677986145,
|
|
"epoch": 0.22174148188209844,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.4544,
|
|
"mean_token_accuracy": 0.0808640293776989,
|
|
"num_tokens": 462053.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 7.913168287277221,
|
|
"epoch": 0.22714981070849108,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.4165,
|
|
"mean_token_accuracy": 0.07268866151571274,
|
|
"num_tokens": 473207.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 7.920434427261353,
|
|
"epoch": 0.23255813953488372,
|
|
"grad_norm": 0.75,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.3957,
|
|
"mean_token_accuracy": 0.07367360889911652,
|
|
"num_tokens": 484999.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 7.8919206142425535,
|
|
"epoch": 0.23796646836127636,
|
|
"grad_norm": 0.76171875,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.3913,
|
|
"mean_token_accuracy": 0.07248864844441413,
|
|
"num_tokens": 496772.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 7.9051414966583256,
|
|
"epoch": 0.24337479718766902,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.3555,
|
|
"mean_token_accuracy": 0.07366742193698883,
|
|
"num_tokens": 507782.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 7.845500946044922,
|
|
"epoch": 0.24878312601406166,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.2834,
|
|
"mean_token_accuracy": 0.07393435426056386,
|
|
"num_tokens": 518950.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 7.782898235321045,
|
|
"epoch": 0.2541914548404543,
|
|
"grad_norm": 0.796875,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.3481,
|
|
"mean_token_accuracy": 0.07736928686499596,
|
|
"num_tokens": 530795.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 7.807211065292359,
|
|
"epoch": 0.25959978366684694,
|
|
"grad_norm": 0.80859375,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.3554,
|
|
"mean_token_accuracy": 0.07271409779787064,
|
|
"num_tokens": 542878.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.8081248760223385,
|
|
"epoch": 0.2650081124932396,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.2632,
|
|
"mean_token_accuracy": 0.07816940769553185,
|
|
"num_tokens": 553682.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.717547512054443,
|
|
"epoch": 0.2704164413196322,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.2621,
|
|
"mean_token_accuracy": 0.08631859943270684,
|
|
"num_tokens": 564433.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 7.729435443878174,
|
|
"epoch": 0.27582477014602486,
|
|
"grad_norm": 0.7109375,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.28,
|
|
"mean_token_accuracy": 0.0831921637058258,
|
|
"num_tokens": 576001.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.762239027023315,
|
|
"epoch": 0.28123309897241755,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.2116,
|
|
"mean_token_accuracy": 0.08321145027875901,
|
|
"num_tokens": 587138.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.712756729125976,
|
|
"epoch": 0.2866414277988102,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.1655,
|
|
"mean_token_accuracy": 0.08543640077114105,
|
|
"num_tokens": 597572.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.693479919433594,
|
|
"epoch": 0.2920497566252028,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 7.1296,
|
|
"mean_token_accuracy": 0.09074783772230148,
|
|
"num_tokens": 608879.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.604578304290771,
|
|
"epoch": 0.29745808545159547,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.1746,
|
|
"mean_token_accuracy": 0.08316268846392631,
|
|
"num_tokens": 620439.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.675649356842041,
|
|
"epoch": 0.3028664142779881,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 7.2274,
|
|
"mean_token_accuracy": 0.0821447916328907,
|
|
"num_tokens": 632543.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.693741321563721,
|
|
"epoch": 0.30827474310438074,
|
|
"grad_norm": 0.75,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.258,
|
|
"mean_token_accuracy": 0.09081905633211136,
|
|
"num_tokens": 643369.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.64626145362854,
|
|
"epoch": 0.3136830719307734,
|
|
"grad_norm": 0.68359375,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.1271,
|
|
"mean_token_accuracy": 0.08910127580165864,
|
|
"num_tokens": 655328.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.68002290725708,
|
|
"epoch": 0.319091400757166,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.1774,
|
|
"mean_token_accuracy": 0.08911841437220573,
|
|
"num_tokens": 665780.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.583356475830078,
|
|
"epoch": 0.32449972958355866,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 7.0876,
|
|
"mean_token_accuracy": 0.09305166527628898,
|
|
"num_tokens": 676566.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.611654996871948,
|
|
"epoch": 0.3299080584099513,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.000152,
|
|
"loss": 7.0847,
|
|
"mean_token_accuracy": 0.09290368705987931,
|
|
"num_tokens": 687407.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.593765163421631,
|
|
"epoch": 0.335316387236344,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 7.0044,
|
|
"mean_token_accuracy": 0.09596289992332459,
|
|
"num_tokens": 698162.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.452439403533935,
|
|
"epoch": 0.34072471606273663,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.000157,
|
|
"loss": 7.0018,
|
|
"mean_token_accuracy": 0.08815811201930046,
|
|
"num_tokens": 708743.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.615065431594848,
|
|
"epoch": 0.34613304488912927,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 7.1876,
|
|
"mean_token_accuracy": 0.09327791407704353,
|
|
"num_tokens": 719828.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.63629150390625,
|
|
"epoch": 0.3515413737155219,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.000162,
|
|
"loss": 7.149,
|
|
"mean_token_accuracy": 0.08833655528724194,
|
|
"num_tokens": 732514.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.527328824996948,
|
|
"epoch": 0.35694970254191455,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 6.9889,
|
|
"mean_token_accuracy": 0.09440577700734139,
|
|
"num_tokens": 742391.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.36641492843628,
|
|
"epoch": 0.3623580313683072,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 6.9959,
|
|
"mean_token_accuracy": 0.08885396271944046,
|
|
"num_tokens": 753638.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.65393443107605,
|
|
"epoch": 0.3677663601946998,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 7.1896,
|
|
"mean_token_accuracy": 0.09420613124966622,
|
|
"num_tokens": 764850.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.468115282058716,
|
|
"epoch": 0.37317468902109246,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 6.9635,
|
|
"mean_token_accuracy": 0.10290685519576073,
|
|
"num_tokens": 776192.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.411421632766723,
|
|
"epoch": 0.3785830178474851,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 6.9167,
|
|
"mean_token_accuracy": 0.10349498242139817,
|
|
"num_tokens": 787051.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.389636278152466,
|
|
"epoch": 0.3839913466738778,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000177,
|
|
"loss": 6.9474,
|
|
"mean_token_accuracy": 0.09752821102738381,
|
|
"num_tokens": 799281.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.4360168933868405,
|
|
"epoch": 0.38939967550027044,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 6.9509,
|
|
"mean_token_accuracy": 0.09831701442599297,
|
|
"num_tokens": 810003.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.410728406906128,
|
|
"epoch": 0.3948080043266631,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000182,
|
|
"loss": 6.9213,
|
|
"mean_token_accuracy": 0.09893173575401307,
|
|
"num_tokens": 819749.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.420523643493652,
|
|
"epoch": 0.4002163331530557,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 6.9334,
|
|
"mean_token_accuracy": 0.10455422475934029,
|
|
"num_tokens": 830552.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.477943468093872,
|
|
"epoch": 0.40562466197944835,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000187,
|
|
"loss": 7.0213,
|
|
"mean_token_accuracy": 0.10054472163319587,
|
|
"num_tokens": 840495.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.325655174255371,
|
|
"epoch": 0.411032990805841,
|
|
"grad_norm": 0.8203125,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 6.9394,
|
|
"mean_token_accuracy": 0.10164839625358582,
|
|
"num_tokens": 852915.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.386038827896118,
|
|
"epoch": 0.41644131963223363,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.000192,
|
|
"loss": 6.9133,
|
|
"mean_token_accuracy": 0.10262698084115982,
|
|
"num_tokens": 862762.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.399918270111084,
|
|
"epoch": 0.42184964845862627,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 6.8485,
|
|
"mean_token_accuracy": 0.10503640621900559,
|
|
"num_tokens": 873023.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.350394630432129,
|
|
"epoch": 0.4272579772850189,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 6.9557,
|
|
"mean_token_accuracy": 0.0971641018986702,
|
|
"num_tokens": 885281.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.2543981075286865,
|
|
"epoch": 0.43266630611141155,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 6.9465,
|
|
"mean_token_accuracy": 0.10591788291931152,
|
|
"num_tokens": 896210.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.301179552078247,
|
|
"epoch": 0.43807463493780424,
|
|
"grad_norm": 0.78515625,
|
|
"learning_rate": 0.000202,
|
|
"loss": 6.8552,
|
|
"mean_token_accuracy": 0.10582882538437843,
|
|
"num_tokens": 909028.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.2928071975708,
|
|
"epoch": 0.4434829637641969,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 6.7897,
|
|
"mean_token_accuracy": 0.10721533969044686,
|
|
"num_tokens": 919710.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.37521538734436,
|
|
"epoch": 0.4488912925905895,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.000207,
|
|
"loss": 6.8504,
|
|
"mean_token_accuracy": 0.10534188896417618,
|
|
"num_tokens": 930019.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.277246475219727,
|
|
"epoch": 0.45429962141698216,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 6.7672,
|
|
"mean_token_accuracy": 0.11057652086019516,
|
|
"num_tokens": 940535.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.120538902282715,
|
|
"epoch": 0.4597079502433748,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000212,
|
|
"loss": 6.7261,
|
|
"mean_token_accuracy": 0.10942267924547196,
|
|
"num_tokens": 951390.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.226644659042359,
|
|
"epoch": 0.46511627906976744,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 6.7662,
|
|
"mean_token_accuracy": 0.11120115816593171,
|
|
"num_tokens": 963010.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.179320001602173,
|
|
"epoch": 0.4705246078961601,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 6.7145,
|
|
"mean_token_accuracy": 0.11507150083780289,
|
|
"num_tokens": 974032.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.230346393585205,
|
|
"epoch": 0.4759329367225527,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 6.7698,
|
|
"mean_token_accuracy": 0.11191772520542145,
|
|
"num_tokens": 983954.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.1874768257141115,
|
|
"epoch": 0.48134126554894535,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000222,
|
|
"loss": 6.7844,
|
|
"mean_token_accuracy": 0.1129900112748146,
|
|
"num_tokens": 994135.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.209841632843018,
|
|
"epoch": 0.48674959437533805,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 6.7424,
|
|
"mean_token_accuracy": 0.10967175662517548,
|
|
"num_tokens": 1006275.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.0370276927947994,
|
|
"epoch": 0.4921579232017307,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 6.5695,
|
|
"mean_token_accuracy": 0.11895088031888008,
|
|
"num_tokens": 1017767.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.171165800094604,
|
|
"epoch": 0.4975662520281233,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 6.6998,
|
|
"mean_token_accuracy": 0.11892824321985244,
|
|
"num_tokens": 1028286.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.08244252204895,
|
|
"epoch": 0.502974580854516,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 6.71,
|
|
"mean_token_accuracy": 0.11281011700630188,
|
|
"num_tokens": 1040555.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.177019929885864,
|
|
"epoch": 0.5083829096809086,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.8046,
|
|
"mean_token_accuracy": 0.11459487229585648,
|
|
"num_tokens": 1053668.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.1339305400848385,
|
|
"epoch": 0.5137912385073012,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.7437,
|
|
"mean_token_accuracy": 0.11368778720498085,
|
|
"num_tokens": 1066042.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.061350917816162,
|
|
"epoch": 0.5191995673336939,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.7603,
|
|
"mean_token_accuracy": 0.11287353932857513,
|
|
"num_tokens": 1077843.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.126706933975219,
|
|
"epoch": 0.5246078961600865,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.6459,
|
|
"mean_token_accuracy": 0.11803325861692429,
|
|
"num_tokens": 1089448.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.057846260070801,
|
|
"epoch": 0.5300162249864792,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.7488,
|
|
"mean_token_accuracy": 0.11246698349714279,
|
|
"num_tokens": 1100989.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.211519956588745,
|
|
"epoch": 0.5354245538128718,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.7773,
|
|
"mean_token_accuracy": 0.11612514406442642,
|
|
"num_tokens": 1113012.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.058304357528686,
|
|
"epoch": 0.5408328826392644,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 6.7171,
|
|
"mean_token_accuracy": 0.11422862857580185,
|
|
"num_tokens": 1123551.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 0.5408328826392644,
|
|
"eval_entropy": 7.070621424371546,
|
|
"eval_loss": 6.859935283660889,
|
|
"eval_mean_token_accuracy": 0.11617732908238064,
|
|
"eval_num_tokens": 1123551.0,
|
|
"eval_runtime": 1.9446,
|
|
"eval_samples_per_second": 1807.585,
|
|
"eval_steps_per_second": 226.27,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.120835399627685,
|
|
"epoch": 0.5462412114656571,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.6374,
|
|
"mean_token_accuracy": 0.11516082882881165,
|
|
"num_tokens": 1133838.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.017620658874511,
|
|
"epoch": 0.5516495402920497,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.6809,
|
|
"mean_token_accuracy": 0.1152262344956398,
|
|
"num_tokens": 1146005.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.052067136764526,
|
|
"epoch": 0.5570578691184424,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.7621,
|
|
"mean_token_accuracy": 0.11798961162567138,
|
|
"num_tokens": 1157328.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 6.988781690597534,
|
|
"epoch": 0.5624661979448351,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.5722,
|
|
"mean_token_accuracy": 0.12257469147443771,
|
|
"num_tokens": 1168592.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.097140264511109,
|
|
"epoch": 0.5678745267712277,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.6913,
|
|
"mean_token_accuracy": 0.11780922934412956,
|
|
"num_tokens": 1178680.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 7.017797470092773,
|
|
"epoch": 0.5732828555976204,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.6329,
|
|
"mean_token_accuracy": 0.12091006338596344,
|
|
"num_tokens": 1188825.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.020737600326538,
|
|
"epoch": 0.578691184424013,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.6649,
|
|
"mean_token_accuracy": 0.121755550801754,
|
|
"num_tokens": 1201375.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 7.082882022857666,
|
|
"epoch": 0.5840995132504057,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.7857,
|
|
"mean_token_accuracy": 0.11599067747592925,
|
|
"num_tokens": 1211915.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 7.0102982997894285,
|
|
"epoch": 0.5895078420767983,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.7015,
|
|
"mean_token_accuracy": 0.12261592373251914,
|
|
"num_tokens": 1223151.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 7.062853527069092,
|
|
"epoch": 0.5949161709031909,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.6242,
|
|
"mean_token_accuracy": 0.12241948544979095,
|
|
"num_tokens": 1234229.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 7.009330701828003,
|
|
"epoch": 0.6003244997295836,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.6335,
|
|
"mean_token_accuracy": 0.12372676730155945,
|
|
"num_tokens": 1245298.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 6.956117057800293,
|
|
"epoch": 0.6057328285559762,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.6296,
|
|
"mean_token_accuracy": 0.12371465265750885,
|
|
"num_tokens": 1256154.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 6.9070703983306885,
|
|
"epoch": 0.6111411573823688,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.5398,
|
|
"mean_token_accuracy": 0.12442566901445389,
|
|
"num_tokens": 1267404.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 6.95917820930481,
|
|
"epoch": 0.6165494862087615,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.5921,
|
|
"mean_token_accuracy": 0.12321074232459069,
|
|
"num_tokens": 1278894.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 6.912023448944092,
|
|
"epoch": 0.6219578150351541,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.5839,
|
|
"mean_token_accuracy": 0.12680861726403236,
|
|
"num_tokens": 1290408.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 7.008191251754761,
|
|
"epoch": 0.6273661438615468,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.6473,
|
|
"mean_token_accuracy": 0.12115282118320465,
|
|
"num_tokens": 1301508.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 7.004787969589233,
|
|
"epoch": 0.6327744726879394,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.624,
|
|
"mean_token_accuracy": 0.12800585329532624,
|
|
"num_tokens": 1312551.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 6.99333176612854,
|
|
"epoch": 0.638182801514332,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.7226,
|
|
"mean_token_accuracy": 0.12242905274033547,
|
|
"num_tokens": 1324876.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 7.026355409622193,
|
|
"epoch": 0.6435911303407247,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.6088,
|
|
"mean_token_accuracy": 0.12180419266223907,
|
|
"num_tokens": 1335401.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 6.767445611953735,
|
|
"epoch": 0.6489994591671173,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.538,
|
|
"mean_token_accuracy": 0.12269261777400971,
|
|
"num_tokens": 1348265.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 6.939475727081299,
|
|
"epoch": 0.65440778799351,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.4984,
|
|
"mean_token_accuracy": 0.12996519654989241,
|
|
"num_tokens": 1359078.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 6.676696491241455,
|
|
"epoch": 0.6598161168199026,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.3855,
|
|
"mean_token_accuracy": 0.12842316627502443,
|
|
"num_tokens": 1369499.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 6.887877798080444,
|
|
"epoch": 0.6652244456462953,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.4993,
|
|
"mean_token_accuracy": 0.12684686109423637,
|
|
"num_tokens": 1380329.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 6.855057859420777,
|
|
"epoch": 0.670632774472688,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.5755,
|
|
"mean_token_accuracy": 0.12310040593147278,
|
|
"num_tokens": 1393047.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 6.847856616973877,
|
|
"epoch": 0.6760411032990806,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.5675,
|
|
"mean_token_accuracy": 0.12307048961520195,
|
|
"num_tokens": 1405600.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 6.890627050399781,
|
|
"epoch": 0.6814494321254733,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.5113,
|
|
"mean_token_accuracy": 0.12754200771450996,
|
|
"num_tokens": 1416670.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 6.845398712158203,
|
|
"epoch": 0.6868577609518659,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.5305,
|
|
"mean_token_accuracy": 0.12670771926641464,
|
|
"num_tokens": 1428929.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 6.809947156906128,
|
|
"epoch": 0.6922660897782585,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.4476,
|
|
"mean_token_accuracy": 0.12762413546442986,
|
|
"num_tokens": 1439298.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 6.87408938407898,
|
|
"epoch": 0.6976744186046512,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.5514,
|
|
"mean_token_accuracy": 0.13081721141934394,
|
|
"num_tokens": 1449853.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 6.944610214233398,
|
|
"epoch": 0.7030827474310438,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.6146,
|
|
"mean_token_accuracy": 0.12372421324253083,
|
|
"num_tokens": 1460816.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 6.768828868865967,
|
|
"epoch": 0.7084910762574365,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.4671,
|
|
"mean_token_accuracy": 0.12459458410739899,
|
|
"num_tokens": 1473003.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 6.815493726730347,
|
|
"epoch": 0.7138994050838291,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.5935,
|
|
"mean_token_accuracy": 0.12046970501542091,
|
|
"num_tokens": 1484745.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 6.918433952331543,
|
|
"epoch": 0.7193077339102217,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.4568,
|
|
"mean_token_accuracy": 0.1293526917695999,
|
|
"num_tokens": 1495803.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 6.756926679611206,
|
|
"epoch": 0.7247160627366144,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.4795,
|
|
"mean_token_accuracy": 0.13275134861469268,
|
|
"num_tokens": 1507504.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 6.846708583831787,
|
|
"epoch": 0.730124391563007,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.5749,
|
|
"mean_token_accuracy": 0.12511131837964057,
|
|
"num_tokens": 1518765.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 6.733471632003784,
|
|
"epoch": 0.7355327203893997,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.4555,
|
|
"mean_token_accuracy": 0.12526600286364556,
|
|
"num_tokens": 1529481.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 6.803652429580689,
|
|
"epoch": 0.7409410492157923,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.4681,
|
|
"mean_token_accuracy": 0.1339200973510742,
|
|
"num_tokens": 1539612.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 6.806907892227173,
|
|
"epoch": 0.7463493780421849,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.6046,
|
|
"mean_token_accuracy": 0.12426809221506119,
|
|
"num_tokens": 1551370.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 6.894266033172608,
|
|
"epoch": 0.7517577068685776,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.4958,
|
|
"mean_token_accuracy": 0.13355423510074615,
|
|
"num_tokens": 1561662.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 6.808643198013305,
|
|
"epoch": 0.7571660356949702,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.4794,
|
|
"mean_token_accuracy": 0.12898182794451712,
|
|
"num_tokens": 1573479.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 6.713399219512939,
|
|
"epoch": 0.7625743645213628,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.4255,
|
|
"mean_token_accuracy": 0.13072630763053894,
|
|
"num_tokens": 1583684.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 6.775609922409058,
|
|
"epoch": 0.7679826933477556,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.4727,
|
|
"mean_token_accuracy": 0.13058030232787132,
|
|
"num_tokens": 1595694.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 6.718004512786865,
|
|
"epoch": 0.7733910221741482,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.4717,
|
|
"mean_token_accuracy": 0.12594751939177512,
|
|
"num_tokens": 1606104.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 6.8197935104370115,
|
|
"epoch": 0.7787993510005409,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.541,
|
|
"mean_token_accuracy": 0.1298747517168522,
|
|
"num_tokens": 1617368.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 6.853786754608154,
|
|
"epoch": 0.7842076798269335,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.5568,
|
|
"mean_token_accuracy": 0.12204690799117088,
|
|
"num_tokens": 1630485.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 6.838800954818725,
|
|
"epoch": 0.7896160086533262,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.3953,
|
|
"mean_token_accuracy": 0.13309185206890106,
|
|
"num_tokens": 1641411.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 6.606268787384034,
|
|
"epoch": 0.7950243374797188,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.4154,
|
|
"mean_token_accuracy": 0.13472312912344933,
|
|
"num_tokens": 1653737.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.835287189483642,
|
|
"epoch": 0.8004326663061114,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.5903,
|
|
"mean_token_accuracy": 0.12430674731731414,
|
|
"num_tokens": 1665910.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 6.817589139938354,
|
|
"epoch": 0.8058409951325041,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.3509,
|
|
"mean_token_accuracy": 0.1289426304399967,
|
|
"num_tokens": 1676927.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.585090637207031,
|
|
"epoch": 0.8112493239588967,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.4032,
|
|
"mean_token_accuracy": 0.13011449351906776,
|
|
"num_tokens": 1688167.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 6.720578002929687,
|
|
"epoch": 0.8166576527852893,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.4835,
|
|
"mean_token_accuracy": 0.13275314942002298,
|
|
"num_tokens": 1699899.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 6.661290502548217,
|
|
"epoch": 0.822065981611682,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.4142,
|
|
"mean_token_accuracy": 0.13195080161094666,
|
|
"num_tokens": 1709862.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.705444717407227,
|
|
"epoch": 0.8274743104380746,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.3877,
|
|
"mean_token_accuracy": 0.13449530974030494,
|
|
"num_tokens": 1719902.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.709910154342651,
|
|
"epoch": 0.8328826392644673,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.4083,
|
|
"mean_token_accuracy": 0.13319506570696832,
|
|
"num_tokens": 1730351.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.61332688331604,
|
|
"epoch": 0.8382909680908599,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.3675,
|
|
"mean_token_accuracy": 0.13003436774015426,
|
|
"num_tokens": 1741783.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.637337017059326,
|
|
"epoch": 0.8436992969172525,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.4219,
|
|
"mean_token_accuracy": 0.13431969806551933,
|
|
"num_tokens": 1752952.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.77474799156189,
|
|
"epoch": 0.8491076257436452,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.4312,
|
|
"mean_token_accuracy": 0.1275618128478527,
|
|
"num_tokens": 1764014.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.689550447463989,
|
|
"epoch": 0.8545159545700378,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.3963,
|
|
"mean_token_accuracy": 0.13133953213691713,
|
|
"num_tokens": 1776154.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.676612186431885,
|
|
"epoch": 0.8599242833964305,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.4523,
|
|
"mean_token_accuracy": 0.1290397100150585,
|
|
"num_tokens": 1787292.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.641109752655029,
|
|
"epoch": 0.8653326122228231,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.4167,
|
|
"mean_token_accuracy": 0.13453051149845124,
|
|
"num_tokens": 1799469.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.551521062850952,
|
|
"epoch": 0.8707409410492158,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.2944,
|
|
"mean_token_accuracy": 0.13961234614253043,
|
|
"num_tokens": 1810091.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.746007585525513,
|
|
"epoch": 0.8761492698756085,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.4587,
|
|
"mean_token_accuracy": 0.13312253430485727,
|
|
"num_tokens": 1820835.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.6614655494689945,
|
|
"epoch": 0.8815575987020011,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.4069,
|
|
"mean_token_accuracy": 0.13295662850141526,
|
|
"num_tokens": 1831868.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.575878047943116,
|
|
"epoch": 0.8869659275283938,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.387,
|
|
"mean_token_accuracy": 0.13565733730793,
|
|
"num_tokens": 1842040.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.7888940334320065,
|
|
"epoch": 0.8923742563547864,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.4603,
|
|
"mean_token_accuracy": 0.13313809186220169,
|
|
"num_tokens": 1853765.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.690069341659546,
|
|
"epoch": 0.897782585181179,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.4548,
|
|
"mean_token_accuracy": 0.13598668873310088,
|
|
"num_tokens": 1866081.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.610725545883179,
|
|
"epoch": 0.9031909140075717,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.364,
|
|
"mean_token_accuracy": 0.1348128743469715,
|
|
"num_tokens": 1876754.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.72492733001709,
|
|
"epoch": 0.9085992428339643,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.4677,
|
|
"mean_token_accuracy": 0.12978390827775002,
|
|
"num_tokens": 1888299.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.47598352432251,
|
|
"epoch": 0.914007571660357,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.2533,
|
|
"mean_token_accuracy": 0.1417229875922203,
|
|
"num_tokens": 1899027.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.667154026031494,
|
|
"epoch": 0.9194159004867496,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.3774,
|
|
"mean_token_accuracy": 0.1327328972518444,
|
|
"num_tokens": 1911043.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.636425971984863,
|
|
"epoch": 0.9248242293131422,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.3034,
|
|
"mean_token_accuracy": 0.13325904086232185,
|
|
"num_tokens": 1921295.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.645001459121704,
|
|
"epoch": 0.9302325581395349,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.4006,
|
|
"mean_token_accuracy": 0.13651155978441237,
|
|
"num_tokens": 1932944.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.538773679733277,
|
|
"epoch": 0.9356408869659275,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.3351,
|
|
"mean_token_accuracy": 0.13037271574139594,
|
|
"num_tokens": 1944072.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.591965293884277,
|
|
"epoch": 0.9410492157923201,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.3428,
|
|
"mean_token_accuracy": 0.13658150658011436,
|
|
"num_tokens": 1954798.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.506063842773438,
|
|
"epoch": 0.9464575446187128,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.3039,
|
|
"mean_token_accuracy": 0.1442844972014427,
|
|
"num_tokens": 1965304.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.660945653915405,
|
|
"epoch": 0.9518658734451054,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.3527,
|
|
"mean_token_accuracy": 0.13428603932261468,
|
|
"num_tokens": 1976797.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.56065354347229,
|
|
"epoch": 0.9572742022714981,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.374,
|
|
"mean_token_accuracy": 0.13659664168953894,
|
|
"num_tokens": 1987448.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.366022777557373,
|
|
"epoch": 0.9626825310978907,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.2006,
|
|
"mean_token_accuracy": 0.13602133318781853,
|
|
"num_tokens": 2000948.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.597631502151489,
|
|
"epoch": 0.9680908599242833,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.3109,
|
|
"mean_token_accuracy": 0.13974019661545753,
|
|
"num_tokens": 2011807.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.633898687362671,
|
|
"epoch": 0.9734991887506761,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.3585,
|
|
"mean_token_accuracy": 0.13105632588267327,
|
|
"num_tokens": 2022835.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.387359571456909,
|
|
"epoch": 0.9789075175770687,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.276,
|
|
"mean_token_accuracy": 0.14153993576765062,
|
|
"num_tokens": 2034331.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.695045137405396,
|
|
"epoch": 0.9843158464034614,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.2744,
|
|
"mean_token_accuracy": 0.13374185785651208,
|
|
"num_tokens": 2047002.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.452011156082153,
|
|
"epoch": 0.989724175229854,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.2301,
|
|
"mean_token_accuracy": 0.14060857594013215,
|
|
"num_tokens": 2056299.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.562105369567871,
|
|
"epoch": 0.9951325040562466,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.3512,
|
|
"mean_token_accuracy": 0.13779174089431762,
|
|
"num_tokens": 2068031.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.541797320048015,
|
|
"epoch": 1.0,
|
|
"grad_norm": 1.8046875,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.2951,
|
|
"mean_token_accuracy": 0.14415054851108128,
|
|
"num_tokens": 2076889.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.501549148559571,
|
|
"epoch": 1.0054083288263926,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.1676,
|
|
"mean_token_accuracy": 0.14561108872294426,
|
|
"num_tokens": 2087316.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.5583666324615475,
|
|
"epoch": 1.0108166576527853,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.1739,
|
|
"mean_token_accuracy": 0.13918173015117646,
|
|
"num_tokens": 2098522.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.392220306396484,
|
|
"epoch": 1.016224986479178,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.1354,
|
|
"mean_token_accuracy": 0.14603151232004166,
|
|
"num_tokens": 2110015.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.509898853302002,
|
|
"epoch": 1.0216333153055706,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.2538,
|
|
"mean_token_accuracy": 0.13781652599573135,
|
|
"num_tokens": 2121333.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.577088499069214,
|
|
"epoch": 1.0270416441319632,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.138,
|
|
"mean_token_accuracy": 0.1490761384367943,
|
|
"num_tokens": 2131385.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.454202699661255,
|
|
"epoch": 1.0324499729583558,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.1985,
|
|
"mean_token_accuracy": 0.1469516322016716,
|
|
"num_tokens": 2143274.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.380458831787109,
|
|
"epoch": 1.0378583017847485,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.0657,
|
|
"mean_token_accuracy": 0.14620195776224137,
|
|
"num_tokens": 2154529.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.4978209972381595,
|
|
"epoch": 1.043266630611141,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.2219,
|
|
"mean_token_accuracy": 0.14443108439445496,
|
|
"num_tokens": 2165949.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.401442909240723,
|
|
"epoch": 1.0486749594375337,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.1906,
|
|
"mean_token_accuracy": 0.14745446890592576,
|
|
"num_tokens": 2176710.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.489724588394165,
|
|
"epoch": 1.0540832882639264,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.1687,
|
|
"mean_token_accuracy": 0.15010684430599214,
|
|
"num_tokens": 2189470.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.376638793945313,
|
|
"epoch": 1.059491617090319,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.1227,
|
|
"mean_token_accuracy": 0.15111407190561293,
|
|
"num_tokens": 2200266.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.467359828948974,
|
|
"epoch": 1.0648999459167117,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.1322,
|
|
"mean_token_accuracy": 0.1447245694696903,
|
|
"num_tokens": 2210736.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.49309549331665,
|
|
"epoch": 1.0703082747431043,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.2527,
|
|
"mean_token_accuracy": 0.14038581252098084,
|
|
"num_tokens": 2221057.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.4171429634094235,
|
|
"epoch": 1.075716603569497,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.2282,
|
|
"mean_token_accuracy": 0.14197226315736772,
|
|
"num_tokens": 2232352.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.583332157135009,
|
|
"epoch": 1.0811249323958896,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.134,
|
|
"mean_token_accuracy": 0.149154694378376,
|
|
"num_tokens": 2244655.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 1.0811249323958896,
|
|
"eval_entropy": 6.15572995597666,
|
|
"eval_loss": 6.430820941925049,
|
|
"eval_mean_token_accuracy": 0.14240490521558308,
|
|
"eval_num_tokens": 2244655.0,
|
|
"eval_runtime": 1.9399,
|
|
"eval_samples_per_second": 1811.932,
|
|
"eval_steps_per_second": 226.814,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.2607872009277346,
|
|
"epoch": 1.0865332612222822,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999997383518051,
|
|
"loss": 6.0974,
|
|
"mean_token_accuracy": 0.15421370416879654,
|
|
"num_tokens": 2255699.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.41339807510376,
|
|
"epoch": 1.0919415900486749,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999986754070562,
|
|
"loss": 6.0937,
|
|
"mean_token_accuracy": 0.151519475877285,
|
|
"num_tokens": 2266627.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.485339260101318,
|
|
"epoch": 1.0973499188750675,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004999967948166009,
|
|
"loss": 6.2596,
|
|
"mean_token_accuracy": 0.1421214833855629,
|
|
"num_tokens": 2277814.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.573788976669311,
|
|
"epoch": 1.1027582477014604,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004999940965872734,
|
|
"loss": 6.2071,
|
|
"mean_token_accuracy": 0.1435778819024563,
|
|
"num_tokens": 2287344.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.232277965545654,
|
|
"epoch": 1.1081665765278528,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999905807288788,
|
|
"loss": 6.0105,
|
|
"mean_token_accuracy": 0.15129713118076324,
|
|
"num_tokens": 2298332.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.4171154499053955,
|
|
"epoch": 1.1135749053542456,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000499986247254194,
|
|
"loss": 6.1208,
|
|
"mean_token_accuracy": 0.14398186057806014,
|
|
"num_tokens": 2309161.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.3788182735443115,
|
|
"epoch": 1.1189832341806383,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004999810961789667,
|
|
"loss": 6.1205,
|
|
"mean_token_accuracy": 0.15155730098485948,
|
|
"num_tokens": 2319516.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.321287155151367,
|
|
"epoch": 1.124391563007031,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.000499975127521916,
|
|
"loss": 6.0164,
|
|
"mean_token_accuracy": 0.1517544224858284,
|
|
"num_tokens": 2330441.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.313332509994507,
|
|
"epoch": 1.1297998918334236,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999683413047319,
|
|
"loss": 6.0256,
|
|
"mean_token_accuracy": 0.15264784991741182,
|
|
"num_tokens": 2340524.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.3693748950958256,
|
|
"epoch": 1.1352082206598162,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004999607375520759,
|
|
"loss": 6.0188,
|
|
"mean_token_accuracy": 0.15174520611763,
|
|
"num_tokens": 2353379.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.395606136322021,
|
|
"epoch": 1.1406165494862088,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004999523162915794,
|
|
"loss": 6.2108,
|
|
"mean_token_accuracy": 0.1419041022658348,
|
|
"num_tokens": 2365708.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.316472816467285,
|
|
"epoch": 1.1460248783126015,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999430775538459,
|
|
"loss": 6.1105,
|
|
"mean_token_accuracy": 0.14867828115820886,
|
|
"num_tokens": 2375630.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.356218957901001,
|
|
"epoch": 1.151433207138994,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999330213724487,
|
|
"loss": 6.1092,
|
|
"mean_token_accuracy": 0.14748374819755555,
|
|
"num_tokens": 2386885.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.318326759338379,
|
|
"epoch": 1.1568415359653867,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999221477839319,
|
|
"loss": 6.1413,
|
|
"mean_token_accuracy": 0.14829123616218567,
|
|
"num_tokens": 2397676.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.429721021652222,
|
|
"epoch": 1.1622498647917794,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004999104568278104,
|
|
"loss": 6.1464,
|
|
"mean_token_accuracy": 0.14260400384664534,
|
|
"num_tokens": 2409039.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.405117321014404,
|
|
"epoch": 1.167658193618172,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000499897948546569,
|
|
"loss": 6.1863,
|
|
"mean_token_accuracy": 0.14825090393424034,
|
|
"num_tokens": 2420103.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.293612098693847,
|
|
"epoch": 1.1730665224445647,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004998846229856629,
|
|
"loss": 6.0965,
|
|
"mean_token_accuracy": 0.14956801235675812,
|
|
"num_tokens": 2430792.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.379774379730224,
|
|
"epoch": 1.1784748512709573,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004998704801935171,
|
|
"loss": 6.1059,
|
|
"mean_token_accuracy": 0.14647614508867263,
|
|
"num_tokens": 2442165.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.316842126846313,
|
|
"epoch": 1.18388318009735,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004998555202215267,
|
|
"loss": 6.0251,
|
|
"mean_token_accuracy": 0.1499698504805565,
|
|
"num_tokens": 2454231.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.376669073104859,
|
|
"epoch": 1.1892915089237426,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004998397431240561,
|
|
"loss": 6.0724,
|
|
"mean_token_accuracy": 0.15592599362134935,
|
|
"num_tokens": 2464828.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.244733238220215,
|
|
"epoch": 1.1946998377501352,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004998231489584395,
|
|
"loss": 6.1251,
|
|
"mean_token_accuracy": 0.14098657220602034,
|
|
"num_tokens": 2476144.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.44529356956482,
|
|
"epoch": 1.2001081665765279,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00049980573778498,
|
|
"loss": 6.295,
|
|
"mean_token_accuracy": 0.14309989288449287,
|
|
"num_tokens": 2489116.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.497919797897339,
|
|
"epoch": 1.2055164954029205,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004997875096669498,
|
|
"loss": 6.144,
|
|
"mean_token_accuracy": 0.14609830528497697,
|
|
"num_tokens": 2501609.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.216084146499634,
|
|
"epoch": 1.2109248242293131,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004997684646705901,
|
|
"loss": 6.0594,
|
|
"mean_token_accuracy": 0.15941857546567917,
|
|
"num_tokens": 2513338.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.397832250595092,
|
|
"epoch": 1.2163331530557058,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004997486028651103,
|
|
"loss": 6.0824,
|
|
"mean_token_accuracy": 0.1556407019495964,
|
|
"num_tokens": 2523199.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.352018690109253,
|
|
"epoch": 1.2217414818820984,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004997279243226884,
|
|
"loss": 6.0961,
|
|
"mean_token_accuracy": 0.1505078837275505,
|
|
"num_tokens": 2535136.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.295426988601685,
|
|
"epoch": 1.227149810708491,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004997064291184701,
|
|
"loss": 6.0032,
|
|
"mean_token_accuracy": 0.14287411645054818,
|
|
"num_tokens": 2546428.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.423472547531128,
|
|
"epoch": 1.2325581395348837,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004996841173305691,
|
|
"loss": 6.2024,
|
|
"mean_token_accuracy": 0.14975845515727998,
|
|
"num_tokens": 2558304.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.165703773498535,
|
|
"epoch": 1.2379664683612763,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004996609890400665,
|
|
"loss": 5.8907,
|
|
"mean_token_accuracy": 0.15885070562362671,
|
|
"num_tokens": 2568717.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.3368885040283205,
|
|
"epoch": 1.243374797187669,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004996370443310107,
|
|
"loss": 6.1036,
|
|
"mean_token_accuracy": 0.14628757089376448,
|
|
"num_tokens": 2578748.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.348951196670532,
|
|
"epoch": 1.2487831260140616,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004996122832904165,
|
|
"loss": 6.0661,
|
|
"mean_token_accuracy": 0.14893802553415297,
|
|
"num_tokens": 2590095.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.312915945053101,
|
|
"epoch": 1.2541914548404542,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004995867060082658,
|
|
"loss": 6.1088,
|
|
"mean_token_accuracy": 0.1505606383085251,
|
|
"num_tokens": 2600929.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.381860637664795,
|
|
"epoch": 1.2595997836668469,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004995603125775065,
|
|
"loss": 6.1062,
|
|
"mean_token_accuracy": 0.1485731616616249,
|
|
"num_tokens": 2611830.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.1977239608764645,
|
|
"epoch": 1.2650081124932395,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004995331030940523,
|
|
"loss": 6.0498,
|
|
"mean_token_accuracy": 0.14882538169622422,
|
|
"num_tokens": 2623177.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.265497636795044,
|
|
"epoch": 1.2704164413196322,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004995050776567826,
|
|
"loss": 6.0423,
|
|
"mean_token_accuracy": 0.15137407928705215,
|
|
"num_tokens": 2635563.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.35697751045227,
|
|
"epoch": 1.2758247701460248,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004994762363675421,
|
|
"loss": 6.0343,
|
|
"mean_token_accuracy": 0.15718608051538469,
|
|
"num_tokens": 2647157.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.22226848602295,
|
|
"epoch": 1.2812330989724177,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004994465793311396,
|
|
"loss": 5.9977,
|
|
"mean_token_accuracy": 0.1555067002773285,
|
|
"num_tokens": 2658175.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.260870838165284,
|
|
"epoch": 1.28664142779881,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004994161066553493,
|
|
"loss": 5.9655,
|
|
"mean_token_accuracy": 0.15602218061685563,
|
|
"num_tokens": 2669458.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.352515697479248,
|
|
"epoch": 1.292049756625203,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004993848184509086,
|
|
"loss": 6.1736,
|
|
"mean_token_accuracy": 0.1515059992671013,
|
|
"num_tokens": 2679727.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.253958749771118,
|
|
"epoch": 1.2974580854515954,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004993527148315193,
|
|
"loss": 6.1329,
|
|
"mean_token_accuracy": 0.1500518336892128,
|
|
"num_tokens": 2691031.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.272245407104492,
|
|
"epoch": 1.3028664142779882,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004993197959138458,
|
|
"loss": 5.9166,
|
|
"mean_token_accuracy": 0.162589730322361,
|
|
"num_tokens": 2701699.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.210496473312378,
|
|
"epoch": 1.3082747431043806,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004992860618175156,
|
|
"loss": 5.9696,
|
|
"mean_token_accuracy": 0.1547999605536461,
|
|
"num_tokens": 2713038.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.211694431304932,
|
|
"epoch": 1.3136830719307735,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004992515126651184,
|
|
"loss": 5.9651,
|
|
"mean_token_accuracy": 0.15444047302007674,
|
|
"num_tokens": 2723685.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.235717296600342,
|
|
"epoch": 1.319091400757166,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004992161485822059,
|
|
"loss": 6.0361,
|
|
"mean_token_accuracy": 0.1552966371178627,
|
|
"num_tokens": 2735341.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.30128288269043,
|
|
"epoch": 1.3244997295835588,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004991799696972913,
|
|
"loss": 5.9831,
|
|
"mean_token_accuracy": 0.15909997522830963,
|
|
"num_tokens": 2746052.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.175243139266968,
|
|
"epoch": 1.3299080584099512,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004991429761418488,
|
|
"loss": 5.9742,
|
|
"mean_token_accuracy": 0.15188155844807624,
|
|
"num_tokens": 2757155.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.244051599502564,
|
|
"epoch": 1.335316387236344,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000499105168050313,
|
|
"loss": 6.0274,
|
|
"mean_token_accuracy": 0.14860790371894836,
|
|
"num_tokens": 2769631.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.3532966613769535,
|
|
"epoch": 1.3407247160627367,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004990665455600787,
|
|
"loss": 6.1346,
|
|
"mean_token_accuracy": 0.1509730860590935,
|
|
"num_tokens": 2780797.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.183553266525268,
|
|
"epoch": 1.3461330448891293,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004990271088114999,
|
|
"loss": 5.985,
|
|
"mean_token_accuracy": 0.1532391682267189,
|
|
"num_tokens": 2792721.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.230117893218994,
|
|
"epoch": 1.351541373715522,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004989868579478901,
|
|
"loss": 6.0428,
|
|
"mean_token_accuracy": 0.16393045336008072,
|
|
"num_tokens": 2804531.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.308541011810303,
|
|
"epoch": 1.3569497025419146,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004989457931155209,
|
|
"loss": 6.0393,
|
|
"mean_token_accuracy": 0.150795978307724,
|
|
"num_tokens": 2816666.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.244772624969483,
|
|
"epoch": 1.3623580313683072,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004989039144636219,
|
|
"loss": 6.066,
|
|
"mean_token_accuracy": 0.15578461140394212,
|
|
"num_tokens": 2827977.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.214469718933105,
|
|
"epoch": 1.3677663601946999,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004988612221443804,
|
|
"loss": 6.0326,
|
|
"mean_token_accuracy": 0.14781438559293747,
|
|
"num_tokens": 2839069.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.328051614761352,
|
|
"epoch": 1.3731746890210925,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004988177163129403,
|
|
"loss": 6.1066,
|
|
"mean_token_accuracy": 0.14855716824531556,
|
|
"num_tokens": 2850632.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.215375328063965,
|
|
"epoch": 1.3785830178474852,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004987733971274018,
|
|
"loss": 5.9962,
|
|
"mean_token_accuracy": 0.1514300212264061,
|
|
"num_tokens": 2862204.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.199917984008789,
|
|
"epoch": 1.3839913466738778,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004987282647488211,
|
|
"loss": 5.9524,
|
|
"mean_token_accuracy": 0.1645523026585579,
|
|
"num_tokens": 2874010.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.22587022781372,
|
|
"epoch": 1.3893996755002704,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004986823193412093,
|
|
"loss": 6.0238,
|
|
"mean_token_accuracy": 0.15714267790317535,
|
|
"num_tokens": 2884315.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.347529888153076,
|
|
"epoch": 1.394808004326663,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000498635561071532,
|
|
"loss": 6.1771,
|
|
"mean_token_accuracy": 0.14605384021997453,
|
|
"num_tokens": 2896327.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.226870584487915,
|
|
"epoch": 1.4002163331530557,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000498587990109709,
|
|
"loss": 5.9394,
|
|
"mean_token_accuracy": 0.15527413636446,
|
|
"num_tokens": 2907007.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.1950747013092045,
|
|
"epoch": 1.4056246619794484,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004985396066286133,
|
|
"loss": 5.9939,
|
|
"mean_token_accuracy": 0.15905235856771469,
|
|
"num_tokens": 2918852.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.203013324737549,
|
|
"epoch": 1.411032990805841,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004984904108040706,
|
|
"loss": 5.9052,
|
|
"mean_token_accuracy": 0.16093855798244477,
|
|
"num_tokens": 2929948.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.176662969589233,
|
|
"epoch": 1.4164413196322336,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004984404028148585,
|
|
"loss": 5.9765,
|
|
"mean_token_accuracy": 0.153941547870636,
|
|
"num_tokens": 2941299.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.12311019897461,
|
|
"epoch": 1.4218496484586263,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004983895828427062,
|
|
"loss": 5.9062,
|
|
"mean_token_accuracy": 0.157071490585804,
|
|
"num_tokens": 2952283.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.472334623336792,
|
|
"epoch": 1.427257977285019,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004983379510722935,
|
|
"loss": 6.2636,
|
|
"mean_token_accuracy": 0.14533184543251992,
|
|
"num_tokens": 2962241.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.201954317092896,
|
|
"epoch": 1.4326663061114115,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004982855076912506,
|
|
"loss": 6.0153,
|
|
"mean_token_accuracy": 0.15568085312843322,
|
|
"num_tokens": 2973408.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.209590101242066,
|
|
"epoch": 1.4380746349378042,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004982322528901568,
|
|
"loss": 6.0346,
|
|
"mean_token_accuracy": 0.15969384759664534,
|
|
"num_tokens": 2985877.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.311101341247559,
|
|
"epoch": 1.4434829637641968,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.00049817818686254,
|
|
"loss": 6.0501,
|
|
"mean_token_accuracy": 0.15240749716758728,
|
|
"num_tokens": 2996935.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.100203990936279,
|
|
"epoch": 1.4488912925905895,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004981233098048765,
|
|
"loss": 5.9407,
|
|
"mean_token_accuracy": 0.15905434638261795,
|
|
"num_tokens": 3007932.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.316351747512817,
|
|
"epoch": 1.454299621416982,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004980676219165894,
|
|
"loss": 6.0265,
|
|
"mean_token_accuracy": 0.1527304731309414,
|
|
"num_tokens": 3019368.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.269512557983399,
|
|
"epoch": 1.4597079502433747,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000498011123400049,
|
|
"loss": 6.0932,
|
|
"mean_token_accuracy": 0.15539545863866805,
|
|
"num_tokens": 3029787.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.171151971817016,
|
|
"epoch": 1.4651162790697674,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004979538144605707,
|
|
"loss": 6.0224,
|
|
"mean_token_accuracy": 0.15601241141557692,
|
|
"num_tokens": 3042100.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.197288656234742,
|
|
"epoch": 1.47052460789616,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004978956953064154,
|
|
"loss": 6.0538,
|
|
"mean_token_accuracy": 0.1524283170700073,
|
|
"num_tokens": 3054160.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.2296675682067875,
|
|
"epoch": 1.4759329367225527,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004978367661487883,
|
|
"loss": 6.0199,
|
|
"mean_token_accuracy": 0.15398170799016953,
|
|
"num_tokens": 3065401.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.1436192989349365,
|
|
"epoch": 1.4813412655489453,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004977770272018379,
|
|
"loss": 5.942,
|
|
"mean_token_accuracy": 0.15897015780210494,
|
|
"num_tokens": 3076137.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.228366565704346,
|
|
"epoch": 1.486749594375338,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000497716478682656,
|
|
"loss": 6.078,
|
|
"mean_token_accuracy": 0.1585587590932846,
|
|
"num_tokens": 3087274.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.243017673492432,
|
|
"epoch": 1.4921579232017308,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004976551208112759,
|
|
"loss": 5.9683,
|
|
"mean_token_accuracy": 0.15684017390012742,
|
|
"num_tokens": 3098476.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.188798952102661,
|
|
"epoch": 1.4975662520281232,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004975929538106721,
|
|
"loss": 5.9849,
|
|
"mean_token_accuracy": 0.1557128444314003,
|
|
"num_tokens": 3110060.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.114158964157104,
|
|
"epoch": 1.502974580854516,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004975299779067598,
|
|
"loss": 5.8907,
|
|
"mean_token_accuracy": 0.15629031956195832,
|
|
"num_tokens": 3121032.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.20700626373291,
|
|
"epoch": 1.5083829096809085,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004974661933283936,
|
|
"loss": 5.9123,
|
|
"mean_token_accuracy": 0.16519911140203475,
|
|
"num_tokens": 3131999.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.059320020675659,
|
|
"epoch": 1.5137912385073014,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004974016003073667,
|
|
"loss": 5.8411,
|
|
"mean_token_accuracy": 0.16118691116571426,
|
|
"num_tokens": 3142693.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.057157564163208,
|
|
"epoch": 1.5191995673336938,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004973361990784104,
|
|
"loss": 5.7987,
|
|
"mean_token_accuracy": 0.15598912984132768,
|
|
"num_tokens": 3154179.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.2198803424835205,
|
|
"epoch": 1.5246078961600866,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004972699898791927,
|
|
"loss": 5.8996,
|
|
"mean_token_accuracy": 0.15782651603221892,
|
|
"num_tokens": 3165314.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.161393499374389,
|
|
"epoch": 1.530016224986479,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004972029729503182,
|
|
"loss": 6.0062,
|
|
"mean_token_accuracy": 0.16119608134031296,
|
|
"num_tokens": 3177646.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.221659851074219,
|
|
"epoch": 1.535424553812872,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004971351485353265,
|
|
"loss": 6.0027,
|
|
"mean_token_accuracy": 0.15873263776302338,
|
|
"num_tokens": 3188401.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.118788003921509,
|
|
"epoch": 1.5408328826392643,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004970665168806917,
|
|
"loss": 5.8784,
|
|
"mean_token_accuracy": 0.15916356444358826,
|
|
"num_tokens": 3200206.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.156543779373169,
|
|
"epoch": 1.5462412114656572,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004969970782358214,
|
|
"loss": 6.0055,
|
|
"mean_token_accuracy": 0.15648946762084961,
|
|
"num_tokens": 3211981.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.123556804656983,
|
|
"epoch": 1.5516495402920496,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004969268328530557,
|
|
"loss": 6.0066,
|
|
"mean_token_accuracy": 0.15547577291727066,
|
|
"num_tokens": 3221997.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.2214514255523685,
|
|
"epoch": 1.5570578691184425,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004968557809876666,
|
|
"loss": 5.9507,
|
|
"mean_token_accuracy": 0.15656527876853943,
|
|
"num_tokens": 3234333.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.149718809127807,
|
|
"epoch": 1.562466197944835,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004967839228978568,
|
|
"loss": 5.986,
|
|
"mean_token_accuracy": 0.15404829382896423,
|
|
"num_tokens": 3245083.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.164693975448609,
|
|
"epoch": 1.5678745267712277,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004967112588447585,
|
|
"loss": 5.8275,
|
|
"mean_token_accuracy": 0.16424602419137954,
|
|
"num_tokens": 3255787.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.083181762695313,
|
|
"epoch": 1.5732828555976204,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004966377890924332,
|
|
"loss": 6.0053,
|
|
"mean_token_accuracy": 0.15939338356256486,
|
|
"num_tokens": 3267793.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.127884101867676,
|
|
"epoch": 1.578691184424013,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004965635139078704,
|
|
"loss": 5.8707,
|
|
"mean_token_accuracy": 0.16731424480676652,
|
|
"num_tokens": 3279100.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.199070119857788,
|
|
"epoch": 1.5840995132504057,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004964884335609857,
|
|
"loss": 6.0326,
|
|
"mean_token_accuracy": 0.1608247846364975,
|
|
"num_tokens": 3290251.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.099456167221069,
|
|
"epoch": 1.5895078420767983,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004964125483246218,
|
|
"loss": 5.9035,
|
|
"mean_token_accuracy": 0.15708550959825515,
|
|
"num_tokens": 3301817.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.218367958068848,
|
|
"epoch": 1.594916170903191,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004963358584745457,
|
|
"loss": 5.9648,
|
|
"mean_token_accuracy": 0.15902439057826995,
|
|
"num_tokens": 3313435.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.103563737869263,
|
|
"epoch": 1.6003244997295836,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004962583642894482,
|
|
"loss": 6.0759,
|
|
"mean_token_accuracy": 0.15265134423971177,
|
|
"num_tokens": 3325489.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.266132020950318,
|
|
"epoch": 1.6057328285559762,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004961800660509437,
|
|
"loss": 5.9367,
|
|
"mean_token_accuracy": 0.16093226224184037,
|
|
"num_tokens": 3337030.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.019805431365967,
|
|
"epoch": 1.6111411573823688,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000496100964043568,
|
|
"loss": 5.8949,
|
|
"mean_token_accuracy": 0.162692628800869,
|
|
"num_tokens": 3347766.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.065528202056885,
|
|
"epoch": 1.6165494862087615,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000496021058554778,
|
|
"loss": 5.8195,
|
|
"mean_token_accuracy": 0.16271883994340897,
|
|
"num_tokens": 3360379.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.094673681259155,
|
|
"epoch": 1.6219578150351541,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004959403498749505,
|
|
"loss": 5.9599,
|
|
"mean_token_accuracy": 0.1654249593615532,
|
|
"num_tokens": 3370962.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"epoch": 1.6219578150351541,
|
|
"eval_entropy": 5.9815694711425085,
|
|
"eval_loss": 6.159132957458496,
|
|
"eval_mean_token_accuracy": 0.15803164498541844,
|
|
"eval_num_tokens": 3370962.0,
|
|
"eval_runtime": 1.9338,
|
|
"eval_samples_per_second": 1817.709,
|
|
"eval_steps_per_second": 227.537,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 6.149770641326905,
|
|
"epoch": 1.6273661438615468,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004958588382973809,
|
|
"loss": 5.8588,
|
|
"mean_token_accuracy": 0.1632590562105179,
|
|
"num_tokens": 3382648.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 6.097417306900025,
|
|
"epoch": 1.6327744726879394,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004957765241182826,
|
|
"loss": 5.9322,
|
|
"mean_token_accuracy": 0.16269588321447373,
|
|
"num_tokens": 3393871.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 6.062821912765503,
|
|
"epoch": 1.638182801514332,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004956934076367854,
|
|
"loss": 5.8248,
|
|
"mean_token_accuracy": 0.1695956364274025,
|
|
"num_tokens": 3405148.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 6.06344313621521,
|
|
"epoch": 1.6435911303407247,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000495609489154935,
|
|
"loss": 5.8578,
|
|
"mean_token_accuracy": 0.16674693375825883,
|
|
"num_tokens": 3416647.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 6.1670889377594,
|
|
"epoch": 1.6489994591671173,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004955247689776911,
|
|
"loss": 5.9581,
|
|
"mean_token_accuracy": 0.15195233076810838,
|
|
"num_tokens": 3428331.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 5.957519054412842,
|
|
"epoch": 1.65440778799351,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.000495439247412927,
|
|
"loss": 5.8089,
|
|
"mean_token_accuracy": 0.16562821343541145,
|
|
"num_tokens": 3440683.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 6.098293161392212,
|
|
"epoch": 1.6598161168199026,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004953529247714285,
|
|
"loss": 5.8107,
|
|
"mean_token_accuracy": 0.16624202877283095,
|
|
"num_tokens": 3452375.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 6.088880920410157,
|
|
"epoch": 1.6652244456462952,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004952658013668922,
|
|
"loss": 5.9661,
|
|
"mean_token_accuracy": 0.15399764329195023,
|
|
"num_tokens": 3463255.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 6.14812479019165,
|
|
"epoch": 1.670632774472688,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004951778775159246,
|
|
"loss": 5.8977,
|
|
"mean_token_accuracy": 0.1587236136198044,
|
|
"num_tokens": 3473642.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 6.053155517578125,
|
|
"epoch": 1.6760411032990805,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004950891535380414,
|
|
"loss": 5.8697,
|
|
"mean_token_accuracy": 0.1589012935757637,
|
|
"num_tokens": 3484877.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 6.028693771362304,
|
|
"epoch": 1.6814494321254734,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004949996297556656,
|
|
"loss": 5.7827,
|
|
"mean_token_accuracy": 0.16816994845867156,
|
|
"num_tokens": 3495500.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 6.077233839035034,
|
|
"epoch": 1.6868577609518658,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004949093064941268,
|
|
"loss": 5.9049,
|
|
"mean_token_accuracy": 0.16012717485427858,
|
|
"num_tokens": 3508057.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 6.154424333572388,
|
|
"epoch": 1.6922660897782587,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004948181840816601,
|
|
"loss": 6.0716,
|
|
"mean_token_accuracy": 0.15230329036712648,
|
|
"num_tokens": 3521110.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 6.148553371429443,
|
|
"epoch": 1.697674418604651,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004947262628494045,
|
|
"loss": 5.982,
|
|
"mean_token_accuracy": 0.1573754370212555,
|
|
"num_tokens": 3530692.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 6.131786251068116,
|
|
"epoch": 1.703082747431044,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004946335431314018,
|
|
"loss": 5.9229,
|
|
"mean_token_accuracy": 0.15379725843667985,
|
|
"num_tokens": 3542195.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 6.160162401199341,
|
|
"epoch": 1.7084910762574363,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004945400252645959,
|
|
"loss": 5.8228,
|
|
"mean_token_accuracy": 0.1686183452606201,
|
|
"num_tokens": 3552676.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 5.978310203552246,
|
|
"epoch": 1.7138994050838292,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004944457095888308,
|
|
"loss": 5.7861,
|
|
"mean_token_accuracy": 0.1679149240255356,
|
|
"num_tokens": 3563308.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 6.119216632843018,
|
|
"epoch": 1.7193077339102216,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00049435059644685,
|
|
"loss": 5.8139,
|
|
"mean_token_accuracy": 0.16842420399188995,
|
|
"num_tokens": 3574085.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 6.048505878448486,
|
|
"epoch": 1.7247160627366145,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004942546861842947,
|
|
"loss": 5.8908,
|
|
"mean_token_accuracy": 0.15561516880989074,
|
|
"num_tokens": 3585816.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 6.069142580032349,
|
|
"epoch": 1.730124391563007,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004941579791497032,
|
|
"loss": 5.9352,
|
|
"mean_token_accuracy": 0.163139209151268,
|
|
"num_tokens": 3597776.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 6.176851987838745,
|
|
"epoch": 1.7355327203893998,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004940604756945091,
|
|
"loss": 5.9478,
|
|
"mean_token_accuracy": 0.16177947521209718,
|
|
"num_tokens": 3607991.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 6.060487985610962,
|
|
"epoch": 1.7409410492157922,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004939621761730399,
|
|
"loss": 5.834,
|
|
"mean_token_accuracy": 0.1683316335082054,
|
|
"num_tokens": 3618204.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 6.032902002334595,
|
|
"epoch": 1.746349378042185,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004938630809425166,
|
|
"loss": 5.9359,
|
|
"mean_token_accuracy": 0.16037531346082687,
|
|
"num_tokens": 3629558.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 6.138170433044434,
|
|
"epoch": 1.7517577068685775,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004937631903630513,
|
|
"loss": 5.8846,
|
|
"mean_token_accuracy": 0.16198209524154664,
|
|
"num_tokens": 3640773.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 6.028919219970703,
|
|
"epoch": 1.7571660356949703,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.0004936625047976468,
|
|
"loss": 5.8051,
|
|
"mean_token_accuracy": 0.16324532479047776,
|
|
"num_tokens": 3652186.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 6.042243909835816,
|
|
"epoch": 1.7625743645213627,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004935610246121944,
|
|
"loss": 5.8584,
|
|
"mean_token_accuracy": 0.16714964359998702,
|
|
"num_tokens": 3663068.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 6.124779558181762,
|
|
"epoch": 1.7679826933477556,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004934587501754736,
|
|
"loss": 5.9893,
|
|
"mean_token_accuracy": 0.15739795863628386,
|
|
"num_tokens": 3673803.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 6.204698705673218,
|
|
"epoch": 1.7733910221741482,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004933556818591498,
|
|
"loss": 5.9585,
|
|
"mean_token_accuracy": 0.16012052744626998,
|
|
"num_tokens": 3685482.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 6.055734586715698,
|
|
"epoch": 1.7787993510005409,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0004932518200377736,
|
|
"loss": 5.8155,
|
|
"mean_token_accuracy": 0.17018294781446458,
|
|
"num_tokens": 3697641.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 6.003864526748657,
|
|
"epoch": 1.7842076798269335,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004931471650887789,
|
|
"loss": 5.8345,
|
|
"mean_token_accuracy": 0.16596303135156631,
|
|
"num_tokens": 3709560.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 6.159448099136353,
|
|
"epoch": 1.7896160086533262,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004930417173924822,
|
|
"loss": 5.9473,
|
|
"mean_token_accuracy": 0.1586322605609894,
|
|
"num_tokens": 3720422.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 5.931419420242309,
|
|
"epoch": 1.7950243374797188,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004929354773320806,
|
|
"loss": 5.7622,
|
|
"mean_token_accuracy": 0.16559358388185502,
|
|
"num_tokens": 3731104.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 6.1085484504699705,
|
|
"epoch": 1.8004326663061114,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004928284452936506,
|
|
"loss": 5.942,
|
|
"mean_token_accuracy": 0.15943334102630616,
|
|
"num_tokens": 3742766.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 6.073635673522949,
|
|
"epoch": 1.805840995132504,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004927206216661473,
|
|
"loss": 5.8155,
|
|
"mean_token_accuracy": 0.16575353890657424,
|
|
"num_tokens": 3753064.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 6.089341020584106,
|
|
"epoch": 1.8112493239588967,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004926120068414015,
|
|
"loss": 5.8684,
|
|
"mean_token_accuracy": 0.16442893743515014,
|
|
"num_tokens": 3765183.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 6.041831970214844,
|
|
"epoch": 1.8166576527852893,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004925026012141201,
|
|
"loss": 5.8593,
|
|
"mean_token_accuracy": 0.1572822079062462,
|
|
"num_tokens": 3777902.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 5.9712233543396,
|
|
"epoch": 1.822065981611682,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004923924051818832,
|
|
"loss": 5.8213,
|
|
"mean_token_accuracy": 0.16846658289432526,
|
|
"num_tokens": 3788461.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 6.023951196670533,
|
|
"epoch": 1.8274743104380746,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004922814191451435,
|
|
"loss": 5.7598,
|
|
"mean_token_accuracy": 0.16275863945484162,
|
|
"num_tokens": 3799574.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 6.105522155761719,
|
|
"epoch": 1.8328826392644673,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004921696435072245,
|
|
"loss": 5.9081,
|
|
"mean_token_accuracy": 0.16315366476774215,
|
|
"num_tokens": 3811889.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 5.98220419883728,
|
|
"epoch": 1.83829096809086,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004920570786743191,
|
|
"loss": 5.8564,
|
|
"mean_token_accuracy": 0.16141254901885987,
|
|
"num_tokens": 3822949.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 6.1289613246917725,
|
|
"epoch": 1.8436992969172525,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004919437250554882,
|
|
"loss": 5.9611,
|
|
"mean_token_accuracy": 0.15678673535585402,
|
|
"num_tokens": 3834734.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 6.035271692276001,
|
|
"epoch": 1.8491076257436452,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004918295830626592,
|
|
"loss": 5.8289,
|
|
"mean_token_accuracy": 0.1687936469912529,
|
|
"num_tokens": 3845508.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 5.979309129714966,
|
|
"epoch": 1.8545159545700378,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004917146531106242,
|
|
"loss": 5.8021,
|
|
"mean_token_accuracy": 0.16257864087820054,
|
|
"num_tokens": 3856500.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 6.1346900939941404,
|
|
"epoch": 1.8599242833964305,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004915989356170389,
|
|
"loss": 5.8853,
|
|
"mean_token_accuracy": 0.16002874821424484,
|
|
"num_tokens": 3867877.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 6.055261754989624,
|
|
"epoch": 1.865332612222823,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004914824310024213,
|
|
"loss": 5.8772,
|
|
"mean_token_accuracy": 0.16758613288402557,
|
|
"num_tokens": 3878613.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 6.0397374629974365,
|
|
"epoch": 1.870740941049216,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004913651396901489,
|
|
"loss": 5.7886,
|
|
"mean_token_accuracy": 0.17041865438222886,
|
|
"num_tokens": 3888418.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 6.031890869140625,
|
|
"epoch": 1.8761492698756084,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000491247062106459,
|
|
"loss": 5.8731,
|
|
"mean_token_accuracy": 0.16119478791952133,
|
|
"num_tokens": 3899359.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 6.117352247238159,
|
|
"epoch": 1.8815575987020012,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004911281986804455,
|
|
"loss": 5.9566,
|
|
"mean_token_accuracy": 0.1571228489279747,
|
|
"num_tokens": 3911568.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 6.012397336959839,
|
|
"epoch": 1.8869659275283936,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004910085498440587,
|
|
"loss": 5.9029,
|
|
"mean_token_accuracy": 0.159549742937088,
|
|
"num_tokens": 3921597.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 6.010352277755738,
|
|
"epoch": 1.8923742563547865,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004908881160321025,
|
|
"loss": 5.7973,
|
|
"mean_token_accuracy": 0.157119719684124,
|
|
"num_tokens": 3933124.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 6.074922895431518,
|
|
"epoch": 1.897782585181179,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004907668976822338,
|
|
"loss": 5.8111,
|
|
"mean_token_accuracy": 0.17123811542987824,
|
|
"num_tokens": 3944388.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 6.051175117492676,
|
|
"epoch": 1.9031909140075718,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004906448952349603,
|
|
"loss": 5.7864,
|
|
"mean_token_accuracy": 0.16755871921777726,
|
|
"num_tokens": 3955024.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 5.917784976959228,
|
|
"epoch": 1.9085992428339642,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004905221091336392,
|
|
"loss": 5.7913,
|
|
"mean_token_accuracy": 0.16885399520397187,
|
|
"num_tokens": 3966029.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 5.945049381256103,
|
|
"epoch": 1.914007571660357,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004903985398244753,
|
|
"loss": 5.7541,
|
|
"mean_token_accuracy": 0.16620747596025467,
|
|
"num_tokens": 3976066.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 6.134118318557739,
|
|
"epoch": 1.9194159004867495,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004902741877565201,
|
|
"loss": 5.9371,
|
|
"mean_token_accuracy": 0.15995117276906967,
|
|
"num_tokens": 3986467.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 6.0786679744720455,
|
|
"epoch": 1.9248242293131423,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000490149053381669,
|
|
"loss": 5.9014,
|
|
"mean_token_accuracy": 0.16296863704919815,
|
|
"num_tokens": 3998208.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 6.069469308853149,
|
|
"epoch": 1.9302325581395348,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004900231371546608,
|
|
"loss": 5.8687,
|
|
"mean_token_accuracy": 0.16599351018667222,
|
|
"num_tokens": 4008202.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 5.939991617202759,
|
|
"epoch": 1.9356408869659276,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004898964395330753,
|
|
"loss": 5.683,
|
|
"mean_token_accuracy": 0.17330397963523864,
|
|
"num_tokens": 4019256.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 5.936979627609253,
|
|
"epoch": 1.94104921579232,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004897689609773318,
|
|
"loss": 5.7298,
|
|
"mean_token_accuracy": 0.1674216002225876,
|
|
"num_tokens": 4029738.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 6.031766033172607,
|
|
"epoch": 1.946457544618713,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004896407019506879,
|
|
"loss": 5.9662,
|
|
"mean_token_accuracy": 0.15475598871707916,
|
|
"num_tokens": 4042277.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 6.056133079528808,
|
|
"epoch": 1.9518658734451053,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004895116629192371,
|
|
"loss": 5.825,
|
|
"mean_token_accuracy": 0.17290453016757965,
|
|
"num_tokens": 4054019.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 5.981416845321656,
|
|
"epoch": 1.9572742022714982,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004893818443519074,
|
|
"loss": 5.8948,
|
|
"mean_token_accuracy": 0.16223520785570145,
|
|
"num_tokens": 4065562.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 6.122199821472168,
|
|
"epoch": 1.9626825310978906,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00048925124672046,
|
|
"loss": 5.9286,
|
|
"mean_token_accuracy": 0.16140141487121581,
|
|
"num_tokens": 4076617.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 6.004253101348877,
|
|
"epoch": 1.9680908599242835,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004891198704994868,
|
|
"loss": 5.7389,
|
|
"mean_token_accuracy": 0.17377816438674926,
|
|
"num_tokens": 4087191.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 6.011477708816528,
|
|
"epoch": 1.973499188750676,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004889877161664096,
|
|
"loss": 5.7747,
|
|
"mean_token_accuracy": 0.16836816668510438,
|
|
"num_tokens": 4099150.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 5.9672447681427006,
|
|
"epoch": 1.9789075175770687,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004888547842014771,
|
|
"loss": 5.7796,
|
|
"mean_token_accuracy": 0.17283790111541747,
|
|
"num_tokens": 4110299.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 5.983891105651855,
|
|
"epoch": 1.9843158464034614,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004887210750877648,
|
|
"loss": 5.7988,
|
|
"mean_token_accuracy": 0.17068280428647994,
|
|
"num_tokens": 4121682.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 5.962355613708496,
|
|
"epoch": 1.989724175229854,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004885865893111716,
|
|
"loss": 5.7528,
|
|
"mean_token_accuracy": 0.1695406287908554,
|
|
"num_tokens": 4132914.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 6.000193357467651,
|
|
"epoch": 1.9951325040562466,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004884513273604195,
|
|
"loss": 5.8527,
|
|
"mean_token_accuracy": 0.15907907485961914,
|
|
"num_tokens": 4144213.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 6.045581499735515,
|
|
"epoch": 2.0,
|
|
"grad_norm": 2.125,
|
|
"learning_rate": 0.0004883152897270502,
|
|
"loss": 5.8794,
|
|
"mean_token_accuracy": 0.16449115839269426,
|
|
"num_tokens": 4153778.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 5.92071852684021,
|
|
"epoch": 2.005408328826393,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004881784769054253,
|
|
"loss": 5.4989,
|
|
"mean_token_accuracy": 0.1751373752951622,
|
|
"num_tokens": 4165536.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 6.05268406867981,
|
|
"epoch": 2.0108166576527853,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004880408893927226,
|
|
"loss": 5.5666,
|
|
"mean_token_accuracy": 0.17167191654443742,
|
|
"num_tokens": 4176182.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 5.833243799209595,
|
|
"epoch": 2.016224986479178,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004879025276889355,
|
|
"loss": 5.5587,
|
|
"mean_token_accuracy": 0.18386996537446976,
|
|
"num_tokens": 4187592.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 5.954632425308228,
|
|
"epoch": 2.0216333153055706,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00048776339229687077,
|
|
"loss": 5.3661,
|
|
"mean_token_accuracy": 0.1883542865514755,
|
|
"num_tokens": 4198199.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 5.8728643417358395,
|
|
"epoch": 2.0270416441319634,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048762348372214683,
|
|
"loss": 5.6029,
|
|
"mean_token_accuracy": 0.17609835863113404,
|
|
"num_tokens": 4208116.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 5.843668127059937,
|
|
"epoch": 2.032449972958356,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00048748280247319165,
|
|
"loss": 5.3806,
|
|
"mean_token_accuracy": 0.18249491304159166,
|
|
"num_tokens": 4218569.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 5.938153409957886,
|
|
"epoch": 2.0378583017847487,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004873413490612412,
|
|
"loss": 5.5383,
|
|
"mean_token_accuracy": 0.18427726328372956,
|
|
"num_tokens": 4228785.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 5.884716176986695,
|
|
"epoch": 2.043266630611141,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00048719912400033757,
|
|
"loss": 5.4917,
|
|
"mean_token_accuracy": 0.18152164071798324,
|
|
"num_tokens": 4238814.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 5.862728834152222,
|
|
"epoch": 2.048674959437534,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.000487056127807327,
|
|
"loss": 5.4584,
|
|
"mean_token_accuracy": 0.18107647448778152,
|
|
"num_tokens": 4250544.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 5.920808267593384,
|
|
"epoch": 2.0540832882639264,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00048691236100185795,
|
|
"loss": 5.5872,
|
|
"mean_token_accuracy": 0.1788840502500534,
|
|
"num_tokens": 4261806.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 5.972072172164917,
|
|
"epoch": 2.0594916170903192,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004867678241063792,
|
|
"loss": 5.5345,
|
|
"mean_token_accuracy": 0.18279671370983125,
|
|
"num_tokens": 4272472.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 5.872320556640625,
|
|
"epoch": 2.0648999459167117,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00048662251764613833,
|
|
"loss": 5.51,
|
|
"mean_token_accuracy": 0.1792232558131218,
|
|
"num_tokens": 4283758.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 5.891624689102173,
|
|
"epoch": 2.0703082747431045,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004864764421491793,
|
|
"loss": 5.6189,
|
|
"mean_token_accuracy": 0.17929046154022216,
|
|
"num_tokens": 4295930.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 6.070260953903198,
|
|
"epoch": 2.075716603569497,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004863295981463408,
|
|
"loss": 5.5463,
|
|
"mean_token_accuracy": 0.18622798025608062,
|
|
"num_tokens": 4306711.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 5.828749132156372,
|
|
"epoch": 2.08112493239589,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004861819861712543,
|
|
"loss": 5.5166,
|
|
"mean_token_accuracy": 0.18105367720127105,
|
|
"num_tokens": 4317694.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 5.99052472114563,
|
|
"epoch": 2.086533261222282,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048603360676034203,
|
|
"loss": 5.5996,
|
|
"mean_token_accuracy": 0.1750706747174263,
|
|
"num_tokens": 4329552.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 5.864745235443115,
|
|
"epoch": 2.091941590048675,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00048588446045281505,
|
|
"loss": 5.547,
|
|
"mean_token_accuracy": 0.17900061458349228,
|
|
"num_tokens": 4340479.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 5.828856182098389,
|
|
"epoch": 2.0973499188750675,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004857345477906714,
|
|
"loss": 5.4167,
|
|
"mean_token_accuracy": 0.18962016105651855,
|
|
"num_tokens": 4351147.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 5.7453841209411625,
|
|
"epoch": 2.1027582477014604,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00048558386931869417,
|
|
"loss": 5.4109,
|
|
"mean_token_accuracy": 0.18178877234458923,
|
|
"num_tokens": 4362962.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 5.89324722290039,
|
|
"epoch": 2.1081665765278528,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004854324255844492,
|
|
"loss": 5.5264,
|
|
"mean_token_accuracy": 0.1859842836856842,
|
|
"num_tokens": 4373723.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 5.784295654296875,
|
|
"epoch": 2.1135749053542456,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00048528021713828336,
|
|
"loss": 5.5126,
|
|
"mean_token_accuracy": 0.1793152168393135,
|
|
"num_tokens": 4383434.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 5.836202096939087,
|
|
"epoch": 2.118983234180638,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004851272445333225,
|
|
"loss": 5.4501,
|
|
"mean_token_accuracy": 0.18879707604646684,
|
|
"num_tokens": 4394847.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 5.928153038024902,
|
|
"epoch": 2.124391563007031,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00048497350832546937,
|
|
"loss": 5.5663,
|
|
"mean_token_accuracy": 0.18257277756929396,
|
|
"num_tokens": 4406367.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 5.943634080886841,
|
|
"epoch": 2.1297998918334233,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004848190090734018,
|
|
"loss": 5.5453,
|
|
"mean_token_accuracy": 0.18376570343971252,
|
|
"num_tokens": 4417636.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 5.86825156211853,
|
|
"epoch": 2.135208220659816,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00048466374733857035,
|
|
"loss": 5.5705,
|
|
"mean_token_accuracy": 0.17977122217416763,
|
|
"num_tokens": 4428473.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 5.888907814025879,
|
|
"epoch": 2.1406165494862086,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004845077236851966,
|
|
"loss": 5.5084,
|
|
"mean_token_accuracy": 0.1778701737523079,
|
|
"num_tokens": 4439780.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 5.801192140579223,
|
|
"epoch": 2.1460248783126015,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048435093868027076,
|
|
"loss": 5.453,
|
|
"mean_token_accuracy": 0.1903861328959465,
|
|
"num_tokens": 4450929.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 5.871478033065796,
|
|
"epoch": 2.151433207138994,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00048419339289354995,
|
|
"loss": 5.4884,
|
|
"mean_token_accuracy": 0.18074046522378923,
|
|
"num_tokens": 4462967.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 5.892301559448242,
|
|
"epoch": 2.1568415359653867,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000484035086897556,
|
|
"loss": 5.5777,
|
|
"mean_token_accuracy": 0.18284107744693756,
|
|
"num_tokens": 4473970.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 5.8949380874633786,
|
|
"epoch": 2.162249864791779,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004838760212675732,
|
|
"loss": 5.5684,
|
|
"mean_token_accuracy": 0.1804433599114418,
|
|
"num_tokens": 4485660.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 2.162249864791779,
|
|
"eval_entropy": 5.781203637339852,
|
|
"eval_loss": 6.039128303527832,
|
|
"eval_mean_token_accuracy": 0.166043197973208,
|
|
"eval_num_tokens": 4485660.0,
|
|
"eval_runtime": 1.928,
|
|
"eval_samples_per_second": 1823.117,
|
|
"eval_steps_per_second": 228.214,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 5.8628510475158695,
|
|
"epoch": 2.167658193618172,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004837161965816464,
|
|
"loss": 5.5494,
|
|
"mean_token_accuracy": 0.1852016642689705,
|
|
"num_tokens": 4497261.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 5.855096387863159,
|
|
"epoch": 2.1730665224445644,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00048355561342057886,
|
|
"loss": 5.436,
|
|
"mean_token_accuracy": 0.1877509295940399,
|
|
"num_tokens": 4508120.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 5.8603614330291744,
|
|
"epoch": 2.1784748512709573,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048339427236793024,
|
|
"loss": 5.5323,
|
|
"mean_token_accuracy": 0.18392907530069352,
|
|
"num_tokens": 4519286.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 5.954311037063599,
|
|
"epoch": 2.1838831800973497,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004832321740100143,
|
|
"loss": 5.5973,
|
|
"mean_token_accuracy": 0.18077260553836821,
|
|
"num_tokens": 4531912.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 5.883333683013916,
|
|
"epoch": 2.1892915089237426,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048306931893589666,
|
|
"loss": 5.6045,
|
|
"mean_token_accuracy": 0.18002939224243164,
|
|
"num_tokens": 4544370.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 5.849424171447754,
|
|
"epoch": 2.194699837750135,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00048290570773739327,
|
|
"loss": 5.5549,
|
|
"mean_token_accuracy": 0.18056917935609818,
|
|
"num_tokens": 4556013.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 5.834095525741577,
|
|
"epoch": 2.200108166576528,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048274134100906747,
|
|
"loss": 5.4963,
|
|
"mean_token_accuracy": 0.18100530803203582,
|
|
"num_tokens": 4566200.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 5.845457983016968,
|
|
"epoch": 2.2055164954029207,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048257621934822835,
|
|
"loss": 5.5482,
|
|
"mean_token_accuracy": 0.1808362513780594,
|
|
"num_tokens": 4576899.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 5.9733531951904295,
|
|
"epoch": 2.210924824229313,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004824103433549285,
|
|
"loss": 5.5573,
|
|
"mean_token_accuracy": 0.17765308022499085,
|
|
"num_tokens": 4588983.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 5.848635292053222,
|
|
"epoch": 2.2163331530557056,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004822437136319616,
|
|
"loss": 5.5024,
|
|
"mean_token_accuracy": 0.19122663736343384,
|
|
"num_tokens": 4599758.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 5.838743019104004,
|
|
"epoch": 2.2217414818820984,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004820763307848604,
|
|
"loss": 5.5567,
|
|
"mean_token_accuracy": 0.1786445513367653,
|
|
"num_tokens": 4610927.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 5.881386804580688,
|
|
"epoch": 2.2271498107084913,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048190819542189456,
|
|
"loss": 5.5703,
|
|
"mean_token_accuracy": 0.18060187101364136,
|
|
"num_tokens": 4623719.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 5.877337074279785,
|
|
"epoch": 2.2325581395348837,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048173930815406847,
|
|
"loss": 5.6264,
|
|
"mean_token_accuracy": 0.1777713567018509,
|
|
"num_tokens": 4635389.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 5.919419860839843,
|
|
"epoch": 2.2379664683612766,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004815696695951187,
|
|
"loss": 5.5731,
|
|
"mean_token_accuracy": 0.17647812217473985,
|
|
"num_tokens": 4645093.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 5.83864803314209,
|
|
"epoch": 2.243374797187669,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004813992803615123,
|
|
"loss": 5.5612,
|
|
"mean_token_accuracy": 0.17630923092365264,
|
|
"num_tokens": 4656490.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 5.885448217391968,
|
|
"epoch": 2.248783126014062,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048122814107244414,
|
|
"loss": 5.5828,
|
|
"mean_token_accuracy": 0.1766236513853073,
|
|
"num_tokens": 4667603.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 5.829516410827637,
|
|
"epoch": 2.2541914548404542,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00048105625234983466,
|
|
"loss": 5.5179,
|
|
"mean_token_accuracy": 0.17676828801631927,
|
|
"num_tokens": 4678669.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 5.76925892829895,
|
|
"epoch": 2.259599783666847,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00048088361481832807,
|
|
"loss": 5.3166,
|
|
"mean_token_accuracy": 0.19174162447452545,
|
|
"num_tokens": 4689757.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 5.735202646255493,
|
|
"epoch": 2.2650081124932395,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048071022910528934,
|
|
"loss": 5.5138,
|
|
"mean_token_accuracy": 0.17985685169696808,
|
|
"num_tokens": 4699332.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 5.899583721160889,
|
|
"epoch": 2.2704164413196324,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048053609584080274,
|
|
"loss": 5.5725,
|
|
"mean_token_accuracy": 0.18425035923719407,
|
|
"num_tokens": 4709485.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 5.814102363586426,
|
|
"epoch": 2.275824770146025,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048036121565766876,
|
|
"loss": 5.5217,
|
|
"mean_token_accuracy": 0.18288214206695558,
|
|
"num_tokens": 4720935.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 5.736945104598999,
|
|
"epoch": 2.2812330989724177,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004801855891914026,
|
|
"loss": 5.4625,
|
|
"mean_token_accuracy": 0.17907819151878357,
|
|
"num_tokens": 4731219.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 5.9266547679901125,
|
|
"epoch": 2.28664142779881,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048000921708023117,
|
|
"loss": 5.5924,
|
|
"mean_token_accuracy": 0.17558808326721193,
|
|
"num_tokens": 4742637.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 5.768679809570313,
|
|
"epoch": 2.292049756625203,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004798320999650911,
|
|
"loss": 5.4868,
|
|
"mean_token_accuracy": 0.1778075650334358,
|
|
"num_tokens": 4753341.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 6.01874303817749,
|
|
"epoch": 2.2974580854515954,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00047965423848962636,
|
|
"loss": 5.6302,
|
|
"mean_token_accuracy": 0.18140974789857864,
|
|
"num_tokens": 4765339.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 5.85342698097229,
|
|
"epoch": 2.302866414277988,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000479475633300186,
|
|
"loss": 5.4893,
|
|
"mean_token_accuracy": 0.18313317447900773,
|
|
"num_tokens": 4776149.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 5.7818248748779295,
|
|
"epoch": 2.3082747431043806,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00047929628504582173,
|
|
"loss": 5.52,
|
|
"mean_token_accuracy": 0.18565733581781388,
|
|
"num_tokens": 4788120.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 5.910288572311401,
|
|
"epoch": 2.3136830719307735,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004791161943782856,
|
|
"loss": 5.6061,
|
|
"mean_token_accuracy": 0.17473552376031876,
|
|
"num_tokens": 4799922.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 5.751388025283814,
|
|
"epoch": 2.319091400757166,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00047893536195202726,
|
|
"loss": 5.5193,
|
|
"mean_token_accuracy": 0.17882945090532304,
|
|
"num_tokens": 4810501.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 5.803964424133301,
|
|
"epoch": 2.3244997295835588,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004787537884241923,
|
|
"loss": 5.484,
|
|
"mean_token_accuracy": 0.18450351655483246,
|
|
"num_tokens": 4820919.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 5.933160972595215,
|
|
"epoch": 2.329908058409951,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00047857147445461943,
|
|
"loss": 5.5729,
|
|
"mean_token_accuracy": 0.18036166429519654,
|
|
"num_tokens": 4833298.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 5.6917722702026365,
|
|
"epoch": 2.335316387236344,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004783884207058379,
|
|
"loss": 5.4448,
|
|
"mean_token_accuracy": 0.18355783820152283,
|
|
"num_tokens": 4845199.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 5.787708139419555,
|
|
"epoch": 2.3407247160627365,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00047820462784306555,
|
|
"loss": 5.5153,
|
|
"mean_token_accuracy": 0.1812552183866501,
|
|
"num_tokens": 4856549.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 5.82427749633789,
|
|
"epoch": 2.3461330448891293,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000478020096534206,
|
|
"loss": 5.5589,
|
|
"mean_token_accuracy": 0.17584991902112962,
|
|
"num_tokens": 4868446.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 5.938240003585816,
|
|
"epoch": 2.3515413737155217,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00047783482744984644,
|
|
"loss": 5.6038,
|
|
"mean_token_accuracy": 0.1734725296497345,
|
|
"num_tokens": 4879047.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 5.851721334457397,
|
|
"epoch": 2.3569497025419146,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00047764882126325527,
|
|
"loss": 5.4952,
|
|
"mean_token_accuracy": 0.18024096488952637,
|
|
"num_tokens": 4891591.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 5.762294912338257,
|
|
"epoch": 2.362358031368307,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004774620786503793,
|
|
"loss": 5.5126,
|
|
"mean_token_accuracy": 0.18283341974020004,
|
|
"num_tokens": 4902710.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 5.901837968826294,
|
|
"epoch": 2.3677663601947,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004772746002898417,
|
|
"loss": 5.5798,
|
|
"mean_token_accuracy": 0.1801070749759674,
|
|
"num_tokens": 4914668.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 5.9564659118652346,
|
|
"epoch": 2.3731746890210923,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00047708638686293903,
|
|
"loss": 5.7156,
|
|
"mean_token_accuracy": 0.1746075138449669,
|
|
"num_tokens": 4926684.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 5.901554822921753,
|
|
"epoch": 2.378583017847485,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00047689743905363963,
|
|
"loss": 5.5811,
|
|
"mean_token_accuracy": 0.18086021691560744,
|
|
"num_tokens": 4938577.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 5.730812931060791,
|
|
"epoch": 2.383991346673878,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004767077575485801,
|
|
"loss": 5.4542,
|
|
"mean_token_accuracy": 0.19141294360160827,
|
|
"num_tokens": 4949070.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 5.8393768787384035,
|
|
"epoch": 2.3893996755002704,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004765173430370635,
|
|
"loss": 5.5104,
|
|
"mean_token_accuracy": 0.18252145498991013,
|
|
"num_tokens": 4959558.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 5.729151821136474,
|
|
"epoch": 2.394808004326663,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004763261962110567,
|
|
"loss": 5.3866,
|
|
"mean_token_accuracy": 0.1876178115606308,
|
|
"num_tokens": 4970310.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 5.719764328002929,
|
|
"epoch": 2.4002163331530557,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004761343177651875,
|
|
"loss": 5.4826,
|
|
"mean_token_accuracy": 0.1792425826191902,
|
|
"num_tokens": 4982300.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 5.820022058486939,
|
|
"epoch": 2.4056246619794486,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004759417083967428,
|
|
"loss": 5.5125,
|
|
"mean_token_accuracy": 0.18094677329063416,
|
|
"num_tokens": 4994425.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 5.796099042892456,
|
|
"epoch": 2.411032990805841,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00047574836880566544,
|
|
"loss": 5.5048,
|
|
"mean_token_accuracy": 0.18813187927007674,
|
|
"num_tokens": 5005872.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 5.757441282272339,
|
|
"epoch": 2.4164413196322334,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004755542996945519,
|
|
"loss": 5.5594,
|
|
"mean_token_accuracy": 0.18093689233064653,
|
|
"num_tokens": 5016903.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 5.891964101791382,
|
|
"epoch": 2.4218496484586263,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00047535950176864976,
|
|
"loss": 5.5853,
|
|
"mean_token_accuracy": 0.183075350522995,
|
|
"num_tokens": 5028407.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 5.827565765380859,
|
|
"epoch": 2.427257977285019,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00047516397573585525,
|
|
"loss": 5.5831,
|
|
"mean_token_accuracy": 0.17881640642881394,
|
|
"num_tokens": 5039693.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 5.8785138607025145,
|
|
"epoch": 2.4326663061114115,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00047496772230671034,
|
|
"loss": 5.5473,
|
|
"mean_token_accuracy": 0.1818876251578331,
|
|
"num_tokens": 5051279.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 5.867937183380127,
|
|
"epoch": 2.4380746349378044,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004747707421944004,
|
|
"loss": 5.4787,
|
|
"mean_token_accuracy": 0.18194486647844316,
|
|
"num_tokens": 5062698.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 5.789582538604736,
|
|
"epoch": 2.443482963764197,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004745730361147517,
|
|
"loss": 5.4718,
|
|
"mean_token_accuracy": 0.18270887881517411,
|
|
"num_tokens": 5074927.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 5.784351396560669,
|
|
"epoch": 2.4488912925905897,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004743746047862286,
|
|
"loss": 5.597,
|
|
"mean_token_accuracy": 0.17477912008762359,
|
|
"num_tokens": 5086298.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 5.859172534942627,
|
|
"epoch": 2.454299621416982,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004741754489299309,
|
|
"loss": 5.5359,
|
|
"mean_token_accuracy": 0.17866628170013427,
|
|
"num_tokens": 5098161.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 5.821074771881103,
|
|
"epoch": 2.459707950243375,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00047397556926959155,
|
|
"loss": 5.5296,
|
|
"mean_token_accuracy": 0.19062084555625916,
|
|
"num_tokens": 5110743.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 5.9080784797668455,
|
|
"epoch": 2.4651162790697674,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00047377496653157374,
|
|
"loss": 5.5719,
|
|
"mean_token_accuracy": 0.18251802325248717,
|
|
"num_tokens": 5121838.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 5.745440864562989,
|
|
"epoch": 2.4705246078961602,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00047357364144486806,
|
|
"loss": 5.5143,
|
|
"mean_token_accuracy": 0.17874232977628707,
|
|
"num_tokens": 5135047.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 5.7803637981414795,
|
|
"epoch": 2.4759329367225527,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00047337159474109057,
|
|
"loss": 5.5,
|
|
"mean_token_accuracy": 0.1746326968073845,
|
|
"num_tokens": 5146329.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 5.859038972854615,
|
|
"epoch": 2.4813412655489455,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00047316882715447934,
|
|
"loss": 5.5434,
|
|
"mean_token_accuracy": 0.18041068613529204,
|
|
"num_tokens": 5157511.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 5.75163950920105,
|
|
"epoch": 2.486749594375338,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004729653394218922,
|
|
"loss": 5.5235,
|
|
"mean_token_accuracy": 0.18174429088830948,
|
|
"num_tokens": 5168456.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 5.918718862533569,
|
|
"epoch": 2.492157923201731,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00047276113228280427,
|
|
"loss": 5.5794,
|
|
"mean_token_accuracy": 0.18340918868780137,
|
|
"num_tokens": 5179294.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 5.8278374671936035,
|
|
"epoch": 2.497566252028123,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004725562064793045,
|
|
"loss": 5.516,
|
|
"mean_token_accuracy": 0.18428950756788254,
|
|
"num_tokens": 5190195.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 5.703502178192139,
|
|
"epoch": 2.502974580854516,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00047235056275609397,
|
|
"loss": 5.4157,
|
|
"mean_token_accuracy": 0.1894952178001404,
|
|
"num_tokens": 5201289.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 5.767619371414185,
|
|
"epoch": 2.5083829096809085,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004721442018604822,
|
|
"loss": 5.4311,
|
|
"mean_token_accuracy": 0.19223134219646454,
|
|
"num_tokens": 5212295.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 5.673934698104858,
|
|
"epoch": 2.5137912385073014,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004719371245423853,
|
|
"loss": 5.4242,
|
|
"mean_token_accuracy": 0.1893549904227257,
|
|
"num_tokens": 5223144.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 5.739576244354248,
|
|
"epoch": 2.5191995673336938,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00047172933155432273,
|
|
"loss": 5.4333,
|
|
"mean_token_accuracy": 0.18600700199604034,
|
|
"num_tokens": 5232840.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 5.767184829711914,
|
|
"epoch": 2.5246078961600866,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00047152082365141466,
|
|
"loss": 5.5476,
|
|
"mean_token_accuracy": 0.18290351927280427,
|
|
"num_tokens": 5243796.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 5.8564368724823,
|
|
"epoch": 2.530016224986479,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004713116015913794,
|
|
"loss": 5.6325,
|
|
"mean_token_accuracy": 0.1742415815591812,
|
|
"num_tokens": 5255990.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 5.871094989776611,
|
|
"epoch": 2.535424553812872,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00047110166613453016,
|
|
"loss": 5.5791,
|
|
"mean_token_accuracy": 0.1803630471229553,
|
|
"num_tokens": 5268134.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 5.678581094741821,
|
|
"epoch": 2.5408328826392643,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00047089101804377305,
|
|
"loss": 5.4102,
|
|
"mean_token_accuracy": 0.18474622070789337,
|
|
"num_tokens": 5278662.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 5.755628442764282,
|
|
"epoch": 2.546241211465657,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004706796580846037,
|
|
"loss": 5.4834,
|
|
"mean_token_accuracy": 0.18088264018297195,
|
|
"num_tokens": 5288382.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 5.777779769897461,
|
|
"epoch": 2.5516495402920496,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00047046758702510465,
|
|
"loss": 5.5423,
|
|
"mean_token_accuracy": 0.1802200809121132,
|
|
"num_tokens": 5299769.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 5.9212644577026365,
|
|
"epoch": 2.5570578691184425,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00047025480563594264,
|
|
"loss": 5.4911,
|
|
"mean_token_accuracy": 0.1891256242990494,
|
|
"num_tokens": 5310662.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 5.712078428268432,
|
|
"epoch": 2.5624661979448353,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00047004131469036576,
|
|
"loss": 5.4646,
|
|
"mean_token_accuracy": 0.1791183829307556,
|
|
"num_tokens": 5321785.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 5.809269905090332,
|
|
"epoch": 2.5678745267712277,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00046982711496420057,
|
|
"loss": 5.5208,
|
|
"mean_token_accuracy": 0.17982801496982576,
|
|
"num_tokens": 5333557.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 5.819075345993042,
|
|
"epoch": 2.57328285559762,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00046961220723584925,
|
|
"loss": 5.5675,
|
|
"mean_token_accuracy": 0.17790673077106475,
|
|
"num_tokens": 5344767.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 5.779632806777954,
|
|
"epoch": 2.578691184424013,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00046939659228628725,
|
|
"loss": 5.4793,
|
|
"mean_token_accuracy": 0.1805744394659996,
|
|
"num_tokens": 5356673.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 5.819158697128296,
|
|
"epoch": 2.584099513250406,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00046918027089905956,
|
|
"loss": 5.5918,
|
|
"mean_token_accuracy": 0.1753435656428337,
|
|
"num_tokens": 5367296.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 5.748660039901734,
|
|
"epoch": 2.5895078420767983,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00046896324386027873,
|
|
"loss": 5.4642,
|
|
"mean_token_accuracy": 0.1818123459815979,
|
|
"num_tokens": 5378081.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 5.837896490097046,
|
|
"epoch": 2.5949161709031907,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00046874551195862145,
|
|
"loss": 5.5489,
|
|
"mean_token_accuracy": 0.17460184395313263,
|
|
"num_tokens": 5390244.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 5.709093618392944,
|
|
"epoch": 2.6003244997295836,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000468527075985326,
|
|
"loss": 5.4206,
|
|
"mean_token_accuracy": 0.1866423398256302,
|
|
"num_tokens": 5400741.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 5.815622282028198,
|
|
"epoch": 2.6057328285559764,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00046830793673418924,
|
|
"loss": 5.5975,
|
|
"mean_token_accuracy": 0.18263377994298935,
|
|
"num_tokens": 5411440.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 5.801127624511719,
|
|
"epoch": 2.611141157382369,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00046808809500156355,
|
|
"loss": 5.4639,
|
|
"mean_token_accuracy": 0.18024436831474305,
|
|
"num_tokens": 5422472.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 5.722740030288696,
|
|
"epoch": 2.6165494862087613,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004678675515863545,
|
|
"loss": 5.5464,
|
|
"mean_token_accuracy": 0.17724482715129852,
|
|
"num_tokens": 5433958.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 5.778017187118531,
|
|
"epoch": 2.621957815035154,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004676463072900173,
|
|
"loss": 5.4411,
|
|
"mean_token_accuracy": 0.19543344378471375,
|
|
"num_tokens": 5444343.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 5.7358941555023195,
|
|
"epoch": 2.627366143861547,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00046742436291655415,
|
|
"loss": 5.4672,
|
|
"mean_token_accuracy": 0.18050388097763062,
|
|
"num_tokens": 5456260.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 5.778013134002686,
|
|
"epoch": 2.6327744726879394,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00046720171927251155,
|
|
"loss": 5.5595,
|
|
"mean_token_accuracy": 0.18578180819749832,
|
|
"num_tokens": 5468281.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 5.67826361656189,
|
|
"epoch": 2.638182801514332,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004669783771669769,
|
|
"loss": 5.4068,
|
|
"mean_token_accuracy": 0.195494344830513,
|
|
"num_tokens": 5480068.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 5.677945947647094,
|
|
"epoch": 2.6435911303407247,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000466754337411576,
|
|
"loss": 5.3989,
|
|
"mean_token_accuracy": 0.18041235506534575,
|
|
"num_tokens": 5490932.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 5.68883695602417,
|
|
"epoch": 2.6489994591671175,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00046652960082046985,
|
|
"loss": 5.5018,
|
|
"mean_token_accuracy": 0.18080479502677918,
|
|
"num_tokens": 5501787.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 5.743030023574829,
|
|
"epoch": 2.65440778799351,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00046630416821035164,
|
|
"loss": 5.4801,
|
|
"mean_token_accuracy": 0.18007345795631408,
|
|
"num_tokens": 5513297.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 5.738504076004029,
|
|
"epoch": 2.6598161168199024,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004660780404004441,
|
|
"loss": 5.4581,
|
|
"mean_token_accuracy": 0.18802803605794907,
|
|
"num_tokens": 5523797.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 5.699677991867065,
|
|
"epoch": 2.6652244456462952,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00046585121821249623,
|
|
"loss": 5.4979,
|
|
"mean_token_accuracy": 0.192328979074955,
|
|
"num_tokens": 5535808.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 5.840962505340576,
|
|
"epoch": 2.670632774472688,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00046562370247078025,
|
|
"loss": 5.5797,
|
|
"mean_token_accuracy": 0.17371595948934554,
|
|
"num_tokens": 5548179.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 5.83707594871521,
|
|
"epoch": 2.6760411032990805,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000465395494002089,
|
|
"loss": 5.5934,
|
|
"mean_token_accuracy": 0.17932454347610474,
|
|
"num_tokens": 5559144.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 5.7753153324127195,
|
|
"epoch": 2.6814494321254734,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004651665936357325,
|
|
"loss": 5.4796,
|
|
"mean_token_accuracy": 0.18297361135482787,
|
|
"num_tokens": 5571103.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 5.699805498123169,
|
|
"epoch": 2.686857760951866,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004649370022035354,
|
|
"loss": 5.4017,
|
|
"mean_token_accuracy": 0.18377552926540375,
|
|
"num_tokens": 5583239.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 5.702951192855835,
|
|
"epoch": 2.6922660897782587,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004647067205398332,
|
|
"loss": 5.3836,
|
|
"mean_token_accuracy": 0.18936756253242493,
|
|
"num_tokens": 5593602.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 5.80911922454834,
|
|
"epoch": 2.697674418604651,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00046447574948147027,
|
|
"loss": 5.5501,
|
|
"mean_token_accuracy": 0.1801131248474121,
|
|
"num_tokens": 5604696.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 5.749916315078735,
|
|
"epoch": 2.703082747431044,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00046424408986779587,
|
|
"loss": 5.5228,
|
|
"mean_token_accuracy": 0.18397949188947677,
|
|
"num_tokens": 5615668.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"epoch": 2.703082747431044,
|
|
"eval_entropy": 5.6709360068494625,
|
|
"eval_loss": 5.93572473526001,
|
|
"eval_mean_token_accuracy": 0.17236155767671085,
|
|
"eval_num_tokens": 5615668.0,
|
|
"eval_runtime": 1.9281,
|
|
"eval_samples_per_second": 1823.04,
|
|
"eval_steps_per_second": 228.204,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 5.761583089828491,
|
|
"epoch": 2.7084910762574363,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004640117425406615,
|
|
"loss": 5.4619,
|
|
"mean_token_accuracy": 0.18198691457509994,
|
|
"num_tokens": 5627344.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 5.761633968353271,
|
|
"epoch": 2.713899405083829,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000463778708344418,
|
|
"loss": 5.4871,
|
|
"mean_token_accuracy": 0.17963782995939254,
|
|
"num_tokens": 5637472.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 5.726103782653809,
|
|
"epoch": 2.7193077339102216,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00046354498812591207,
|
|
"loss": 5.4858,
|
|
"mean_token_accuracy": 0.1846323385834694,
|
|
"num_tokens": 5647766.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 5.665884637832642,
|
|
"epoch": 2.7247160627366145,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00046331058273448364,
|
|
"loss": 5.4059,
|
|
"mean_token_accuracy": 0.18580469489097595,
|
|
"num_tokens": 5658343.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 5.725853347778321,
|
|
"epoch": 2.730124391563007,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004630754930219624,
|
|
"loss": 5.3959,
|
|
"mean_token_accuracy": 0.1896525055170059,
|
|
"num_tokens": 5669640.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 5.617433071136475,
|
|
"epoch": 2.7355327203893998,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004628397198426648,
|
|
"loss": 5.3875,
|
|
"mean_token_accuracy": 0.17847614586353303,
|
|
"num_tokens": 5682421.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 5.701791667938233,
|
|
"epoch": 2.740941049215792,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00046260326405339126,
|
|
"loss": 5.3657,
|
|
"mean_token_accuracy": 0.19074373990297316,
|
|
"num_tokens": 5692797.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 5.759147357940674,
|
|
"epoch": 2.746349378042185,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004623661265134226,
|
|
"loss": 5.5506,
|
|
"mean_token_accuracy": 0.18549878895282745,
|
|
"num_tokens": 5702779.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 5.791321182250977,
|
|
"epoch": 2.7517577068685775,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004621283080845172,
|
|
"loss": 5.5228,
|
|
"mean_token_accuracy": 0.1855258360505104,
|
|
"num_tokens": 5714147.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 5.6605089664459225,
|
|
"epoch": 2.7571660356949703,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00046188980963090783,
|
|
"loss": 5.3695,
|
|
"mean_token_accuracy": 0.19154659062623977,
|
|
"num_tokens": 5725718.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 5.684929466247558,
|
|
"epoch": 2.7625743645213627,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00046165063201929825,
|
|
"loss": 5.4512,
|
|
"mean_token_accuracy": 0.18923905193805696,
|
|
"num_tokens": 5737441.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 5.7932765007019045,
|
|
"epoch": 2.7679826933477556,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00046141077611886074,
|
|
"loss": 5.5537,
|
|
"mean_token_accuracy": 0.18146031498908996,
|
|
"num_tokens": 5748735.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 5.86720724105835,
|
|
"epoch": 2.773391022174148,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004611702428012319,
|
|
"loss": 5.5994,
|
|
"mean_token_accuracy": 0.1744624376296997,
|
|
"num_tokens": 5760262.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 5.69360876083374,
|
|
"epoch": 2.778799351000541,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00046092903294051057,
|
|
"loss": 5.5074,
|
|
"mean_token_accuracy": 0.18178244829177856,
|
|
"num_tokens": 5771909.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 5.600863695144653,
|
|
"epoch": 2.7842076798269337,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00046068714741325385,
|
|
"loss": 5.3953,
|
|
"mean_token_accuracy": 0.187199105322361,
|
|
"num_tokens": 5782028.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 5.855996894836426,
|
|
"epoch": 2.789616008653326,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004604445870984743,
|
|
"loss": 5.5433,
|
|
"mean_token_accuracy": 0.18408290147781373,
|
|
"num_tokens": 5793641.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 5.810896348953247,
|
|
"epoch": 2.7950243374797186,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004602013528776367,
|
|
"loss": 5.4894,
|
|
"mean_token_accuracy": 0.1814708650112152,
|
|
"num_tokens": 5805392.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 5.730097723007202,
|
|
"epoch": 2.8004326663061114,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004599574456346548,
|
|
"loss": 5.4681,
|
|
"mean_token_accuracy": 0.18465488255023957,
|
|
"num_tokens": 5816362.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 5.692508840560913,
|
|
"epoch": 2.8058409951325043,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00045971286625588796,
|
|
"loss": 5.4051,
|
|
"mean_token_accuracy": 0.18702610284090043,
|
|
"num_tokens": 5827518.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 5.823817777633667,
|
|
"epoch": 2.8112493239588967,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004594676156301383,
|
|
"loss": 5.5418,
|
|
"mean_token_accuracy": 0.18507158160209655,
|
|
"num_tokens": 5838455.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 5.692004680633545,
|
|
"epoch": 2.816657652785289,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00045922169464864725,
|
|
"loss": 5.4214,
|
|
"mean_token_accuracy": 0.18905350863933562,
|
|
"num_tokens": 5850065.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 5.760145282745361,
|
|
"epoch": 2.822065981611682,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000458975104205092,
|
|
"loss": 5.5392,
|
|
"mean_token_accuracy": 0.1797675982117653,
|
|
"num_tokens": 5860444.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 5.838974380493164,
|
|
"epoch": 2.827474310438075,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00045872784519558304,
|
|
"loss": 5.5214,
|
|
"mean_token_accuracy": 0.18014506101608277,
|
|
"num_tokens": 5873984.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 5.708714008331299,
|
|
"epoch": 2.8328826392644673,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004584799185186599,
|
|
"loss": 5.5432,
|
|
"mean_token_accuracy": 0.18505633622407913,
|
|
"num_tokens": 5885807.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 5.69213171005249,
|
|
"epoch": 2.8382909680908597,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00045823132507528903,
|
|
"loss": 5.4259,
|
|
"mean_token_accuracy": 0.1890205755829811,
|
|
"num_tokens": 5897122.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 5.682660722732544,
|
|
"epoch": 2.8436992969172525,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004579820657688593,
|
|
"loss": 5.4322,
|
|
"mean_token_accuracy": 0.18685290813446045,
|
|
"num_tokens": 5909189.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 5.7754181861877445,
|
|
"epoch": 2.8491076257436454,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004577321415051798,
|
|
"loss": 5.5229,
|
|
"mean_token_accuracy": 0.17633915394544603,
|
|
"num_tokens": 5920135.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 5.774295663833618,
|
|
"epoch": 2.854515954570038,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00045748155319247564,
|
|
"loss": 5.505,
|
|
"mean_token_accuracy": 0.18883296996355056,
|
|
"num_tokens": 5932279.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 5.654811239242553,
|
|
"epoch": 2.8599242833964302,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004572303017413856,
|
|
"loss": 5.4297,
|
|
"mean_token_accuracy": 0.1837768778204918,
|
|
"num_tokens": 5942082.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 5.7737048149108885,
|
|
"epoch": 2.865332612222823,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004569783880649578,
|
|
"loss": 5.4819,
|
|
"mean_token_accuracy": 0.18235635608434678,
|
|
"num_tokens": 5952856.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 5.608963918685913,
|
|
"epoch": 2.870740941049216,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004567258130786474,
|
|
"loss": 5.3882,
|
|
"mean_token_accuracy": 0.1902555212378502,
|
|
"num_tokens": 5963937.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 5.749698257446289,
|
|
"epoch": 2.8761492698756084,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00045647257770031215,
|
|
"loss": 5.4559,
|
|
"mean_token_accuracy": 0.18624959141016006,
|
|
"num_tokens": 5974852.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 5.716911697387696,
|
|
"epoch": 2.8815575987020012,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00045621868285021023,
|
|
"loss": 5.4713,
|
|
"mean_token_accuracy": 0.1868917316198349,
|
|
"num_tokens": 5985616.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 5.724991273880005,
|
|
"epoch": 2.8869659275283936,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00045596412945099615,
|
|
"loss": 5.4173,
|
|
"mean_token_accuracy": 0.1917577862739563,
|
|
"num_tokens": 5997199.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 5.721166038513184,
|
|
"epoch": 2.8923742563547865,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00045570891842771727,
|
|
"loss": 5.4732,
|
|
"mean_token_accuracy": 0.18812378197908403,
|
|
"num_tokens": 6008654.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 5.751707315444946,
|
|
"epoch": 2.897782585181179,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00045545305070781135,
|
|
"loss": 5.5244,
|
|
"mean_token_accuracy": 0.18074266314506532,
|
|
"num_tokens": 6020662.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 5.737688779830933,
|
|
"epoch": 2.903190914007572,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00045519652722110206,
|
|
"loss": 5.4109,
|
|
"mean_token_accuracy": 0.2018731340765953,
|
|
"num_tokens": 6031793.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 5.680935049057007,
|
|
"epoch": 2.908599242833964,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00045493934889979655,
|
|
"loss": 5.3914,
|
|
"mean_token_accuracy": 0.19068088084459306,
|
|
"num_tokens": 6042181.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 5.580568599700928,
|
|
"epoch": 2.914007571660357,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00045468151667848116,
|
|
"loss": 5.3457,
|
|
"mean_token_accuracy": 0.19399455934762955,
|
|
"num_tokens": 6052774.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 5.776365852355957,
|
|
"epoch": 2.9194159004867495,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000454423031494119,
|
|
"loss": 5.3801,
|
|
"mean_token_accuracy": 0.19814668893814086,
|
|
"num_tokens": 6063273.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 5.641749668121338,
|
|
"epoch": 2.9248242293131423,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004541638942860458,
|
|
"loss": 5.3857,
|
|
"mean_token_accuracy": 0.1883971869945526,
|
|
"num_tokens": 6075121.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 5.563340091705323,
|
|
"epoch": 2.9302325581395348,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004539041059959667,
|
|
"loss": 5.3579,
|
|
"mean_token_accuracy": 0.18671021461486817,
|
|
"num_tokens": 6085666.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 5.682371330261231,
|
|
"epoch": 2.9356408869659276,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00045364366756795295,
|
|
"loss": 5.3795,
|
|
"mean_token_accuracy": 0.1973988965153694,
|
|
"num_tokens": 6096619.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 5.762141609191895,
|
|
"epoch": 2.94104921579232,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00045338257994843845,
|
|
"loss": 5.5387,
|
|
"mean_token_accuracy": 0.18189419358968734,
|
|
"num_tokens": 6108800.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 5.74236831665039,
|
|
"epoch": 2.946457544618713,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004531208440862162,
|
|
"loss": 5.4583,
|
|
"mean_token_accuracy": 0.18336043655872344,
|
|
"num_tokens": 6119789.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 5.630492210388184,
|
|
"epoch": 2.9518658734451053,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00045285846093243485,
|
|
"loss": 5.4389,
|
|
"mean_token_accuracy": 0.18647215962409974,
|
|
"num_tokens": 6131786.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 5.80919828414917,
|
|
"epoch": 2.957274202271498,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00045259543144059557,
|
|
"loss": 5.5263,
|
|
"mean_token_accuracy": 0.18384593725204468,
|
|
"num_tokens": 6143012.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 5.719038486480713,
|
|
"epoch": 2.9626825310978906,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004523317565665479,
|
|
"loss": 5.4015,
|
|
"mean_token_accuracy": 0.1907029166817665,
|
|
"num_tokens": 6153972.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 5.703077077865601,
|
|
"epoch": 2.9680908599242835,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00045206743726848716,
|
|
"loss": 5.5821,
|
|
"mean_token_accuracy": 0.17692120671272277,
|
|
"num_tokens": 6166202.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 5.633232641220093,
|
|
"epoch": 2.973499188750676,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004518024745069502,
|
|
"loss": 5.2955,
|
|
"mean_token_accuracy": 0.19987683147192,
|
|
"num_tokens": 6177253.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 5.6717733383178714,
|
|
"epoch": 2.9789075175770687,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004515368692448123,
|
|
"loss": 5.5737,
|
|
"mean_token_accuracy": 0.17942172437906265,
|
|
"num_tokens": 6188857.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 5.79138708114624,
|
|
"epoch": 2.9843158464034616,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00045127062244728353,
|
|
"loss": 5.5201,
|
|
"mean_token_accuracy": 0.17747295051813125,
|
|
"num_tokens": 6199945.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 5.7016438961029055,
|
|
"epoch": 2.989724175229854,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004510037350819055,
|
|
"loss": 5.5043,
|
|
"mean_token_accuracy": 0.18678925037384034,
|
|
"num_tokens": 6209866.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 5.75241117477417,
|
|
"epoch": 2.9951325040562464,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.00045073620811854744,
|
|
"loss": 5.4992,
|
|
"mean_token_accuracy": 0.17622313499450684,
|
|
"num_tokens": 6221636.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 5.672918478647868,
|
|
"epoch": 3.0,
|
|
"grad_norm": 2.203125,
|
|
"learning_rate": 0.00045046804252940294,
|
|
"loss": 5.3729,
|
|
"mean_token_accuracy": 0.18995803760157692,
|
|
"num_tokens": 6230667.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 5.71522889137268,
|
|
"epoch": 3.005408328826393,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00045019923928898644,
|
|
"loss": 5.1674,
|
|
"mean_token_accuracy": 0.20795877426862716,
|
|
"num_tokens": 6242242.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 5.758365535736084,
|
|
"epoch": 3.0108166576527853,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004499297993741294,
|
|
"loss": 5.198,
|
|
"mean_token_accuracy": 0.19910841882228852,
|
|
"num_tokens": 6253337.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 5.590553951263428,
|
|
"epoch": 3.016224986479178,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004496597237639773,
|
|
"loss": 5.0968,
|
|
"mean_token_accuracy": 0.20121320188045502,
|
|
"num_tokens": 6264582.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 5.688187074661255,
|
|
"epoch": 3.0216333153055706,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00044938901343998536,
|
|
"loss": 5.1225,
|
|
"mean_token_accuracy": 0.21101101785898208,
|
|
"num_tokens": 6276172.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 5.615479421615601,
|
|
"epoch": 3.0270416441319634,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004491176693859156,
|
|
"loss": 4.9837,
|
|
"mean_token_accuracy": 0.22034686654806138,
|
|
"num_tokens": 6286038.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 5.574426507949829,
|
|
"epoch": 3.032449972958356,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004488456925878329,
|
|
"loss": 5.0812,
|
|
"mean_token_accuracy": 0.20797832012176515,
|
|
"num_tokens": 6296546.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 5.635595321655273,
|
|
"epoch": 3.0378583017847487,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00044857308403410163,
|
|
"loss": 5.1657,
|
|
"mean_token_accuracy": 0.21319493353366853,
|
|
"num_tokens": 6308001.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 5.6665009498596195,
|
|
"epoch": 3.043266630611141,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00044829984471538183,
|
|
"loss": 5.1758,
|
|
"mean_token_accuracy": 0.19861980825662612,
|
|
"num_tokens": 6320896.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 5.716265916824341,
|
|
"epoch": 3.048674959437534,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004480259756246258,
|
|
"loss": 5.247,
|
|
"mean_token_accuracy": 0.19757085740566255,
|
|
"num_tokens": 6332710.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 5.661328935623169,
|
|
"epoch": 3.0540832882639264,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00044775147775707465,
|
|
"loss": 5.1193,
|
|
"mean_token_accuracy": 0.20423852056264877,
|
|
"num_tokens": 6343563.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 5.571894311904908,
|
|
"epoch": 3.0594916170903192,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000447476352110254,
|
|
"loss": 5.2229,
|
|
"mean_token_accuracy": 0.20744810849428177,
|
|
"num_tokens": 6354064.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 5.621117734909058,
|
|
"epoch": 3.0648999459167117,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004472005996839712,
|
|
"loss": 5.1263,
|
|
"mean_token_accuracy": 0.2032922774553299,
|
|
"num_tokens": 6366551.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 5.582510614395142,
|
|
"epoch": 3.0703082747431045,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004469242214803112,
|
|
"loss": 5.0556,
|
|
"mean_token_accuracy": 0.20877836942672728,
|
|
"num_tokens": 6377086.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 5.650656747817993,
|
|
"epoch": 3.075716603569497,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004466472185036331,
|
|
"loss": 5.1701,
|
|
"mean_token_accuracy": 0.20498382300138474,
|
|
"num_tokens": 6388795.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 5.648554182052612,
|
|
"epoch": 3.08112493239589,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004463695917605663,
|
|
"loss": 5.1666,
|
|
"mean_token_accuracy": 0.20774204283952713,
|
|
"num_tokens": 6400174.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 5.600169658660889,
|
|
"epoch": 3.086533261222282,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000446091342260007,
|
|
"loss": 5.1142,
|
|
"mean_token_accuracy": 0.2033505067229271,
|
|
"num_tokens": 6411297.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 5.5910125255584715,
|
|
"epoch": 3.091941590048675,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004458124710131145,
|
|
"loss": 5.1462,
|
|
"mean_token_accuracy": 0.20421304404735566,
|
|
"num_tokens": 6423487.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 5.72054705619812,
|
|
"epoch": 3.0973499188750675,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00044553297903330766,
|
|
"loss": 5.3422,
|
|
"mean_token_accuracy": 0.19205281138420105,
|
|
"num_tokens": 6435460.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 5.74812421798706,
|
|
"epoch": 3.1027582477014604,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00044525286733626085,
|
|
"loss": 5.2598,
|
|
"mean_token_accuracy": 0.2062794089317322,
|
|
"num_tokens": 6445664.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 5.606708955764771,
|
|
"epoch": 3.1081665765278528,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004449721369399008,
|
|
"loss": 5.1243,
|
|
"mean_token_accuracy": 0.20224094688892363,
|
|
"num_tokens": 6457182.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 5.550483560562133,
|
|
"epoch": 3.1135749053542456,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00044469078886440227,
|
|
"loss": 4.9953,
|
|
"mean_token_accuracy": 0.21901510357856752,
|
|
"num_tokens": 6468378.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 5.597189617156983,
|
|
"epoch": 3.118983234180638,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00044440882413218484,
|
|
"loss": 5.2089,
|
|
"mean_token_accuracy": 0.2041062295436859,
|
|
"num_tokens": 6479625.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 5.585499286651611,
|
|
"epoch": 3.124391563007031,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00044412624376790905,
|
|
"loss": 5.1143,
|
|
"mean_token_accuracy": 0.20767037719488143,
|
|
"num_tokens": 6490713.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 5.538257694244384,
|
|
"epoch": 3.1297998918334233,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004438430487984726,
|
|
"loss": 5.1737,
|
|
"mean_token_accuracy": 0.1948377251625061,
|
|
"num_tokens": 6502161.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 5.620874691009521,
|
|
"epoch": 3.135208220659816,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004435592402530066,
|
|
"loss": 5.0946,
|
|
"mean_token_accuracy": 0.2029889017343521,
|
|
"num_tokens": 6513639.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 5.549866485595703,
|
|
"epoch": 3.1406165494862086,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000443274819162872,
|
|
"loss": 5.1063,
|
|
"mean_token_accuracy": 0.2103414848446846,
|
|
"num_tokens": 6524893.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 5.462497997283935,
|
|
"epoch": 3.1460248783126015,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004429897865616557,
|
|
"loss": 5.103,
|
|
"mean_token_accuracy": 0.2072642683982849,
|
|
"num_tokens": 6535797.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 5.660726404190063,
|
|
"epoch": 3.151433207138994,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004427041434851668,
|
|
"loss": 5.2122,
|
|
"mean_token_accuracy": 0.19863755255937576,
|
|
"num_tokens": 6547117.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 5.688783025741577,
|
|
"epoch": 3.1568415359653867,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00044241789097143293,
|
|
"loss": 5.2377,
|
|
"mean_token_accuracy": 0.20508474260568618,
|
|
"num_tokens": 6558771.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 5.510453128814698,
|
|
"epoch": 3.162249864791779,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00044213103006069635,
|
|
"loss": 5.1885,
|
|
"mean_token_accuracy": 0.19535900950431823,
|
|
"num_tokens": 6569922.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 5.695133113861084,
|
|
"epoch": 3.167658193618172,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00044184356179541035,
|
|
"loss": 5.1933,
|
|
"mean_token_accuracy": 0.20060930252075196,
|
|
"num_tokens": 6580561.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 5.553457403182984,
|
|
"epoch": 3.1730665224445644,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004415554872202352,
|
|
"loss": 5.0304,
|
|
"mean_token_accuracy": 0.20982445776462555,
|
|
"num_tokens": 6591636.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 5.580715703964233,
|
|
"epoch": 3.1784748512709573,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00044126680738203466,
|
|
"loss": 5.2507,
|
|
"mean_token_accuracy": 0.20107281804084778,
|
|
"num_tokens": 6603736.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 5.721163749694824,
|
|
"epoch": 3.1838831800973497,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004409775233298718,
|
|
"loss": 5.1828,
|
|
"mean_token_accuracy": 0.20554571151733397,
|
|
"num_tokens": 6615415.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 5.559901380538941,
|
|
"epoch": 3.1892915089237426,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004406876361150056,
|
|
"loss": 5.2233,
|
|
"mean_token_accuracy": 0.19856353998184204,
|
|
"num_tokens": 6625970.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 5.632291460037232,
|
|
"epoch": 3.194699837750135,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004403971467908869,
|
|
"loss": 5.1647,
|
|
"mean_token_accuracy": 0.19651600122451782,
|
|
"num_tokens": 6636326.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 5.587097072601319,
|
|
"epoch": 3.200108166576528,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004401060564131545,
|
|
"loss": 5.202,
|
|
"mean_token_accuracy": 0.2035299852490425,
|
|
"num_tokens": 6646825.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 5.617121410369873,
|
|
"epoch": 3.2055164954029207,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004398143660396315,
|
|
"loss": 5.2015,
|
|
"mean_token_accuracy": 0.20159223824739456,
|
|
"num_tokens": 6657748.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 5.583178424835205,
|
|
"epoch": 3.210924824229313,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004395220767303215,
|
|
"loss": 5.1047,
|
|
"mean_token_accuracy": 0.21189998239278793,
|
|
"num_tokens": 6668422.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 5.585834217071533,
|
|
"epoch": 3.2163331530557056,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004392291895474045,
|
|
"loss": 5.1819,
|
|
"mean_token_accuracy": 0.20014864057302476,
|
|
"num_tokens": 6678898.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 5.667119741439819,
|
|
"epoch": 3.2217414818820984,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004389357055552331,
|
|
"loss": 5.2188,
|
|
"mean_token_accuracy": 0.20729435980319977,
|
|
"num_tokens": 6688691.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 5.576155710220337,
|
|
"epoch": 3.2271498107084913,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00043864162582032897,
|
|
"loss": 5.1007,
|
|
"mean_token_accuracy": 0.2111068606376648,
|
|
"num_tokens": 6700279.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 5.550846433639526,
|
|
"epoch": 3.2325581395348837,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004383469514113784,
|
|
"loss": 5.0379,
|
|
"mean_token_accuracy": 0.21483268588781357,
|
|
"num_tokens": 6710440.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 5.549640083312989,
|
|
"epoch": 3.2379664683612766,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004380516833992289,
|
|
"loss": 5.2199,
|
|
"mean_token_accuracy": 0.20157641917467117,
|
|
"num_tokens": 6722347.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 5.6257239818573,
|
|
"epoch": 3.243374797187669,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000437755822856885,
|
|
"loss": 5.2397,
|
|
"mean_token_accuracy": 0.20499618500471115,
|
|
"num_tokens": 6732936.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 3.243374797187669,
|
|
"eval_entropy": 5.435353024439379,
|
|
"eval_loss": 5.9132513999938965,
|
|
"eval_mean_token_accuracy": 0.17473087745972654,
|
|
"eval_num_tokens": 6732936.0,
|
|
"eval_runtime": 1.9417,
|
|
"eval_samples_per_second": 1810.257,
|
|
"eval_steps_per_second": 226.604,
|
|
"step": 3000
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 9240,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 10,
|
|
"save_steps": 500,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 1561509553029120.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|