6046 lines
164 KiB
JSON
6046 lines
164 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.29655990510083036,
|
|
"eval_steps": 3000,
|
|
"global_step": 3000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.742559432983398,
|
|
"epoch": 0.0004942665085013839,
|
|
"grad_norm": 5.34375,
|
|
"learning_rate": 2e-06,
|
|
"loss": 10.7918,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 11289.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.742565155029297,
|
|
"epoch": 0.0009885330170027679,
|
|
"grad_norm": 5.21875,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 10.7677,
|
|
"mean_token_accuracy": 0.00010660981060937048,
|
|
"num_tokens": 21541.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.74260196685791,
|
|
"epoch": 0.0014827995255041518,
|
|
"grad_norm": 5.28125,
|
|
"learning_rate": 7e-06,
|
|
"loss": 10.7469,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 32257.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.742596530914307,
|
|
"epoch": 0.0019770660340055358,
|
|
"grad_norm": 5.09375,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 10.7082,
|
|
"mean_token_accuracy": 0.00043735416838899257,
|
|
"num_tokens": 44176.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.742553424835204,
|
|
"epoch": 0.0024713325425069197,
|
|
"grad_norm": 4.4375,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 10.6063,
|
|
"mean_token_accuracy": 0.005003847437910736,
|
|
"num_tokens": 56115.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.742230701446534,
|
|
"epoch": 0.0029655990510083037,
|
|
"grad_norm": 4.0,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 10.4933,
|
|
"mean_token_accuracy": 0.02219230718910694,
|
|
"num_tokens": 68828.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.74111728668213,
|
|
"epoch": 0.0034598655595096876,
|
|
"grad_norm": 3.390625,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 10.3232,
|
|
"mean_token_accuracy": 0.037596833519637586,
|
|
"num_tokens": 80097.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.73831787109375,
|
|
"epoch": 0.0039541320680110716,
|
|
"grad_norm": 2.546875,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 10.1706,
|
|
"mean_token_accuracy": 0.042161832191050054,
|
|
"num_tokens": 91510.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.734669494628907,
|
|
"epoch": 0.004448398576512456,
|
|
"grad_norm": 2.25,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 10.0532,
|
|
"mean_token_accuracy": 0.04492984525859356,
|
|
"num_tokens": 103443.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.731356620788574,
|
|
"epoch": 0.0049426650850138395,
|
|
"grad_norm": 2.171875,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 9.9944,
|
|
"mean_token_accuracy": 0.04110746793448925,
|
|
"num_tokens": 114629.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.72984676361084,
|
|
"epoch": 0.005436931593515224,
|
|
"grad_norm": 1.984375,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.9354,
|
|
"mean_token_accuracy": 0.0402316827327013,
|
|
"num_tokens": 125012.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.72827968597412,
|
|
"epoch": 0.005931198102016607,
|
|
"grad_norm": 1.96875,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.8745,
|
|
"mean_token_accuracy": 0.0415823869407177,
|
|
"num_tokens": 135834.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.726681423187255,
|
|
"epoch": 0.006425464610517992,
|
|
"grad_norm": 1.90625,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.8094,
|
|
"mean_token_accuracy": 0.0378471277654171,
|
|
"num_tokens": 147413.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.725395107269287,
|
|
"epoch": 0.006919731119019375,
|
|
"grad_norm": 2.0,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.7766,
|
|
"mean_token_accuracy": 0.04534405171871185,
|
|
"num_tokens": 158046.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.722041702270507,
|
|
"epoch": 0.00741399762752076,
|
|
"grad_norm": 2.0,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.7003,
|
|
"mean_token_accuracy": 0.04153196755796671,
|
|
"num_tokens": 168564.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.717565536499023,
|
|
"epoch": 0.007908264136022143,
|
|
"grad_norm": 2.0,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.6649,
|
|
"mean_token_accuracy": 0.040395910106599334,
|
|
"num_tokens": 181168.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.712557983398437,
|
|
"epoch": 0.008402530644523527,
|
|
"grad_norm": 1.984375,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.5927,
|
|
"mean_token_accuracy": 0.040117064118385316,
|
|
"num_tokens": 192666.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.706313228607177,
|
|
"epoch": 0.008896797153024912,
|
|
"grad_norm": 1.9609375,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 9.5324,
|
|
"mean_token_accuracy": 0.04307716451585293,
|
|
"num_tokens": 203688.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.696890449523925,
|
|
"epoch": 0.009391063661526295,
|
|
"grad_norm": 1.9609375,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 9.4292,
|
|
"mean_token_accuracy": 0.04244718514382839,
|
|
"num_tokens": 214399.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.68424196243286,
|
|
"epoch": 0.009885330170027679,
|
|
"grad_norm": 1.8046875,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 9.4064,
|
|
"mean_token_accuracy": 0.03980784080922604,
|
|
"num_tokens": 227549.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.669907379150391,
|
|
"epoch": 0.010379596678529062,
|
|
"grad_norm": 2.09375,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 9.2501,
|
|
"mean_token_accuracy": 0.05166649520397186,
|
|
"num_tokens": 239612.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.647725677490234,
|
|
"epoch": 0.010873863187030448,
|
|
"grad_norm": 1.890625,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 9.1891,
|
|
"mean_token_accuracy": 0.05595755912363529,
|
|
"num_tokens": 251163.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 10.614391994476318,
|
|
"epoch": 0.011368129695531831,
|
|
"grad_norm": 1.984375,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 9.0675,
|
|
"mean_token_accuracy": 0.0543642945587635,
|
|
"num_tokens": 263995.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 10.580703353881836,
|
|
"epoch": 0.011862396204033215,
|
|
"grad_norm": 1.765625,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 9.0334,
|
|
"mean_token_accuracy": 0.05128365270793438,
|
|
"num_tokens": 275103.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 10.53465690612793,
|
|
"epoch": 0.012356662712534598,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.9026,
|
|
"mean_token_accuracy": 0.05212775617837906,
|
|
"num_tokens": 286838.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 10.470762634277344,
|
|
"epoch": 0.012850929221035983,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 8.8066,
|
|
"mean_token_accuracy": 0.05225841067731381,
|
|
"num_tokens": 298750.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 10.38624849319458,
|
|
"epoch": 0.013345195729537367,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 8.6142,
|
|
"mean_token_accuracy": 0.06268712393939495,
|
|
"num_tokens": 310451.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 10.291066646575928,
|
|
"epoch": 0.01383946223803875,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 8.5675,
|
|
"mean_token_accuracy": 0.05591914691030979,
|
|
"num_tokens": 321232.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 10.217416381835937,
|
|
"epoch": 0.014333728746540134,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 8.426,
|
|
"mean_token_accuracy": 0.06510277166962623,
|
|
"num_tokens": 332531.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 10.091275501251221,
|
|
"epoch": 0.01482799525504152,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 8.3371,
|
|
"mean_token_accuracy": 0.060455797240138054,
|
|
"num_tokens": 346002.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 9.966124248504638,
|
|
"epoch": 0.015322261763542903,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 8.1965,
|
|
"mean_token_accuracy": 0.05812167599797249,
|
|
"num_tokens": 356691.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 9.774528980255127,
|
|
"epoch": 0.015816528272044286,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 8.1137,
|
|
"mean_token_accuracy": 0.06779846139252185,
|
|
"num_tokens": 368749.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 9.564791774749756,
|
|
"epoch": 0.01631079478054567,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.9752,
|
|
"mean_token_accuracy": 0.06878971830010414,
|
|
"num_tokens": 379484.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 9.371025466918946,
|
|
"epoch": 0.016805061289047053,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.9252,
|
|
"mean_token_accuracy": 0.065206453576684,
|
|
"num_tokens": 389751.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 9.126683902740478,
|
|
"epoch": 0.017299327797548437,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.8734,
|
|
"mean_token_accuracy": 0.06619854085147381,
|
|
"num_tokens": 400771.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.916962146759033,
|
|
"epoch": 0.017793594306049824,
|
|
"grad_norm": 0.7734375,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.8145,
|
|
"mean_token_accuracy": 0.07182282060384751,
|
|
"num_tokens": 413575.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.716337394714355,
|
|
"epoch": 0.018287860814551207,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.7284,
|
|
"mean_token_accuracy": 0.07276704236865043,
|
|
"num_tokens": 424825.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 8.58464298248291,
|
|
"epoch": 0.01878212732305259,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.7373,
|
|
"mean_token_accuracy": 0.07814876064658165,
|
|
"num_tokens": 435610.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 8.413296604156494,
|
|
"epoch": 0.019276393831553974,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.6992,
|
|
"mean_token_accuracy": 0.07458564341068268,
|
|
"num_tokens": 445855.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 8.39066514968872,
|
|
"epoch": 0.019770660340055358,
|
|
"grad_norm": 0.7578125,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.5975,
|
|
"mean_token_accuracy": 0.07509246952831745,
|
|
"num_tokens": 456895.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 8.318842697143555,
|
|
"epoch": 0.02026492684855674,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.6975,
|
|
"mean_token_accuracy": 0.07416085563600064,
|
|
"num_tokens": 467446.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 8.206272411346436,
|
|
"epoch": 0.020759193357058125,
|
|
"grad_norm": 0.765625,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.6073,
|
|
"mean_token_accuracy": 0.07490024492144584,
|
|
"num_tokens": 478988.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 8.203931045532226,
|
|
"epoch": 0.02125345986555951,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.6028,
|
|
"mean_token_accuracy": 0.0804927557706833,
|
|
"num_tokens": 491053.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 8.194980716705322,
|
|
"epoch": 0.021747726374060895,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.6693,
|
|
"mean_token_accuracy": 0.07337137088179588,
|
|
"num_tokens": 502279.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 8.176432991027832,
|
|
"epoch": 0.02224199288256228,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.5236,
|
|
"mean_token_accuracy": 0.08306471854448319,
|
|
"num_tokens": 513470.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 8.073592710494996,
|
|
"epoch": 0.022736259391063662,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.5779,
|
|
"mean_token_accuracy": 0.07856447473168374,
|
|
"num_tokens": 524380.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 8.152504539489746,
|
|
"epoch": 0.023230525899565046,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.5837,
|
|
"mean_token_accuracy": 0.08571179807186127,
|
|
"num_tokens": 535798.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 8.032267951965332,
|
|
"epoch": 0.02372479240806643,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.4937,
|
|
"mean_token_accuracy": 0.08659510985016823,
|
|
"num_tokens": 545720.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 8.007299280166626,
|
|
"epoch": 0.024219058916567813,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.5277,
|
|
"mean_token_accuracy": 0.0827151209115982,
|
|
"num_tokens": 557110.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 8.001366901397706,
|
|
"epoch": 0.024713325425069196,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.4975,
|
|
"mean_token_accuracy": 0.08423097357153893,
|
|
"num_tokens": 568966.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 8.03173599243164,
|
|
"epoch": 0.02520759193357058,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.4969,
|
|
"mean_token_accuracy": 0.08669283166527748,
|
|
"num_tokens": 580441.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 8.057562732696534,
|
|
"epoch": 0.025701858442071967,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.4982,
|
|
"mean_token_accuracy": 0.08434496074914932,
|
|
"num_tokens": 592698.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.953376913070679,
|
|
"epoch": 0.02619612495057335,
|
|
"grad_norm": 0.79296875,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.4669,
|
|
"mean_token_accuracy": 0.08737746998667717,
|
|
"num_tokens": 604787.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.998782920837402,
|
|
"epoch": 0.026690391459074734,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 7.4532,
|
|
"mean_token_accuracy": 0.08598756566643714,
|
|
"num_tokens": 616445.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.9657186508178714,
|
|
"epoch": 0.027184657967576117,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.4802,
|
|
"mean_token_accuracy": 0.08283287510275841,
|
|
"num_tokens": 627723.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 8.00689115524292,
|
|
"epoch": 0.0276789244760775,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 7.5189,
|
|
"mean_token_accuracy": 0.08700487017631531,
|
|
"num_tokens": 638987.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.936768245697022,
|
|
"epoch": 0.028173190984578884,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.334,
|
|
"mean_token_accuracy": 0.09142324700951576,
|
|
"num_tokens": 650626.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.860474920272827,
|
|
"epoch": 0.028667457493080268,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.3572,
|
|
"mean_token_accuracy": 0.09691204503178596,
|
|
"num_tokens": 661911.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.903319931030273,
|
|
"epoch": 0.02916172400158165,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.4067,
|
|
"mean_token_accuracy": 0.09089745208621025,
|
|
"num_tokens": 674298.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.882914972305298,
|
|
"epoch": 0.02965599051008304,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 7.3443,
|
|
"mean_token_accuracy": 0.0963382288813591,
|
|
"num_tokens": 686000.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.80755648612976,
|
|
"epoch": 0.030150257018584422,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000152,
|
|
"loss": 7.362,
|
|
"mean_token_accuracy": 0.0944715328514576,
|
|
"num_tokens": 698464.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.860896110534668,
|
|
"epoch": 0.030644523527085805,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 7.379,
|
|
"mean_token_accuracy": 0.09046986699104309,
|
|
"num_tokens": 709702.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.968799591064453,
|
|
"epoch": 0.03113879003558719,
|
|
"grad_norm": 2.203125,
|
|
"learning_rate": 0.000157,
|
|
"loss": 7.4846,
|
|
"mean_token_accuracy": 0.0890492059290409,
|
|
"num_tokens": 722234.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.776518249511719,
|
|
"epoch": 0.03163305654408857,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 7.4068,
|
|
"mean_token_accuracy": 0.0861303061246872,
|
|
"num_tokens": 733001.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.8235656261444095,
|
|
"epoch": 0.03212732305258996,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.000162,
|
|
"loss": 7.3732,
|
|
"mean_token_accuracy": 0.09485758617520332,
|
|
"num_tokens": 743315.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.8506021976470945,
|
|
"epoch": 0.03262158956109134,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 7.3153,
|
|
"mean_token_accuracy": 0.09164077565073966,
|
|
"num_tokens": 754386.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.832219028472901,
|
|
"epoch": 0.033115856069592726,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 7.4069,
|
|
"mean_token_accuracy": 0.09743229635059833,
|
|
"num_tokens": 766504.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.739070034027099,
|
|
"epoch": 0.033610122578094107,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 7.2425,
|
|
"mean_token_accuracy": 0.10150521248579025,
|
|
"num_tokens": 777039.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.80741925239563,
|
|
"epoch": 0.034104389086595494,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 7.3291,
|
|
"mean_token_accuracy": 0.09152510985732079,
|
|
"num_tokens": 787890.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.745412302017212,
|
|
"epoch": 0.034598655595096874,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 7.2991,
|
|
"mean_token_accuracy": 0.09431157857179642,
|
|
"num_tokens": 800129.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.7933450698852536,
|
|
"epoch": 0.03509292210359826,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.000177,
|
|
"loss": 7.3748,
|
|
"mean_token_accuracy": 0.08783866837620735,
|
|
"num_tokens": 812517.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.6894533157348635,
|
|
"epoch": 0.03558718861209965,
|
|
"grad_norm": 0.8203125,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 7.2982,
|
|
"mean_token_accuracy": 0.09622510373592377,
|
|
"num_tokens": 824570.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.741975736618042,
|
|
"epoch": 0.03608145512060103,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.000182,
|
|
"loss": 7.2824,
|
|
"mean_token_accuracy": 0.09666576460003853,
|
|
"num_tokens": 836506.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.75258502960205,
|
|
"epoch": 0.036575721629102415,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 7.3016,
|
|
"mean_token_accuracy": 0.09653399288654327,
|
|
"num_tokens": 846841.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.799630975723266,
|
|
"epoch": 0.037069988137603795,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000187,
|
|
"loss": 7.3364,
|
|
"mean_token_accuracy": 0.10044656544923783,
|
|
"num_tokens": 858770.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.680880880355835,
|
|
"epoch": 0.03756425464610518,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 7.1812,
|
|
"mean_token_accuracy": 0.09969076812267304,
|
|
"num_tokens": 870184.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.6216026782989506,
|
|
"epoch": 0.03805852115460656,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.000192,
|
|
"loss": 7.2248,
|
|
"mean_token_accuracy": 0.10294909402728081,
|
|
"num_tokens": 881843.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.7437238693237305,
|
|
"epoch": 0.03855278766310795,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 7.2613,
|
|
"mean_token_accuracy": 0.09593532085418702,
|
|
"num_tokens": 893453.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.66199746131897,
|
|
"epoch": 0.03904705417160933,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 7.3009,
|
|
"mean_token_accuracy": 0.08956974968314171,
|
|
"num_tokens": 904417.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.71559739112854,
|
|
"epoch": 0.039541320680110716,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 7.223,
|
|
"mean_token_accuracy": 0.09316879063844681,
|
|
"num_tokens": 915134.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.676138925552368,
|
|
"epoch": 0.0400355871886121,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.000202,
|
|
"loss": 7.2417,
|
|
"mean_token_accuracy": 0.09755214676260948,
|
|
"num_tokens": 926361.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.596137046813965,
|
|
"epoch": 0.04052985369711348,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 7.1619,
|
|
"mean_token_accuracy": 0.09669081121683121,
|
|
"num_tokens": 937474.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.70306715965271,
|
|
"epoch": 0.04102412020561487,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000207,
|
|
"loss": 7.2242,
|
|
"mean_token_accuracy": 0.10178504958748817,
|
|
"num_tokens": 948897.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.578281736373901,
|
|
"epoch": 0.04151838671411625,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 7.2445,
|
|
"mean_token_accuracy": 0.09385722428560257,
|
|
"num_tokens": 960669.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.606976747512817,
|
|
"epoch": 0.04201265322261764,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000212,
|
|
"loss": 7.097,
|
|
"mean_token_accuracy": 0.10118152871727944,
|
|
"num_tokens": 973001.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.5266501903533936,
|
|
"epoch": 0.04250691973111902,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 7.1573,
|
|
"mean_token_accuracy": 0.1002947136759758,
|
|
"num_tokens": 986214.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.537716960906982,
|
|
"epoch": 0.043001186239620404,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 7.1675,
|
|
"mean_token_accuracy": 0.09322678521275521,
|
|
"num_tokens": 997363.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.530834531784057,
|
|
"epoch": 0.04349545274812179,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 7.092,
|
|
"mean_token_accuracy": 0.10840804129838943,
|
|
"num_tokens": 1009543.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.578037929534912,
|
|
"epoch": 0.04398971925662317,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000222,
|
|
"loss": 7.1736,
|
|
"mean_token_accuracy": 0.0948575533926487,
|
|
"num_tokens": 1020561.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.589271831512451,
|
|
"epoch": 0.04448398576512456,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 7.1256,
|
|
"mean_token_accuracy": 0.09639807492494583,
|
|
"num_tokens": 1032587.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.5890984535217285,
|
|
"epoch": 0.04497825227362594,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 7.1359,
|
|
"mean_token_accuracy": 0.09905373975634575,
|
|
"num_tokens": 1044249.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.547998476028442,
|
|
"epoch": 0.045472518782127325,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 7.1744,
|
|
"mean_token_accuracy": 0.09762973859906196,
|
|
"num_tokens": 1055415.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.5688958168029785,
|
|
"epoch": 0.045966785290628705,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 7.1466,
|
|
"mean_token_accuracy": 0.09966595321893693,
|
|
"num_tokens": 1066593.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.529252672195435,
|
|
"epoch": 0.04646105179913009,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 7.1369,
|
|
"mean_token_accuracy": 0.10283891633152961,
|
|
"num_tokens": 1076966.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.504359388351441,
|
|
"epoch": 0.04695531830763147,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.000237,
|
|
"loss": 7.1676,
|
|
"mean_token_accuracy": 0.09871028363704681,
|
|
"num_tokens": 1089927.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.551082754135132,
|
|
"epoch": 0.04744958481613286,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 7.1255,
|
|
"mean_token_accuracy": 0.09758824929594993,
|
|
"num_tokens": 1101920.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.489306211471558,
|
|
"epoch": 0.047943851324634246,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000242,
|
|
"loss": 7.1816,
|
|
"mean_token_accuracy": 0.10394024401903153,
|
|
"num_tokens": 1115078.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.6370282649993895,
|
|
"epoch": 0.048438117833135626,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 7.2542,
|
|
"mean_token_accuracy": 0.09503929167985917,
|
|
"num_tokens": 1127138.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.449513244628906,
|
|
"epoch": 0.04893238434163701,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000247,
|
|
"loss": 7.0705,
|
|
"mean_token_accuracy": 0.0948334887623787,
|
|
"num_tokens": 1138390.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.407353210449219,
|
|
"epoch": 0.04942665085013839,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 7.0009,
|
|
"mean_token_accuracy": 0.10578849464654923,
|
|
"num_tokens": 1149124.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.487055921554566,
|
|
"epoch": 0.04992091735863978,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000252,
|
|
"loss": 7.0787,
|
|
"mean_token_accuracy": 0.10215417668223381,
|
|
"num_tokens": 1161091.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.527550268173218,
|
|
"epoch": 0.05041518386714116,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 7.062,
|
|
"mean_token_accuracy": 0.10265694484114647,
|
|
"num_tokens": 1171919.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.452037620544433,
|
|
"epoch": 0.05090945037564255,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000257,
|
|
"loss": 7.0832,
|
|
"mean_token_accuracy": 0.10110483095049858,
|
|
"num_tokens": 1183688.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.405346298217774,
|
|
"epoch": 0.051403716884143934,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 7.012,
|
|
"mean_token_accuracy": 0.0998148925602436,
|
|
"num_tokens": 1195323.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.492317199707031,
|
|
"epoch": 0.051897983392645314,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.000262,
|
|
"loss": 7.1082,
|
|
"mean_token_accuracy": 0.09472725465893746,
|
|
"num_tokens": 1206782.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 7.3935158252716064,
|
|
"epoch": 0.0523922499011467,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 7.0087,
|
|
"mean_token_accuracy": 0.10457788184285163,
|
|
"num_tokens": 1218102.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.366737365722656,
|
|
"epoch": 0.05288651640964808,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.979,
|
|
"mean_token_accuracy": 0.10683351457118988,
|
|
"num_tokens": 1230199.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 7.456813716888428,
|
|
"epoch": 0.05338078291814947,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 7.1109,
|
|
"mean_token_accuracy": 0.09833906963467598,
|
|
"num_tokens": 1242127.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 7.489393091201782,
|
|
"epoch": 0.05387504942665085,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.9608,
|
|
"mean_token_accuracy": 0.10372178331017494,
|
|
"num_tokens": 1252329.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 7.225747203826904,
|
|
"epoch": 0.054369315935152235,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.9648,
|
|
"mean_token_accuracy": 0.10731954500079155,
|
|
"num_tokens": 1264832.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 7.370330858230591,
|
|
"epoch": 0.054863582443653615,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000277,
|
|
"loss": 7.041,
|
|
"mean_token_accuracy": 0.10336996018886566,
|
|
"num_tokens": 1276708.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 7.453816318511963,
|
|
"epoch": 0.055357848952155,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.9314,
|
|
"mean_token_accuracy": 0.11433103457093238,
|
|
"num_tokens": 1288089.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 7.364833116531372,
|
|
"epoch": 0.05585211546065639,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 7.0866,
|
|
"mean_token_accuracy": 0.09704122841358184,
|
|
"num_tokens": 1299411.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 7.38861722946167,
|
|
"epoch": 0.05634638196915777,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 7.0718,
|
|
"mean_token_accuracy": 0.09769185036420822,
|
|
"num_tokens": 1311368.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 7.383206033706665,
|
|
"epoch": 0.056840648477659156,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.9885,
|
|
"mean_token_accuracy": 0.10403251945972443,
|
|
"num_tokens": 1321962.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 7.367150115966797,
|
|
"epoch": 0.057334914986160536,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.9718,
|
|
"mean_token_accuracy": 0.10916787758469582,
|
|
"num_tokens": 1335150.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 7.376563596725464,
|
|
"epoch": 0.05782918149466192,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.000292,
|
|
"loss": 7.0122,
|
|
"mean_token_accuracy": 0.10337089449167251,
|
|
"num_tokens": 1346986.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 7.361315155029297,
|
|
"epoch": 0.0583234480031633,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.951,
|
|
"mean_token_accuracy": 0.10719594955444336,
|
|
"num_tokens": 1357820.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 7.284826040267944,
|
|
"epoch": 0.05881771451166469,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.929,
|
|
"mean_token_accuracy": 0.11087205111980439,
|
|
"num_tokens": 1368269.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 7.249915885925293,
|
|
"epoch": 0.05931198102016608,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.9547,
|
|
"mean_token_accuracy": 0.10180922672152519,
|
|
"num_tokens": 1380364.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 7.395902109146118,
|
|
"epoch": 0.05980624752866746,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.9082,
|
|
"mean_token_accuracy": 0.11050013303756714,
|
|
"num_tokens": 1390667.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 7.20961561203003,
|
|
"epoch": 0.060300514037168844,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.8605,
|
|
"mean_token_accuracy": 0.10693677067756653,
|
|
"num_tokens": 1401452.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 7.270598030090332,
|
|
"epoch": 0.060794780545670224,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.8652,
|
|
"mean_token_accuracy": 0.10748695507645607,
|
|
"num_tokens": 1412947.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 7.267968559265137,
|
|
"epoch": 0.06128904705417161,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.923,
|
|
"mean_token_accuracy": 0.10039002001285553,
|
|
"num_tokens": 1425123.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 7.293630933761596,
|
|
"epoch": 0.06178331356267299,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.8859,
|
|
"mean_token_accuracy": 0.10513018518686294,
|
|
"num_tokens": 1436538.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 7.242138385772705,
|
|
"epoch": 0.06227758007117438,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.9292,
|
|
"mean_token_accuracy": 0.1024887353181839,
|
|
"num_tokens": 1448269.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 7.26104211807251,
|
|
"epoch": 0.06277184657967576,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.904,
|
|
"mean_token_accuracy": 0.11182300597429276,
|
|
"num_tokens": 1459729.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 7.269317722320556,
|
|
"epoch": 0.06326611308817714,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.9178,
|
|
"mean_token_accuracy": 0.10927679911255836,
|
|
"num_tokens": 1470684.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 7.160611963272094,
|
|
"epoch": 0.06376037959667853,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.8722,
|
|
"mean_token_accuracy": 0.10590490102767944,
|
|
"num_tokens": 1483571.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 7.248231744766235,
|
|
"epoch": 0.06425464610517992,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.8625,
|
|
"mean_token_accuracy": 0.113711516559124,
|
|
"num_tokens": 1494671.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 7.255923461914063,
|
|
"epoch": 0.06474891261368129,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.8742,
|
|
"mean_token_accuracy": 0.10942275002598763,
|
|
"num_tokens": 1505435.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 7.311585712432861,
|
|
"epoch": 0.06524317912218268,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.9542,
|
|
"mean_token_accuracy": 0.10462261438369751,
|
|
"num_tokens": 1518510.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 7.2363609790802,
|
|
"epoch": 0.06573744563068407,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.8723,
|
|
"mean_token_accuracy": 0.11201808229088783,
|
|
"num_tokens": 1530218.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 7.211708068847656,
|
|
"epoch": 0.06623171213918545,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.9092,
|
|
"mean_token_accuracy": 0.10970637276768684,
|
|
"num_tokens": 1542486.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 7.21020245552063,
|
|
"epoch": 0.06672597864768683,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.9146,
|
|
"mean_token_accuracy": 0.10766631141304969,
|
|
"num_tokens": 1553873.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 7.245796585083008,
|
|
"epoch": 0.06722024515618821,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.8705,
|
|
"mean_token_accuracy": 0.11166855990886689,
|
|
"num_tokens": 1565343.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 7.190037250518799,
|
|
"epoch": 0.0677145116646896,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.9103,
|
|
"mean_token_accuracy": 0.10737355202436447,
|
|
"num_tokens": 1577061.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 7.27009220123291,
|
|
"epoch": 0.06820877817319099,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.9156,
|
|
"mean_token_accuracy": 0.10903515294194221,
|
|
"num_tokens": 1588083.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 7.203428411483765,
|
|
"epoch": 0.06870304468169237,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.856,
|
|
"mean_token_accuracy": 0.10634591579437255,
|
|
"num_tokens": 1599796.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 7.246427440643311,
|
|
"epoch": 0.06919731119019375,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.9223,
|
|
"mean_token_accuracy": 0.10735742151737213,
|
|
"num_tokens": 1611994.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 7.121251058578491,
|
|
"epoch": 0.06969157769869513,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.8412,
|
|
"mean_token_accuracy": 0.10536888688802719,
|
|
"num_tokens": 1623280.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 7.183412694931031,
|
|
"epoch": 0.07018584420719652,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.9013,
|
|
"mean_token_accuracy": 0.10780779123306275,
|
|
"num_tokens": 1634989.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 7.263358879089355,
|
|
"epoch": 0.07068011071569791,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.9441,
|
|
"mean_token_accuracy": 0.10218141227960587,
|
|
"num_tokens": 1647042.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 7.1384925365448,
|
|
"epoch": 0.0711743772241993,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.8193,
|
|
"mean_token_accuracy": 0.11825829073786735,
|
|
"num_tokens": 1659245.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 7.110760354995728,
|
|
"epoch": 0.07166864373270067,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.7993,
|
|
"mean_token_accuracy": 0.1118646964430809,
|
|
"num_tokens": 1670271.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 7.166925048828125,
|
|
"epoch": 0.07216291024120206,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.8701,
|
|
"mean_token_accuracy": 0.10942979305982589,
|
|
"num_tokens": 1682648.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 7.17819299697876,
|
|
"epoch": 0.07265717674970344,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.8884,
|
|
"mean_token_accuracy": 0.10658517628908157,
|
|
"num_tokens": 1694542.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 7.1897674083709715,
|
|
"epoch": 0.07315144325820483,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.7792,
|
|
"mean_token_accuracy": 0.1051691822707653,
|
|
"num_tokens": 1705737.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 7.1092109203338625,
|
|
"epoch": 0.0736457097667062,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.8445,
|
|
"mean_token_accuracy": 0.10925951823592187,
|
|
"num_tokens": 1718474.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 7.171104812622071,
|
|
"epoch": 0.07413997627520759,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.8287,
|
|
"mean_token_accuracy": 0.1083482749760151,
|
|
"num_tokens": 1729360.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 7.082499885559082,
|
|
"epoch": 0.07463424278370898,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.8145,
|
|
"mean_token_accuracy": 0.1140282928943634,
|
|
"num_tokens": 1739654.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 7.164748859405518,
|
|
"epoch": 0.07512850929221036,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.8316,
|
|
"mean_token_accuracy": 0.1102795235812664,
|
|
"num_tokens": 1752158.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 7.129274892807007,
|
|
"epoch": 0.07562277580071175,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.804,
|
|
"mean_token_accuracy": 0.11611431241035461,
|
|
"num_tokens": 1764545.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 7.115315723419189,
|
|
"epoch": 0.07611704230921312,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.8045,
|
|
"mean_token_accuracy": 0.1117110326886177,
|
|
"num_tokens": 1775422.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 7.086942100524903,
|
|
"epoch": 0.07661130881771451,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.8269,
|
|
"mean_token_accuracy": 0.10424647778272629,
|
|
"num_tokens": 1787833.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 7.1043376445770265,
|
|
"epoch": 0.0771055753262159,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.6876,
|
|
"mean_token_accuracy": 0.12525334507226943,
|
|
"num_tokens": 1798508.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 7.1196941375732425,
|
|
"epoch": 0.07759984183471728,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.7504,
|
|
"mean_token_accuracy": 0.11390295699238777,
|
|
"num_tokens": 1809873.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 7.046653699874878,
|
|
"epoch": 0.07809410834321866,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.8219,
|
|
"mean_token_accuracy": 0.10968291312456131,
|
|
"num_tokens": 1820876.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 7.074091911315918,
|
|
"epoch": 0.07858837485172004,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.7882,
|
|
"mean_token_accuracy": 0.10766826942563057,
|
|
"num_tokens": 1833233.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 7.204562473297119,
|
|
"epoch": 0.07908264136022143,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.8631,
|
|
"mean_token_accuracy": 0.1043733924627304,
|
|
"num_tokens": 1844520.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.992415571212769,
|
|
"epoch": 0.07957690786872282,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.7529,
|
|
"mean_token_accuracy": 0.11640457883477211,
|
|
"num_tokens": 1856814.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 7.083125209808349,
|
|
"epoch": 0.0800711743772242,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.7458,
|
|
"mean_token_accuracy": 0.11934995800256729,
|
|
"num_tokens": 1867581.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 7.114955091476441,
|
|
"epoch": 0.08056544088572558,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.793,
|
|
"mean_token_accuracy": 0.10545687675476074,
|
|
"num_tokens": 1879302.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 7.047843027114868,
|
|
"epoch": 0.08105970739422697,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.7703,
|
|
"mean_token_accuracy": 0.10801751762628556,
|
|
"num_tokens": 1891697.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 7.011522197723389,
|
|
"epoch": 0.08155397390272835,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.6972,
|
|
"mean_token_accuracy": 0.113658956438303,
|
|
"num_tokens": 1903544.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 7.099178838729858,
|
|
"epoch": 0.08204824041122974,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.7716,
|
|
"mean_token_accuracy": 0.11163810566067696,
|
|
"num_tokens": 1915286.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.956659364700317,
|
|
"epoch": 0.08254250691973111,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.6435,
|
|
"mean_token_accuracy": 0.1189879409968853,
|
|
"num_tokens": 1926361.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 7.089762735366821,
|
|
"epoch": 0.0830367734282325,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.7388,
|
|
"mean_token_accuracy": 0.11271768435835838,
|
|
"num_tokens": 1938540.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 7.055161714553833,
|
|
"epoch": 0.08353103993673389,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.7549,
|
|
"mean_token_accuracy": 0.11166792660951615,
|
|
"num_tokens": 1950532.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 7.115563917160034,
|
|
"epoch": 0.08402530644523527,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.7771,
|
|
"mean_token_accuracy": 0.11082297042012215,
|
|
"num_tokens": 1960899.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.951827669143677,
|
|
"epoch": 0.08451957295373666,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.7352,
|
|
"mean_token_accuracy": 0.11721526905894279,
|
|
"num_tokens": 1972278.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 7.075967121124267,
|
|
"epoch": 0.08501383946223803,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.753,
|
|
"mean_token_accuracy": 0.11036830991506577,
|
|
"num_tokens": 1985079.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.958264875411987,
|
|
"epoch": 0.08550810597073942,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.7833,
|
|
"mean_token_accuracy": 0.10577104613184929,
|
|
"num_tokens": 1996467.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 7.0288783550262455,
|
|
"epoch": 0.08600237247924081,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.6732,
|
|
"mean_token_accuracy": 0.11622288599610328,
|
|
"num_tokens": 2007339.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 7.031373453140259,
|
|
"epoch": 0.0864966389877422,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.6974,
|
|
"mean_token_accuracy": 0.11922055259346961,
|
|
"num_tokens": 2019962.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 7.015153598785401,
|
|
"epoch": 0.08699090549624358,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.6811,
|
|
"mean_token_accuracy": 0.1218427762389183,
|
|
"num_tokens": 2031251.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.9253716468811035,
|
|
"epoch": 0.08748517200474495,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.6908,
|
|
"mean_token_accuracy": 0.11059122309088706,
|
|
"num_tokens": 2042412.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.9423582553863525,
|
|
"epoch": 0.08797943851324634,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.7154,
|
|
"mean_token_accuracy": 0.11710638850927353,
|
|
"num_tokens": 2054423.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.971608352661133,
|
|
"epoch": 0.08847370502174773,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.6994,
|
|
"mean_token_accuracy": 0.11089581474661828,
|
|
"num_tokens": 2065657.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 7.033730363845825,
|
|
"epoch": 0.08896797153024912,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.6673,
|
|
"mean_token_accuracy": 0.11892515495419502,
|
|
"num_tokens": 2076305.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.9093526840209964,
|
|
"epoch": 0.08946223803875049,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.6538,
|
|
"mean_token_accuracy": 0.11571658477187156,
|
|
"num_tokens": 2087734.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.8397496223449705,
|
|
"epoch": 0.08995650454725188,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.5604,
|
|
"mean_token_accuracy": 0.12399489283561707,
|
|
"num_tokens": 2098016.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.953526353836059,
|
|
"epoch": 0.09045077105575326,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.7456,
|
|
"mean_token_accuracy": 0.11191561967134475,
|
|
"num_tokens": 2110293.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.92013955116272,
|
|
"epoch": 0.09094503756425465,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.5414,
|
|
"mean_token_accuracy": 0.12161664664745331,
|
|
"num_tokens": 2120441.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.89482569694519,
|
|
"epoch": 0.09143930407275604,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.6421,
|
|
"mean_token_accuracy": 0.12238077148795128,
|
|
"num_tokens": 2131520.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.9351709365844725,
|
|
"epoch": 0.09193357058125741,
|
|
"grad_norm": 0.79296875,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.6623,
|
|
"mean_token_accuracy": 0.11628105267882347,
|
|
"num_tokens": 2145005.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.952813577651978,
|
|
"epoch": 0.0924278370897588,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.6147,
|
|
"mean_token_accuracy": 0.12266910672187806,
|
|
"num_tokens": 2156608.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.844784832000732,
|
|
"epoch": 0.09292210359826018,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.6121,
|
|
"mean_token_accuracy": 0.12510473579168319,
|
|
"num_tokens": 2168341.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.867721843719482,
|
|
"epoch": 0.09341637010676157,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.622,
|
|
"mean_token_accuracy": 0.11190108060836793,
|
|
"num_tokens": 2178704.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 7.011946821212769,
|
|
"epoch": 0.09391063661526294,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.7787,
|
|
"mean_token_accuracy": 0.1083400510251522,
|
|
"num_tokens": 2190884.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.949655151367187,
|
|
"epoch": 0.09440490312376433,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.6793,
|
|
"mean_token_accuracy": 0.11808931306004525,
|
|
"num_tokens": 2201822.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.885033130645752,
|
|
"epoch": 0.09489916963226572,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.6691,
|
|
"mean_token_accuracy": 0.1151963710784912,
|
|
"num_tokens": 2213362.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.988611841201783,
|
|
"epoch": 0.0953934361407671,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.6834,
|
|
"mean_token_accuracy": 0.11506200209259987,
|
|
"num_tokens": 2224366.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.796273279190063,
|
|
"epoch": 0.09588770264926849,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.5308,
|
|
"mean_token_accuracy": 0.11869918927550316,
|
|
"num_tokens": 2234645.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.887414264678955,
|
|
"epoch": 0.09638196915776986,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.6156,
|
|
"mean_token_accuracy": 0.12557216882705688,
|
|
"num_tokens": 2245904.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.760451459884644,
|
|
"epoch": 0.09687623566627125,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.5608,
|
|
"mean_token_accuracy": 0.11967702507972718,
|
|
"num_tokens": 2257753.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.836030530929565,
|
|
"epoch": 0.09737050217477264,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.5764,
|
|
"mean_token_accuracy": 0.1251159615814686,
|
|
"num_tokens": 2269009.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.922193813323974,
|
|
"epoch": 0.09786476868327403,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.616,
|
|
"mean_token_accuracy": 0.11580619663000107,
|
|
"num_tokens": 2279840.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.835604095458985,
|
|
"epoch": 0.0983590351917754,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.5989,
|
|
"mean_token_accuracy": 0.11514367312192916,
|
|
"num_tokens": 2290577.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.809481859207153,
|
|
"epoch": 0.09885330170027679,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.5868,
|
|
"mean_token_accuracy": 0.12064301818609238,
|
|
"num_tokens": 2301341.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.813262176513672,
|
|
"epoch": 0.09934756820877817,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004999999982287888,
|
|
"loss": 6.6368,
|
|
"mean_token_accuracy": 0.11775564104318618,
|
|
"num_tokens": 2312505.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.9637086391448975,
|
|
"epoch": 0.09984183471727956,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004999999910332434,
|
|
"loss": 6.5746,
|
|
"mean_token_accuracy": 0.11561667025089264,
|
|
"num_tokens": 2323443.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.791029787063598,
|
|
"epoch": 0.10033610122578095,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999999783026635,
|
|
"loss": 6.5993,
|
|
"mean_token_accuracy": 0.11769351288676262,
|
|
"num_tokens": 2335686.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.783609867095947,
|
|
"epoch": 0.10083036773428232,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999999600370491,
|
|
"loss": 6.5922,
|
|
"mean_token_accuracy": 0.11608155444264412,
|
|
"num_tokens": 2347273.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.764084768295288,
|
|
"epoch": 0.1013246342427837,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004999999362364008,
|
|
"loss": 6.5001,
|
|
"mean_token_accuracy": 0.13010589852929116,
|
|
"num_tokens": 2358011.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.786945724487305,
|
|
"epoch": 0.1018189007512851,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004999999069007191,
|
|
"loss": 6.6375,
|
|
"mean_token_accuracy": 0.1150654211640358,
|
|
"num_tokens": 2371390.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.861456680297851,
|
|
"epoch": 0.10231316725978648,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999998720300048,
|
|
"loss": 6.5951,
|
|
"mean_token_accuracy": 0.12022169157862664,
|
|
"num_tokens": 2381887.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.88414945602417,
|
|
"epoch": 0.10280743376828787,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004999998316242589,
|
|
"loss": 6.535,
|
|
"mean_token_accuracy": 0.12180361151695251,
|
|
"num_tokens": 2393632.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.735341882705688,
|
|
"epoch": 0.10330170027678924,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999997856834821,
|
|
"loss": 6.5544,
|
|
"mean_token_accuracy": 0.12011176198720933,
|
|
"num_tokens": 2405686.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.720200252532959,
|
|
"epoch": 0.10379596678529063,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004999997342076757,
|
|
"loss": 6.5159,
|
|
"mean_token_accuracy": 0.11903640404343604,
|
|
"num_tokens": 2417459.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.890789365768432,
|
|
"epoch": 0.10429023329379201,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004999996771968408,
|
|
"loss": 6.5692,
|
|
"mean_token_accuracy": 0.11979348957538605,
|
|
"num_tokens": 2428689.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.733220434188842,
|
|
"epoch": 0.1047844998022934,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999996146509791,
|
|
"loss": 6.5962,
|
|
"mean_token_accuracy": 0.11043051034212112,
|
|
"num_tokens": 2439437.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.875022935867309,
|
|
"epoch": 0.10527876631079477,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004999995465700919,
|
|
"loss": 6.5727,
|
|
"mean_token_accuracy": 0.11816314235329628,
|
|
"num_tokens": 2450576.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.892560863494873,
|
|
"epoch": 0.10577303281929616,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999994729541809,
|
|
"loss": 6.6961,
|
|
"mean_token_accuracy": 0.11508011296391488,
|
|
"num_tokens": 2462684.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.847589826583862,
|
|
"epoch": 0.10626729932779755,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004999993938032481,
|
|
"loss": 6.5992,
|
|
"mean_token_accuracy": 0.12335227355360985,
|
|
"num_tokens": 2474193.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.71039776802063,
|
|
"epoch": 0.10676156583629894,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004999993091172952,
|
|
"loss": 6.5155,
|
|
"mean_token_accuracy": 0.118268733471632,
|
|
"num_tokens": 2486264.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.779750728607178,
|
|
"epoch": 0.10725583234480032,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004999992188963245,
|
|
"loss": 6.5201,
|
|
"mean_token_accuracy": 0.12872469574213027,
|
|
"num_tokens": 2497668.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.727497005462647,
|
|
"epoch": 0.1077500988533017,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.000499999123140338,
|
|
"loss": 6.5714,
|
|
"mean_token_accuracy": 0.11793504282832146,
|
|
"num_tokens": 2509635.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.829061508178711,
|
|
"epoch": 0.10824436536180308,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999990218493382,
|
|
"loss": 6.5694,
|
|
"mean_token_accuracy": 0.11895573884248734,
|
|
"num_tokens": 2521980.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.751840114593506,
|
|
"epoch": 0.10873863187030447,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004999989150233276,
|
|
"loss": 6.525,
|
|
"mean_token_accuracy": 0.12123857289552689,
|
|
"num_tokens": 2533750.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.78528938293457,
|
|
"epoch": 0.10923289837880586,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004999988026623088,
|
|
"loss": 6.622,
|
|
"mean_token_accuracy": 0.11473045870661736,
|
|
"num_tokens": 2545620.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.696118307113648,
|
|
"epoch": 0.10972716488730723,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999986847662843,
|
|
"loss": 6.3993,
|
|
"mean_token_accuracy": 0.1327408328652382,
|
|
"num_tokens": 2556463.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.784327554702759,
|
|
"epoch": 0.11022143139580862,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999985613352576,
|
|
"loss": 6.5923,
|
|
"mean_token_accuracy": 0.11819555461406708,
|
|
"num_tokens": 2567396.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.7881745338439945,
|
|
"epoch": 0.11071569790431,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004999984323692311,
|
|
"loss": 6.5291,
|
|
"mean_token_accuracy": 0.12449823319911957,
|
|
"num_tokens": 2579208.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.713969278335571,
|
|
"epoch": 0.11120996441281139,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004999982978682085,
|
|
"loss": 6.4819,
|
|
"mean_token_accuracy": 0.12787408530712127,
|
|
"num_tokens": 2590940.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.781643867492676,
|
|
"epoch": 0.11170423092131278,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004999981578321927,
|
|
"loss": 6.5599,
|
|
"mean_token_accuracy": 0.12885867655277253,
|
|
"num_tokens": 2602491.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.6775916576385494,
|
|
"epoch": 0.11219849742981415,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004999980122611874,
|
|
"loss": 6.3952,
|
|
"mean_token_accuracy": 0.13354044929146766,
|
|
"num_tokens": 2612139.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.752026987075806,
|
|
"epoch": 0.11269276393831554,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.000499997861155196,
|
|
"loss": 6.5509,
|
|
"mean_token_accuracy": 0.12556230872869492,
|
|
"num_tokens": 2624906.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.673003530502319,
|
|
"epoch": 0.11318703044681692,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999977045142223,
|
|
"loss": 6.4808,
|
|
"mean_token_accuracy": 0.1187933087348938,
|
|
"num_tokens": 2636046.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.753292274475098,
|
|
"epoch": 0.11368129695531831,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999975423382703,
|
|
"loss": 6.4753,
|
|
"mean_token_accuracy": 0.12741708233952523,
|
|
"num_tokens": 2645617.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.790705585479737,
|
|
"epoch": 0.11417556346381968,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999973746273438,
|
|
"loss": 6.5141,
|
|
"mean_token_accuracy": 0.122653928399086,
|
|
"num_tokens": 2656780.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.733772039413452,
|
|
"epoch": 0.11466982997232107,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000499997201381447,
|
|
"loss": 6.5012,
|
|
"mean_token_accuracy": 0.12425200417637824,
|
|
"num_tokens": 2668078.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.741680097579956,
|
|
"epoch": 0.11516409648082246,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999970226005841,
|
|
"loss": 6.5639,
|
|
"mean_token_accuracy": 0.11674718782305718,
|
|
"num_tokens": 2679049.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.803271245956421,
|
|
"epoch": 0.11565836298932385,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004999968382847595,
|
|
"loss": 6.5366,
|
|
"mean_token_accuracy": 0.12377994433045388,
|
|
"num_tokens": 2690165.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.783146047592163,
|
|
"epoch": 0.11615262949782523,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999966484339778,
|
|
"loss": 6.5556,
|
|
"mean_token_accuracy": 0.12106125354766846,
|
|
"num_tokens": 2701965.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.6661967754364015,
|
|
"epoch": 0.1166468960063266,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999964530482437,
|
|
"loss": 6.4637,
|
|
"mean_token_accuracy": 0.12662513479590415,
|
|
"num_tokens": 2713479.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.716840696334839,
|
|
"epoch": 0.11714116251482799,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999962521275619,
|
|
"loss": 6.4722,
|
|
"mean_token_accuracy": 0.1229412630200386,
|
|
"num_tokens": 2725629.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.520010709762573,
|
|
"epoch": 0.11763542902332938,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999960456719374,
|
|
"loss": 6.3949,
|
|
"mean_token_accuracy": 0.12263003587722779,
|
|
"num_tokens": 2736620.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.698487615585327,
|
|
"epoch": 0.11812969553183077,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004999958336813753,
|
|
"loss": 6.3743,
|
|
"mean_token_accuracy": 0.13159746527671815,
|
|
"num_tokens": 2747118.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.5850576877594,
|
|
"epoch": 0.11862396204033215,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999956161558807,
|
|
"loss": 6.4496,
|
|
"mean_token_accuracy": 0.12530086562037468,
|
|
"num_tokens": 2758371.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.637788963317871,
|
|
"epoch": 0.11911822854883353,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004999953930954591,
|
|
"loss": 6.4311,
|
|
"mean_token_accuracy": 0.12403330877423287,
|
|
"num_tokens": 2770575.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.686907768249512,
|
|
"epoch": 0.11961249505733491,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004999951645001159,
|
|
"loss": 6.3959,
|
|
"mean_token_accuracy": 0.12846756651997565,
|
|
"num_tokens": 2783103.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.6527008533477785,
|
|
"epoch": 0.1201067615658363,
|
|
"grad_norm": 0.8203125,
|
|
"learning_rate": 0.0004999949303698568,
|
|
"loss": 6.516,
|
|
"mean_token_accuracy": 0.12293659523129463,
|
|
"num_tokens": 2796636.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.695577192306518,
|
|
"epoch": 0.12060102807433769,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004999946907046875,
|
|
"loss": 6.3694,
|
|
"mean_token_accuracy": 0.13485616445541382,
|
|
"num_tokens": 2807798.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.6003742694854735,
|
|
"epoch": 0.12109529458283906,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004999944455046139,
|
|
"loss": 6.487,
|
|
"mean_token_accuracy": 0.12794808968901633,
|
|
"num_tokens": 2819255.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.676394367218018,
|
|
"epoch": 0.12158956109134045,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000499994194769642,
|
|
"loss": 6.4363,
|
|
"mean_token_accuracy": 0.12319996431469918,
|
|
"num_tokens": 2829952.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.776423740386963,
|
|
"epoch": 0.12208382759984183,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000499993938499778,
|
|
"loss": 6.5218,
|
|
"mean_token_accuracy": 0.1313498340547085,
|
|
"num_tokens": 2841056.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.614255428314209,
|
|
"epoch": 0.12257809410834322,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999936766950283,
|
|
"loss": 6.3736,
|
|
"mean_token_accuracy": 0.13403896167874335,
|
|
"num_tokens": 2852045.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.6999880313873295,
|
|
"epoch": 0.12307236061684461,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999934093553992,
|
|
"loss": 6.4759,
|
|
"mean_token_accuracy": 0.1261597417294979,
|
|
"num_tokens": 2863440.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.664808034896851,
|
|
"epoch": 0.12356662712534598,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004999931364808972,
|
|
"loss": 6.4951,
|
|
"mean_token_accuracy": 0.12566187903285025,
|
|
"num_tokens": 2874017.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.6710155487060545,
|
|
"epoch": 0.12406089363384737,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999928580715293,
|
|
"loss": 6.4019,
|
|
"mean_token_accuracy": 0.1276327282190323,
|
|
"num_tokens": 2885677.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.632281970977783,
|
|
"epoch": 0.12455516014234876,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.000499992574127302,
|
|
"loss": 6.5095,
|
|
"mean_token_accuracy": 0.12587906420230865,
|
|
"num_tokens": 2897825.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.72322564125061,
|
|
"epoch": 0.12504942665085014,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004999922846482226,
|
|
"loss": 6.558,
|
|
"mean_token_accuracy": 0.11891639679670334,
|
|
"num_tokens": 2909304.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.680528593063355,
|
|
"epoch": 0.12554369315935152,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004999919896342981,
|
|
"loss": 6.494,
|
|
"mean_token_accuracy": 0.11919789239764214,
|
|
"num_tokens": 2920918.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.711538314819336,
|
|
"epoch": 0.12603795966785292,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004999916890855358,
|
|
"loss": 6.4429,
|
|
"mean_token_accuracy": 0.12809634804725648,
|
|
"num_tokens": 2932294.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.614521741867065,
|
|
"epoch": 0.1265322261763543,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004999913830019431,
|
|
"loss": 6.3589,
|
|
"mean_token_accuracy": 0.13480542078614235,
|
|
"num_tokens": 2943372.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.5310547828674315,
|
|
"epoch": 0.12702649268485566,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999910713835272,
|
|
"loss": 6.3364,
|
|
"mean_token_accuracy": 0.12778233140707015,
|
|
"num_tokens": 2953807.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.607894229888916,
|
|
"epoch": 0.12752075919335706,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999907542302962,
|
|
"loss": 6.2839,
|
|
"mean_token_accuracy": 0.13762608990073205,
|
|
"num_tokens": 2964372.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.600684118270874,
|
|
"epoch": 0.12801502570185844,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999904315422577,
|
|
"loss": 6.3847,
|
|
"mean_token_accuracy": 0.13021055534482,
|
|
"num_tokens": 2975537.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.563079786300659,
|
|
"epoch": 0.12850929221035984,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999901033194198,
|
|
"loss": 6.3563,
|
|
"mean_token_accuracy": 0.12701485753059388,
|
|
"num_tokens": 2986094.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.685067892074585,
|
|
"epoch": 0.1290035587188612,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999897695617904,
|
|
"loss": 6.4705,
|
|
"mean_token_accuracy": 0.12639627903699874,
|
|
"num_tokens": 2996023.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.640681552886963,
|
|
"epoch": 0.12949782522736258,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004999894302693778,
|
|
"loss": 6.4669,
|
|
"mean_token_accuracy": 0.12245377749204636,
|
|
"num_tokens": 3007584.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.764276647567749,
|
|
"epoch": 0.12999209173586398,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004999890854421902,
|
|
"loss": 6.4915,
|
|
"mean_token_accuracy": 0.12133662179112434,
|
|
"num_tokens": 3018995.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.470447969436646,
|
|
"epoch": 0.13048635824436536,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004999887350802362,
|
|
"loss": 6.3006,
|
|
"mean_token_accuracy": 0.13123086243867874,
|
|
"num_tokens": 3030576.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.557611608505249,
|
|
"epoch": 0.13098062475286676,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999883791835246,
|
|
"loss": 6.3631,
|
|
"mean_token_accuracy": 0.12982181310653687,
|
|
"num_tokens": 3040765.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.625185441970825,
|
|
"epoch": 0.13147489126136813,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004999880177520637,
|
|
"loss": 6.4162,
|
|
"mean_token_accuracy": 0.12189083769917489,
|
|
"num_tokens": 3053912.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.515785074234008,
|
|
"epoch": 0.1319691577698695,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004999876507858628,
|
|
"loss": 6.3911,
|
|
"mean_token_accuracy": 0.12720990628004075,
|
|
"num_tokens": 3065921.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.7070118427276615,
|
|
"epoch": 0.1324634242783709,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004999872782849308,
|
|
"loss": 6.4407,
|
|
"mean_token_accuracy": 0.1257806733250618,
|
|
"num_tokens": 3078643.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.503523445129394,
|
|
"epoch": 0.13295769078687228,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004999869002492768,
|
|
"loss": 6.3656,
|
|
"mean_token_accuracy": 0.12997100427746772,
|
|
"num_tokens": 3090300.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.547778797149658,
|
|
"epoch": 0.13345195729537365,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004999865166789102,
|
|
"loss": 6.2962,
|
|
"mean_token_accuracy": 0.13570954278111458,
|
|
"num_tokens": 3102307.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.671496915817261,
|
|
"epoch": 0.13394622380387505,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004999861275738404,
|
|
"loss": 6.438,
|
|
"mean_token_accuracy": 0.122230763733387,
|
|
"num_tokens": 3113817.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.534291315078735,
|
|
"epoch": 0.13444049031237643,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999857329340769,
|
|
"loss": 6.2984,
|
|
"mean_token_accuracy": 0.1316831812262535,
|
|
"num_tokens": 3124719.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.49489164352417,
|
|
"epoch": 0.13493475682087783,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004999853327596295,
|
|
"loss": 6.3292,
|
|
"mean_token_accuracy": 0.12992627099156379,
|
|
"num_tokens": 3136136.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.607687616348267,
|
|
"epoch": 0.1354290233293792,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.000499984927050508,
|
|
"loss": 6.3386,
|
|
"mean_token_accuracy": 0.13101759776473046,
|
|
"num_tokens": 3148977.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.594228982925415,
|
|
"epoch": 0.13592328983788057,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999845158067223,
|
|
"loss": 6.3974,
|
|
"mean_token_accuracy": 0.12740953639149666,
|
|
"num_tokens": 3160520.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.579832983016968,
|
|
"epoch": 0.13641755634638197,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004999840990282828,
|
|
"loss": 6.3672,
|
|
"mean_token_accuracy": 0.1273616522550583,
|
|
"num_tokens": 3172895.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.562347316741944,
|
|
"epoch": 0.13691182285488335,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004999836767151994,
|
|
"loss": 6.3811,
|
|
"mean_token_accuracy": 0.1285211369395256,
|
|
"num_tokens": 3183960.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.513716793060302,
|
|
"epoch": 0.13740608936338475,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999832488674827,
|
|
"loss": 6.4157,
|
|
"mean_token_accuracy": 0.12590486258268357,
|
|
"num_tokens": 3194276.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.599906539916992,
|
|
"epoch": 0.13790035587188612,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004999828154851432,
|
|
"loss": 6.2909,
|
|
"mean_token_accuracy": 0.1359431803226471,
|
|
"num_tokens": 3206084.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.534882593154907,
|
|
"epoch": 0.1383946223803875,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999823765681915,
|
|
"loss": 6.3305,
|
|
"mean_token_accuracy": 0.1354965254664421,
|
|
"num_tokens": 3217941.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.533507204055786,
|
|
"epoch": 0.1388888888888889,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004999819321166385,
|
|
"loss": 6.3004,
|
|
"mean_token_accuracy": 0.13300728499889375,
|
|
"num_tokens": 3229088.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.476707124710083,
|
|
"epoch": 0.13938315539739027,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000499981482130495,
|
|
"loss": 6.2574,
|
|
"mean_token_accuracy": 0.13597408458590507,
|
|
"num_tokens": 3240708.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.52806544303894,
|
|
"epoch": 0.13987742190589167,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004999810266097721,
|
|
"loss": 6.3747,
|
|
"mean_token_accuracy": 0.12898992374539375,
|
|
"num_tokens": 3251040.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.505051183700561,
|
|
"epoch": 0.14037168841439304,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999805655544811,
|
|
"loss": 6.3206,
|
|
"mean_token_accuracy": 0.13407422080636025,
|
|
"num_tokens": 3262655.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.514699459075928,
|
|
"epoch": 0.14086595492289442,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999800989646333,
|
|
"loss": 6.2762,
|
|
"mean_token_accuracy": 0.1369866333901882,
|
|
"num_tokens": 3273160.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.538471126556397,
|
|
"epoch": 0.14136022143139582,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999796268402401,
|
|
"loss": 6.3008,
|
|
"mean_token_accuracy": 0.129919171333313,
|
|
"num_tokens": 3285932.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.4485719203948975,
|
|
"epoch": 0.1418544879398972,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004999791491813132,
|
|
"loss": 6.3325,
|
|
"mean_token_accuracy": 0.12975751757621765,
|
|
"num_tokens": 3298834.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.628458499908447,
|
|
"epoch": 0.1423487544483986,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999786659878643,
|
|
"loss": 6.3121,
|
|
"mean_token_accuracy": 0.13101852536201478,
|
|
"num_tokens": 3310473.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.355752944946289,
|
|
"epoch": 0.14284302095689996,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999781772599055,
|
|
"loss": 6.2774,
|
|
"mean_token_accuracy": 0.1346455655992031,
|
|
"num_tokens": 3322662.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.666700553894043,
|
|
"epoch": 0.14333728746540134,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004999776829974484,
|
|
"loss": 6.377,
|
|
"mean_token_accuracy": 0.13617630526423455,
|
|
"num_tokens": 3334321.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.4790514469146725,
|
|
"epoch": 0.14383155397390274,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999771832005055,
|
|
"loss": 6.2639,
|
|
"mean_token_accuracy": 0.12925363183021546,
|
|
"num_tokens": 3345130.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.453184318542481,
|
|
"epoch": 0.1443258204824041,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000499976677869089,
|
|
"loss": 6.2703,
|
|
"mean_token_accuracy": 0.13220407143235208,
|
|
"num_tokens": 3355601.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.477377080917359,
|
|
"epoch": 0.14482008699090548,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004999761670032115,
|
|
"loss": 6.2541,
|
|
"mean_token_accuracy": 0.13375383466482163,
|
|
"num_tokens": 3367075.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.511325311660767,
|
|
"epoch": 0.14531435349940688,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004999756506028851,
|
|
"loss": 6.2408,
|
|
"mean_token_accuracy": 0.13651637956500054,
|
|
"num_tokens": 3378888.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.501529884338379,
|
|
"epoch": 0.14580862000790826,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004999751286681229,
|
|
"loss": 6.2783,
|
|
"mean_token_accuracy": 0.13225537091493605,
|
|
"num_tokens": 3390705.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.47571210861206,
|
|
"epoch": 0.14630288651640966,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004999746011989376,
|
|
"loss": 6.2827,
|
|
"mean_token_accuracy": 0.1402031034231186,
|
|
"num_tokens": 3401710.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.556623744964599,
|
|
"epoch": 0.14679715302491103,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999740681953422,
|
|
"loss": 6.3976,
|
|
"mean_token_accuracy": 0.1264283038675785,
|
|
"num_tokens": 3413467.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.510135364532471,
|
|
"epoch": 0.1472914195334124,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004999735296573499,
|
|
"loss": 6.3265,
|
|
"mean_token_accuracy": 0.1363067075610161,
|
|
"num_tokens": 3425885.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.520489978790283,
|
|
"epoch": 0.1477856860419138,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004999729855849738,
|
|
"loss": 6.3524,
|
|
"mean_token_accuracy": 0.1277867116034031,
|
|
"num_tokens": 3437952.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.465743494033814,
|
|
"epoch": 0.14827995255041518,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004999724359782274,
|
|
"loss": 6.2184,
|
|
"mean_token_accuracy": 0.14817326366901398,
|
|
"num_tokens": 3449680.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 6.495955276489258,
|
|
"epoch": 0.14877421905891658,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999718808371241,
|
|
"loss": 6.3814,
|
|
"mean_token_accuracy": 0.1254202604293823,
|
|
"num_tokens": 3460699.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 6.596548366546631,
|
|
"epoch": 0.14926848556741795,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004999713201616776,
|
|
"loss": 6.2471,
|
|
"mean_token_accuracy": 0.13369383811950683,
|
|
"num_tokens": 3471919.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 6.353741025924682,
|
|
"epoch": 0.14976275207591933,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000499970753951902,
|
|
"loss": 6.356,
|
|
"mean_token_accuracy": 0.12923398613929749,
|
|
"num_tokens": 3484110.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 6.576723194122314,
|
|
"epoch": 0.15025701858442073,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999701822078107,
|
|
"loss": 6.265,
|
|
"mean_token_accuracy": 0.1321140117943287,
|
|
"num_tokens": 3496103.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 6.493707942962646,
|
|
"epoch": 0.1507512850929221,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000499969604929418,
|
|
"loss": 6.3409,
|
|
"mean_token_accuracy": 0.13155030012130736,
|
|
"num_tokens": 3507291.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 6.532199859619141,
|
|
"epoch": 0.1512455516014235,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004999690221167382,
|
|
"loss": 6.3266,
|
|
"mean_token_accuracy": 0.1285949692130089,
|
|
"num_tokens": 3519564.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 6.396489763259888,
|
|
"epoch": 0.15173981810992487,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004999684337697855,
|
|
"loss": 6.1979,
|
|
"mean_token_accuracy": 0.14119874984025954,
|
|
"num_tokens": 3530679.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 6.489102125167847,
|
|
"epoch": 0.15223408461842625,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004999678398885746,
|
|
"loss": 6.2593,
|
|
"mean_token_accuracy": 0.13449762463569642,
|
|
"num_tokens": 3542662.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 6.484030818939209,
|
|
"epoch": 0.15272835112692765,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004999672404731197,
|
|
"loss": 6.3726,
|
|
"mean_token_accuracy": 0.12962616756558418,
|
|
"num_tokens": 3554760.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 6.430119800567627,
|
|
"epoch": 0.15322261763542902,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000499966635523436,
|
|
"loss": 6.184,
|
|
"mean_token_accuracy": 0.1408859372138977,
|
|
"num_tokens": 3565604.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 6.50899305343628,
|
|
"epoch": 0.1537168841439304,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999660250395381,
|
|
"loss": 6.3469,
|
|
"mean_token_accuracy": 0.12966826558113098,
|
|
"num_tokens": 3577535.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 6.513133192062378,
|
|
"epoch": 0.1542111506524318,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000499965409021441,
|
|
"loss": 6.3448,
|
|
"mean_token_accuracy": 0.12810506224632262,
|
|
"num_tokens": 3590458.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 6.503912401199341,
|
|
"epoch": 0.15470541716093317,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00049996478746916,
|
|
"loss": 6.3018,
|
|
"mean_token_accuracy": 0.13346393704414367,
|
|
"num_tokens": 3600916.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 6.413891839981079,
|
|
"epoch": 0.15519968366943457,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004999641603827104,
|
|
"loss": 6.3234,
|
|
"mean_token_accuracy": 0.13152189254760743,
|
|
"num_tokens": 3612794.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 6.491737747192383,
|
|
"epoch": 0.15569395017793594,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999635277621073,
|
|
"loss": 6.2369,
|
|
"mean_token_accuracy": 0.1417726531624794,
|
|
"num_tokens": 3624186.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 6.421962738037109,
|
|
"epoch": 0.15618821668643731,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004999628896073668,
|
|
"loss": 6.2647,
|
|
"mean_token_accuracy": 0.13585594519972802,
|
|
"num_tokens": 3636041.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 6.4938102722167965,
|
|
"epoch": 0.15668248319493872,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004999622459185042,
|
|
"loss": 6.3248,
|
|
"mean_token_accuracy": 0.1295533798635006,
|
|
"num_tokens": 3646859.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 6.432663345336914,
|
|
"epoch": 0.1571767497034401,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004999615966955354,
|
|
"loss": 6.1457,
|
|
"mean_token_accuracy": 0.14173236414790152,
|
|
"num_tokens": 3657853.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 6.514078617095947,
|
|
"epoch": 0.1576710162119415,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004999609419384764,
|
|
"loss": 6.2523,
|
|
"mean_token_accuracy": 0.1346098266541958,
|
|
"num_tokens": 3668341.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 6.446404361724854,
|
|
"epoch": 0.15816528272044286,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004999602816473434,
|
|
"loss": 6.2758,
|
|
"mean_token_accuracy": 0.13876884281635285,
|
|
"num_tokens": 3679020.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 6.419503450393677,
|
|
"epoch": 0.15865954922894424,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004999596158221526,
|
|
"loss": 6.1899,
|
|
"mean_token_accuracy": 0.14101243168115615,
|
|
"num_tokens": 3691121.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 6.448402786254883,
|
|
"epoch": 0.15915381573744564,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004999589444629203,
|
|
"loss": 6.2489,
|
|
"mean_token_accuracy": 0.13868038430809976,
|
|
"num_tokens": 3702584.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 6.412480592727661,
|
|
"epoch": 0.159648082245947,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999582675696631,
|
|
"loss": 6.1982,
|
|
"mean_token_accuracy": 0.14014509692788124,
|
|
"num_tokens": 3713085.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 6.4948835372924805,
|
|
"epoch": 0.1601423487544484,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999575851423976,
|
|
"loss": 6.322,
|
|
"mean_token_accuracy": 0.1277170367538929,
|
|
"num_tokens": 3724318.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 6.488896036148072,
|
|
"epoch": 0.16063661526294978,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999568971811407,
|
|
"loss": 6.2169,
|
|
"mean_token_accuracy": 0.14014409109950066,
|
|
"num_tokens": 3734288.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 6.383061933517456,
|
|
"epoch": 0.16113088177145116,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000499956203685909,
|
|
"loss": 6.2253,
|
|
"mean_token_accuracy": 0.1334229089319706,
|
|
"num_tokens": 3744645.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 6.456633996963501,
|
|
"epoch": 0.16162514827995256,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00049995550465672,
|
|
"loss": 6.219,
|
|
"mean_token_accuracy": 0.1377868391573429,
|
|
"num_tokens": 3755660.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 6.372065544128418,
|
|
"epoch": 0.16211941478845393,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999548000935906,
|
|
"loss": 6.1684,
|
|
"mean_token_accuracy": 0.1421402059495449,
|
|
"num_tokens": 3767614.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 6.441373538970947,
|
|
"epoch": 0.16261368129695533,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004999540899965383,
|
|
"loss": 6.3184,
|
|
"mean_token_accuracy": 0.1364623002707958,
|
|
"num_tokens": 3780750.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 6.428185367584229,
|
|
"epoch": 0.1631079478054567,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004999533743655803,
|
|
"loss": 6.3161,
|
|
"mean_token_accuracy": 0.1284065067768097,
|
|
"num_tokens": 3793195.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 6.457364511489868,
|
|
"epoch": 0.16360221431395808,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004999526532007346,
|
|
"loss": 6.2676,
|
|
"mean_token_accuracy": 0.13861094564199447,
|
|
"num_tokens": 3804256.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 6.42730746269226,
|
|
"epoch": 0.16409648082245948,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999519265020187,
|
|
"loss": 6.2388,
|
|
"mean_token_accuracy": 0.13844978511333467,
|
|
"num_tokens": 3815403.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 6.320225238800049,
|
|
"epoch": 0.16459074733096085,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004999511942694504,
|
|
"loss": 6.1467,
|
|
"mean_token_accuracy": 0.14466449469327927,
|
|
"num_tokens": 3826682.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 6.43041729927063,
|
|
"epoch": 0.16508501383946222,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004999504565030479,
|
|
"loss": 6.1678,
|
|
"mean_token_accuracy": 0.13993233293294907,
|
|
"num_tokens": 3837780.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 6.429787445068359,
|
|
"epoch": 0.16557928034796363,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004999497132028293,
|
|
"loss": 6.2239,
|
|
"mean_token_accuracy": 0.1318391665816307,
|
|
"num_tokens": 3848782.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 6.428545904159546,
|
|
"epoch": 0.166073546856465,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004999489643688127,
|
|
"loss": 6.1696,
|
|
"mean_token_accuracy": 0.13959652185440063,
|
|
"num_tokens": 3859220.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 6.30494270324707,
|
|
"epoch": 0.1665678133649664,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999482100010169,
|
|
"loss": 6.1018,
|
|
"mean_token_accuracy": 0.1421866834163666,
|
|
"num_tokens": 3869794.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 6.35781421661377,
|
|
"epoch": 0.16706207987346777,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004999474500994601,
|
|
"loss": 6.2396,
|
|
"mean_token_accuracy": 0.13421173691749572,
|
|
"num_tokens": 3881133.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 6.413172864913941,
|
|
"epoch": 0.16755634638196915,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004999466846641611,
|
|
"loss": 6.1624,
|
|
"mean_token_accuracy": 0.13713917806744574,
|
|
"num_tokens": 3891600.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 6.333802795410156,
|
|
"epoch": 0.16805061289047055,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004999459136951389,
|
|
"loss": 6.2631,
|
|
"mean_token_accuracy": 0.13335662260651587,
|
|
"num_tokens": 3903287.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 6.439137983322143,
|
|
"epoch": 0.16854487939897192,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999451371924123,
|
|
"loss": 6.1806,
|
|
"mean_token_accuracy": 0.13854067176580429,
|
|
"num_tokens": 3914497.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 6.338579225540161,
|
|
"epoch": 0.16903914590747332,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004999443551560003,
|
|
"loss": 6.2093,
|
|
"mean_token_accuracy": 0.13935932666063308,
|
|
"num_tokens": 3925536.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 6.43532075881958,
|
|
"epoch": 0.1695334124159747,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004999435675859224,
|
|
"loss": 6.1608,
|
|
"mean_token_accuracy": 0.14295403137803078,
|
|
"num_tokens": 3936705.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 6.352819776535034,
|
|
"epoch": 0.17002767892447607,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999427744821977,
|
|
"loss": 6.0594,
|
|
"mean_token_accuracy": 0.1423388808965683,
|
|
"num_tokens": 3948422.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 6.321838235855102,
|
|
"epoch": 0.17052194543297747,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000499941975844846,
|
|
"loss": 6.1211,
|
|
"mean_token_accuracy": 0.14932032078504562,
|
|
"num_tokens": 3958697.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 6.3387940406799315,
|
|
"epoch": 0.17101621194147884,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004999411716738866,
|
|
"loss": 6.1359,
|
|
"mean_token_accuracy": 0.13918534070253372,
|
|
"num_tokens": 3970255.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 6.376590156555176,
|
|
"epoch": 0.17151047844998024,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004999403619693397,
|
|
"loss": 6.2221,
|
|
"mean_token_accuracy": 0.1375705450773239,
|
|
"num_tokens": 3980892.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 6.290405988693237,
|
|
"epoch": 0.17200474495848161,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999395467312249,
|
|
"loss": 6.159,
|
|
"mean_token_accuracy": 0.13679868206381798,
|
|
"num_tokens": 3991375.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 6.445168209075928,
|
|
"epoch": 0.172499011466983,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004999387259595623,
|
|
"loss": 6.2982,
|
|
"mean_token_accuracy": 0.13123658969998359,
|
|
"num_tokens": 4003352.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 6.349807596206665,
|
|
"epoch": 0.1729932779754844,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004999378996543723,
|
|
"loss": 6.1895,
|
|
"mean_token_accuracy": 0.13942418098449708,
|
|
"num_tokens": 4014429.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 6.37531623840332,
|
|
"epoch": 0.17348754448398576,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999370678156749,
|
|
"loss": 6.1121,
|
|
"mean_token_accuracy": 0.1444903664290905,
|
|
"num_tokens": 4025213.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 6.307505178451538,
|
|
"epoch": 0.17398181099248716,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999362304434909,
|
|
"loss": 6.1071,
|
|
"mean_token_accuracy": 0.1425596944987774,
|
|
"num_tokens": 4036244.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 6.413739204406738,
|
|
"epoch": 0.17447607750098854,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999353875378406,
|
|
"loss": 6.1546,
|
|
"mean_token_accuracy": 0.1389375738799572,
|
|
"num_tokens": 4047944.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 6.28089075088501,
|
|
"epoch": 0.1749703440094899,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.000499934539098745,
|
|
"loss": 6.1216,
|
|
"mean_token_accuracy": 0.14509597718715667,
|
|
"num_tokens": 4058865.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 6.326784420013428,
|
|
"epoch": 0.1754646105179913,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004999336851262247,
|
|
"loss": 6.1735,
|
|
"mean_token_accuracy": 0.13881123885512353,
|
|
"num_tokens": 4070627.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 6.431045770645142,
|
|
"epoch": 0.17595887702649268,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004999328256203009,
|
|
"loss": 6.221,
|
|
"mean_token_accuracy": 0.13756831139326095,
|
|
"num_tokens": 4081406.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 6.324437522888184,
|
|
"epoch": 0.17645314353499406,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004999319605809947,
|
|
"loss": 6.1871,
|
|
"mean_token_accuracy": 0.12888796105980874,
|
|
"num_tokens": 4092957.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 6.411680078506469,
|
|
"epoch": 0.17694741004349546,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004999310900083274,
|
|
"loss": 6.231,
|
|
"mean_token_accuracy": 0.1384389355778694,
|
|
"num_tokens": 4104381.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 6.3523437023162845,
|
|
"epoch": 0.17744167655199683,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004999302139023204,
|
|
"loss": 6.1525,
|
|
"mean_token_accuracy": 0.14232124909758567,
|
|
"num_tokens": 4117190.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 6.305995988845825,
|
|
"epoch": 0.17793594306049823,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004999293322629951,
|
|
"loss": 6.0964,
|
|
"mean_token_accuracy": 0.14318788796663284,
|
|
"num_tokens": 4127870.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 6.298418235778809,
|
|
"epoch": 0.1784302095689996,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999284450903736,
|
|
"loss": 6.1353,
|
|
"mean_token_accuracy": 0.14193641096353532,
|
|
"num_tokens": 4139240.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 6.373674917221069,
|
|
"epoch": 0.17892447607750098,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999275523844772,
|
|
"loss": 6.1209,
|
|
"mean_token_accuracy": 0.14574467837810517,
|
|
"num_tokens": 4149885.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 6.44624924659729,
|
|
"epoch": 0.17941874258600238,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999266541453283,
|
|
"loss": 6.2778,
|
|
"mean_token_accuracy": 0.13498160094022751,
|
|
"num_tokens": 4161824.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 6.335614776611328,
|
|
"epoch": 0.17991300909450375,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004999257503729487,
|
|
"loss": 6.2461,
|
|
"mean_token_accuracy": 0.13315987810492516,
|
|
"num_tokens": 4173581.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 6.3903834342956545,
|
|
"epoch": 0.18040727560300515,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999248410673607,
|
|
"loss": 6.1128,
|
|
"mean_token_accuracy": 0.14748075008392333,
|
|
"num_tokens": 4184843.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 6.380693054199218,
|
|
"epoch": 0.18090154211150652,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999239262285869,
|
|
"loss": 6.2069,
|
|
"mean_token_accuracy": 0.1320282258093357,
|
|
"num_tokens": 4195016.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 6.386849069595337,
|
|
"epoch": 0.1813958086200079,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004999230058566495,
|
|
"loss": 6.1467,
|
|
"mean_token_accuracy": 0.14474730044603348,
|
|
"num_tokens": 4206798.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 6.370251035690307,
|
|
"epoch": 0.1818900751285093,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004999220799515713,
|
|
"loss": 6.131,
|
|
"mean_token_accuracy": 0.13890684843063356,
|
|
"num_tokens": 4218784.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 6.337589168548584,
|
|
"epoch": 0.18238434163701067,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.000499921148513375,
|
|
"loss": 6.157,
|
|
"mean_token_accuracy": 0.13808465152978897,
|
|
"num_tokens": 4229875.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 6.241511249542237,
|
|
"epoch": 0.18287860814551207,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999202115420836,
|
|
"loss": 6.1431,
|
|
"mean_token_accuracy": 0.1402451813220978,
|
|
"num_tokens": 4241801.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 6.385967588424682,
|
|
"epoch": 0.18337287465401345,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999192690377201,
|
|
"loss": 6.1187,
|
|
"mean_token_accuracy": 0.1410462327301502,
|
|
"num_tokens": 4253168.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 6.331370687484741,
|
|
"epoch": 0.18386714116251482,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004999183210003076,
|
|
"loss": 6.1497,
|
|
"mean_token_accuracy": 0.13985029086470604,
|
|
"num_tokens": 4264791.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 6.294579124450683,
|
|
"epoch": 0.18436140767101622,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004999173674298696,
|
|
"loss": 6.0931,
|
|
"mean_token_accuracy": 0.14586635753512384,
|
|
"num_tokens": 4275816.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 6.413454818725586,
|
|
"epoch": 0.1848556741795176,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004999164083264294,
|
|
"loss": 6.1696,
|
|
"mean_token_accuracy": 0.13626813441514968,
|
|
"num_tokens": 4287806.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 6.289838790893555,
|
|
"epoch": 0.185349940688019,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004999154436900108,
|
|
"loss": 6.2006,
|
|
"mean_token_accuracy": 0.14326421394944192,
|
|
"num_tokens": 4300319.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 6.346887874603271,
|
|
"epoch": 0.18584420719652037,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004999144735206373,
|
|
"loss": 6.2017,
|
|
"mean_token_accuracy": 0.13601267710328102,
|
|
"num_tokens": 4310987.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 6.371985960006714,
|
|
"epoch": 0.18633847370502174,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004999134978183328,
|
|
"loss": 6.1595,
|
|
"mean_token_accuracy": 0.14411325454711915,
|
|
"num_tokens": 4322438.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 6.385655212402344,
|
|
"epoch": 0.18683274021352314,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004999125165831215,
|
|
"loss": 6.1898,
|
|
"mean_token_accuracy": 0.1465175122022629,
|
|
"num_tokens": 4333242.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 6.255407238006592,
|
|
"epoch": 0.1873270067220245,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999115298150273,
|
|
"loss": 6.0597,
|
|
"mean_token_accuracy": 0.14152660295367242,
|
|
"num_tokens": 4343660.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 6.372375345230102,
|
|
"epoch": 0.1878212732305259,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004999105375140746,
|
|
"loss": 6.1347,
|
|
"mean_token_accuracy": 0.14555889666080474,
|
|
"num_tokens": 4353382.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 6.268469953536988,
|
|
"epoch": 0.1883155397390273,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999095396802878,
|
|
"loss": 6.1092,
|
|
"mean_token_accuracy": 0.13386452049016953,
|
|
"num_tokens": 4364614.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 6.378190708160401,
|
|
"epoch": 0.18880980624752866,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004999085363136913,
|
|
"loss": 6.1705,
|
|
"mean_token_accuracy": 0.13516660034656525,
|
|
"num_tokens": 4376001.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 6.323903131484985,
|
|
"epoch": 0.18930407275603006,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00049990752741431,
|
|
"loss": 6.1771,
|
|
"mean_token_accuracy": 0.13419201746582984,
|
|
"num_tokens": 4387083.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 6.297012424468994,
|
|
"epoch": 0.18979833926453143,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004999065129821687,
|
|
"loss": 6.1337,
|
|
"mean_token_accuracy": 0.141158889234066,
|
|
"num_tokens": 4399977.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 6.267847394943237,
|
|
"epoch": 0.1902926057730328,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0004999054930172923,
|
|
"loss": 6.066,
|
|
"mean_token_accuracy": 0.14234692901372908,
|
|
"num_tokens": 4411609.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 6.402806425094605,
|
|
"epoch": 0.1907868722815342,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999044675197058,
|
|
"loss": 6.1835,
|
|
"mean_token_accuracy": 0.14248162806034087,
|
|
"num_tokens": 4423567.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 6.22545223236084,
|
|
"epoch": 0.19128113879003558,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999034364894345,
|
|
"loss": 6.141,
|
|
"mean_token_accuracy": 0.1415740117430687,
|
|
"num_tokens": 4435410.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 6.422749614715576,
|
|
"epoch": 0.19177540529853698,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999023999265039,
|
|
"loss": 6.1708,
|
|
"mean_token_accuracy": 0.13967961594462394,
|
|
"num_tokens": 4445279.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 6.232316732406616,
|
|
"epoch": 0.19226967180703836,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999013578309392,
|
|
"loss": 6.0375,
|
|
"mean_token_accuracy": 0.14595175608992578,
|
|
"num_tokens": 4456017.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 6.297853326797485,
|
|
"epoch": 0.19276393831553973,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999003102027663,
|
|
"loss": 6.0303,
|
|
"mean_token_accuracy": 0.15286524295806886,
|
|
"num_tokens": 4466759.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 6.268013620376587,
|
|
"epoch": 0.19325820482404113,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000499899257042011,
|
|
"loss": 6.1067,
|
|
"mean_token_accuracy": 0.14023480713367462,
|
|
"num_tokens": 4478316.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 6.377132701873779,
|
|
"epoch": 0.1937524713325425,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004998981983486988,
|
|
"loss": 6.1813,
|
|
"mean_token_accuracy": 0.13629069328308105,
|
|
"num_tokens": 4490775.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 6.301769161224366,
|
|
"epoch": 0.1942467378410439,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004998971341228563,
|
|
"loss": 6.03,
|
|
"mean_token_accuracy": 0.14630968272686004,
|
|
"num_tokens": 4501622.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 6.303157186508178,
|
|
"epoch": 0.19474100434954528,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004998960643645093,
|
|
"loss": 6.1343,
|
|
"mean_token_accuracy": 0.1427514761686325,
|
|
"num_tokens": 4512279.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 6.305225229263305,
|
|
"epoch": 0.19523527085804665,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004998949890736841,
|
|
"loss": 6.121,
|
|
"mean_token_accuracy": 0.1444181725382805,
|
|
"num_tokens": 4524087.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 6.310341024398804,
|
|
"epoch": 0.19572953736654805,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004998939082504075,
|
|
"loss": 6.1019,
|
|
"mean_token_accuracy": 0.14459830969572068,
|
|
"num_tokens": 4534816.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 6.280261421203614,
|
|
"epoch": 0.19622380387504942,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004998928218947057,
|
|
"loss": 6.074,
|
|
"mean_token_accuracy": 0.1454021230340004,
|
|
"num_tokens": 4545725.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 6.377605772018432,
|
|
"epoch": 0.1967180703835508,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004998917300066056,
|
|
"loss": 6.1899,
|
|
"mean_token_accuracy": 0.1378871500492096,
|
|
"num_tokens": 4557003.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 6.222435426712036,
|
|
"epoch": 0.1972123368920522,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004998906325861342,
|
|
"loss": 6.101,
|
|
"mean_token_accuracy": 0.14017595276236533,
|
|
"num_tokens": 4568768.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 6.241748237609864,
|
|
"epoch": 0.19770660340055357,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004998895296333183,
|
|
"loss": 6.0654,
|
|
"mean_token_accuracy": 0.14961813986301423,
|
|
"num_tokens": 4581116.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 6.334244871139527,
|
|
"epoch": 0.19820086990905497,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000499888421148185,
|
|
"loss": 6.145,
|
|
"mean_token_accuracy": 0.14341819882392884,
|
|
"num_tokens": 4592660.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 6.315072679519654,
|
|
"epoch": 0.19869513641755635,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004998873071307617,
|
|
"loss": 6.115,
|
|
"mean_token_accuracy": 0.1400900736451149,
|
|
"num_tokens": 4603057.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 6.237052297592163,
|
|
"epoch": 0.19918940292605772,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004998861875810758,
|
|
"loss": 6.0078,
|
|
"mean_token_accuracy": 0.15424127280712127,
|
|
"num_tokens": 4613546.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 6.281518363952637,
|
|
"epoch": 0.19968366943455912,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004998850624991548,
|
|
"loss": 6.0869,
|
|
"mean_token_accuracy": 0.14481984600424766,
|
|
"num_tokens": 4624613.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 6.22469654083252,
|
|
"epoch": 0.2001779359430605,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004998839318850263,
|
|
"loss": 6.0521,
|
|
"mean_token_accuracy": 0.14788607954978944,
|
|
"num_tokens": 4636004.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 6.255212354660034,
|
|
"epoch": 0.2006722024515619,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004998827957387182,
|
|
"loss": 6.1074,
|
|
"mean_token_accuracy": 0.14652306362986564,
|
|
"num_tokens": 4646450.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 6.283095216751098,
|
|
"epoch": 0.20116646896006327,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004998816540602584,
|
|
"loss": 6.094,
|
|
"mean_token_accuracy": 0.1446727216243744,
|
|
"num_tokens": 4658928.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 6.248164463043213,
|
|
"epoch": 0.20166073546856464,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004998805068496752,
|
|
"loss": 6.1172,
|
|
"mean_token_accuracy": 0.14286085069179535,
|
|
"num_tokens": 4670392.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 6.35498571395874,
|
|
"epoch": 0.20215500197706604,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004998793541069964,
|
|
"loss": 6.0934,
|
|
"mean_token_accuracy": 0.1423612579703331,
|
|
"num_tokens": 4680605.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 6.207654523849487,
|
|
"epoch": 0.2026492684855674,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004998781958322507,
|
|
"loss": 6.0135,
|
|
"mean_token_accuracy": 0.14874968454241752,
|
|
"num_tokens": 4694250.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 6.302755451202392,
|
|
"epoch": 0.20314353499406881,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004998770320254666,
|
|
"loss": 6.1632,
|
|
"mean_token_accuracy": 0.1457796052098274,
|
|
"num_tokens": 4705754.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 6.281993579864502,
|
|
"epoch": 0.2036378015025702,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004998758626866725,
|
|
"loss": 6.0476,
|
|
"mean_token_accuracy": 0.1467272847890854,
|
|
"num_tokens": 4716393.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 6.196063613891601,
|
|
"epoch": 0.20413206801107156,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004998746878158973,
|
|
"loss": 6.0481,
|
|
"mean_token_accuracy": 0.14491303488612176,
|
|
"num_tokens": 4726894.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 6.244138574600219,
|
|
"epoch": 0.20462633451957296,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004998735074131699,
|
|
"loss": 6.0629,
|
|
"mean_token_accuracy": 0.14195707365870475,
|
|
"num_tokens": 4737488.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 6.279796409606933,
|
|
"epoch": 0.20512060102807433,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004998723214785194,
|
|
"loss": 6.1494,
|
|
"mean_token_accuracy": 0.1365630991756916,
|
|
"num_tokens": 4749131.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 6.2584418773651125,
|
|
"epoch": 0.20561486753657574,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004998711300119749,
|
|
"loss": 6.0195,
|
|
"mean_token_accuracy": 0.1438528060913086,
|
|
"num_tokens": 4760503.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 6.319554138183594,
|
|
"epoch": 0.2061091340450771,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004998699330135656,
|
|
"loss": 6.1319,
|
|
"mean_token_accuracy": 0.14361703097820283,
|
|
"num_tokens": 4771965.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 6.23747992515564,
|
|
"epoch": 0.20660340055357848,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004998687304833212,
|
|
"loss": 6.1247,
|
|
"mean_token_accuracy": 0.14026476591825485,
|
|
"num_tokens": 4784215.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 6.243879270553589,
|
|
"epoch": 0.20709766706207988,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000499867522421271,
|
|
"loss": 6.0339,
|
|
"mean_token_accuracy": 0.1493344783782959,
|
|
"num_tokens": 4796410.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 6.184952163696289,
|
|
"epoch": 0.20759193357058126,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000499866308827445,
|
|
"loss": 6.0151,
|
|
"mean_token_accuracy": 0.15359587073326111,
|
|
"num_tokens": 4806990.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 6.302822113037109,
|
|
"epoch": 0.20808620007908263,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004998650897018729,
|
|
"loss": 6.0734,
|
|
"mean_token_accuracy": 0.1439187079668045,
|
|
"num_tokens": 4818903.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 6.244305086135864,
|
|
"epoch": 0.20858046658758403,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004998638650445847,
|
|
"loss": 6.0081,
|
|
"mean_token_accuracy": 0.1505424477159977,
|
|
"num_tokens": 4829130.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 6.217615032196045,
|
|
"epoch": 0.2090747330960854,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004998626348556105,
|
|
"loss": 6.0298,
|
|
"mean_token_accuracy": 0.15414684265851974,
|
|
"num_tokens": 4840424.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 6.263989543914795,
|
|
"epoch": 0.2095689996045868,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004998613991349807,
|
|
"loss": 6.0376,
|
|
"mean_token_accuracy": 0.14594117775559426,
|
|
"num_tokens": 4851674.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 6.096542119979858,
|
|
"epoch": 0.21006326611308818,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004998601578827255,
|
|
"loss": 5.9601,
|
|
"mean_token_accuracy": 0.15645498782396317,
|
|
"num_tokens": 4863884.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 6.19989767074585,
|
|
"epoch": 0.21055753262158955,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004998589110988755,
|
|
"loss": 6.0415,
|
|
"mean_token_accuracy": 0.15132073909044266,
|
|
"num_tokens": 4875582.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 6.274530458450317,
|
|
"epoch": 0.21105179913009095,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004998576587834614,
|
|
"loss": 6.0528,
|
|
"mean_token_accuracy": 0.14546677693724633,
|
|
"num_tokens": 4888325.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 6.172981452941895,
|
|
"epoch": 0.21154606563859232,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000499856400936514,
|
|
"loss": 6.0524,
|
|
"mean_token_accuracy": 0.14551248848438264,
|
|
"num_tokens": 4899631.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 6.245505332946777,
|
|
"epoch": 0.21204033214709372,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004998551375580643,
|
|
"loss": 6.0628,
|
|
"mean_token_accuracy": 0.14277451261878013,
|
|
"num_tokens": 4910954.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 6.237064695358276,
|
|
"epoch": 0.2125345986555951,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004998538686481435,
|
|
"loss": 6.0531,
|
|
"mean_token_accuracy": 0.14683079421520234,
|
|
"num_tokens": 4921952.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 6.203374338150025,
|
|
"epoch": 0.21302886516409647,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004998525942067824,
|
|
"loss": 5.997,
|
|
"mean_token_accuracy": 0.15139593183994293,
|
|
"num_tokens": 4933984.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 6.283549213409424,
|
|
"epoch": 0.21352313167259787,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004998513142340128,
|
|
"loss": 6.1433,
|
|
"mean_token_accuracy": 0.14827414751052856,
|
|
"num_tokens": 4945926.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 6.204256439208985,
|
|
"epoch": 0.21401739818109924,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004998500287298657,
|
|
"loss": 6.0549,
|
|
"mean_token_accuracy": 0.15452619791030883,
|
|
"num_tokens": 4958123.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 6.319782829284668,
|
|
"epoch": 0.21451166468960065,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004998487376943733,
|
|
"loss": 6.1369,
|
|
"mean_token_accuracy": 0.1432581551373005,
|
|
"num_tokens": 4970409.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 6.231132888793946,
|
|
"epoch": 0.21500593119810202,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000499847441127567,
|
|
"loss": 6.0345,
|
|
"mean_token_accuracy": 0.1420576423406601,
|
|
"num_tokens": 4980429.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 6.213157796859742,
|
|
"epoch": 0.2155001977066034,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004998461390294787,
|
|
"loss": 6.0472,
|
|
"mean_token_accuracy": 0.14993677660822868,
|
|
"num_tokens": 4991790.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 6.261713266372681,
|
|
"epoch": 0.2159944642151048,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004998448314001405,
|
|
"loss": 6.069,
|
|
"mean_token_accuracy": 0.1435249201953411,
|
|
"num_tokens": 5003611.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 6.302479600906372,
|
|
"epoch": 0.21648873072360617,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004998435182395845,
|
|
"loss": 6.1232,
|
|
"mean_token_accuracy": 0.13771253526210786,
|
|
"num_tokens": 5014824.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 6.258434915542603,
|
|
"epoch": 0.21698299723210757,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004998421995478432,
|
|
"loss": 6.0768,
|
|
"mean_token_accuracy": 0.14596307873725892,
|
|
"num_tokens": 5026937.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 6.298146438598633,
|
|
"epoch": 0.21747726374060894,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004998408753249489,
|
|
"loss": 6.0927,
|
|
"mean_token_accuracy": 0.14147113561630248,
|
|
"num_tokens": 5037996.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 6.2773766040802,
|
|
"epoch": 0.2179715302491103,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004998395455709342,
|
|
"loss": 6.0422,
|
|
"mean_token_accuracy": 0.15115779191255568,
|
|
"num_tokens": 5050944.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 6.2505774974823,
|
|
"epoch": 0.2184657967576117,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004998382102858317,
|
|
"loss": 6.0871,
|
|
"mean_token_accuracy": 0.13831509053707122,
|
|
"num_tokens": 5062129.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 6.262403440475464,
|
|
"epoch": 0.2189600632661131,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004998368694696743,
|
|
"loss": 6.0806,
|
|
"mean_token_accuracy": 0.1436243861913681,
|
|
"num_tokens": 5073318.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 6.185823059082031,
|
|
"epoch": 0.21945432977461446,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004998355231224952,
|
|
"loss": 6.1047,
|
|
"mean_token_accuracy": 0.14238689020276069,
|
|
"num_tokens": 5084607.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 6.298590517044067,
|
|
"epoch": 0.21994859628311586,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004998341712443273,
|
|
"loss": 6.0071,
|
|
"mean_token_accuracy": 0.1467602863907814,
|
|
"num_tokens": 5096004.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 6.199511909484864,
|
|
"epoch": 0.22044286279161723,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000499832813835204,
|
|
"loss": 6.0052,
|
|
"mean_token_accuracy": 0.14269362017512321,
|
|
"num_tokens": 5107171.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 6.184848213195801,
|
|
"epoch": 0.22093712930011863,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004998314508951584,
|
|
"loss": 5.9611,
|
|
"mean_token_accuracy": 0.15424952656030655,
|
|
"num_tokens": 5119249.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 6.14829421043396,
|
|
"epoch": 0.22143139580862,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004998300824242244,
|
|
"loss": 5.976,
|
|
"mean_token_accuracy": 0.14679246842861177,
|
|
"num_tokens": 5130798.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 6.275025987625122,
|
|
"epoch": 0.22192566231712138,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004998287084224353,
|
|
"loss": 6.1437,
|
|
"mean_token_accuracy": 0.13429683446884155,
|
|
"num_tokens": 5141376.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 6.154032039642334,
|
|
"epoch": 0.22241992882562278,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004998273288898252,
|
|
"loss": 5.9357,
|
|
"mean_token_accuracy": 0.15065955817699433,
|
|
"num_tokens": 5152888.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 6.156353521347046,
|
|
"epoch": 0.22291419533412415,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.000499825943826428,
|
|
"loss": 6.0392,
|
|
"mean_token_accuracy": 0.14408660531044007,
|
|
"num_tokens": 5165720.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 6.161768770217895,
|
|
"epoch": 0.22340846184262556,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004998245532322776,
|
|
"loss": 5.9084,
|
|
"mean_token_accuracy": 0.15456337854266167,
|
|
"num_tokens": 5177345.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 6.202211046218872,
|
|
"epoch": 0.22390272835112693,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004998231571074083,
|
|
"loss": 5.9329,
|
|
"mean_token_accuracy": 0.1534419000148773,
|
|
"num_tokens": 5187385.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 6.170833683013916,
|
|
"epoch": 0.2243969948596283,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004998217554518545,
|
|
"loss": 6.0403,
|
|
"mean_token_accuracy": 0.14258744418621064,
|
|
"num_tokens": 5199552.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 6.200662231445312,
|
|
"epoch": 0.2248912613681297,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004998203482656507,
|
|
"loss": 6.007,
|
|
"mean_token_accuracy": 0.15123232156038285,
|
|
"num_tokens": 5210316.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 6.234782314300537,
|
|
"epoch": 0.22538552787663108,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004998189355488314,
|
|
"loss": 6.0744,
|
|
"mean_token_accuracy": 0.14858508333563805,
|
|
"num_tokens": 5221595.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 6.227956914901734,
|
|
"epoch": 0.22587979438513248,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004998175173014313,
|
|
"loss": 6.0504,
|
|
"mean_token_accuracy": 0.14020669236779212,
|
|
"num_tokens": 5232882.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 6.238587951660156,
|
|
"epoch": 0.22637406089363385,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004998160935234857,
|
|
"loss": 6.1441,
|
|
"mean_token_accuracy": 0.14422304853796958,
|
|
"num_tokens": 5244841.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 6.146130466461182,
|
|
"epoch": 0.22686832740213522,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000499814664215029,
|
|
"loss": 5.9581,
|
|
"mean_token_accuracy": 0.152134919911623,
|
|
"num_tokens": 5256162.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 6.190051794052124,
|
|
"epoch": 0.22736259391063662,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004998132293760968,
|
|
"loss": 5.8762,
|
|
"mean_token_accuracy": 0.15613968968391417,
|
|
"num_tokens": 5267130.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 6.0447431087493895,
|
|
"epoch": 0.227856860419138,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004998117890067243,
|
|
"loss": 6.0008,
|
|
"mean_token_accuracy": 0.1474076971411705,
|
|
"num_tokens": 5279394.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 6.172190380096436,
|
|
"epoch": 0.22835112692763937,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004998103431069469,
|
|
"loss": 5.9787,
|
|
"mean_token_accuracy": 0.14816260784864427,
|
|
"num_tokens": 5291068.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 6.2173741340637205,
|
|
"epoch": 0.22884539343614077,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004998088916768002,
|
|
"loss": 5.9391,
|
|
"mean_token_accuracy": 0.15463171005249024,
|
|
"num_tokens": 5300924.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 6.124597406387329,
|
|
"epoch": 0.22933965994464214,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004998074347163199,
|
|
"loss": 5.9419,
|
|
"mean_token_accuracy": 0.15559005588293076,
|
|
"num_tokens": 5311998.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 6.211737871170044,
|
|
"epoch": 0.22983392645314354,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004998059722255418,
|
|
"loss": 5.9836,
|
|
"mean_token_accuracy": 0.14932580888271332,
|
|
"num_tokens": 5322163.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 6.127090263366699,
|
|
"epoch": 0.23032819296164492,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004998045042045019,
|
|
"loss": 5.981,
|
|
"mean_token_accuracy": 0.15460044890642166,
|
|
"num_tokens": 5331638.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 6.200595951080322,
|
|
"epoch": 0.2308224594701463,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004998030306532363,
|
|
"loss": 5.9814,
|
|
"mean_token_accuracy": 0.15659119933843613,
|
|
"num_tokens": 5342399.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 6.172941446304321,
|
|
"epoch": 0.2313167259786477,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004998015515717813,
|
|
"loss": 6.028,
|
|
"mean_token_accuracy": 0.14542844295501708,
|
|
"num_tokens": 5353481.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 6.180023050308227,
|
|
"epoch": 0.23181099248714906,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004998000669601733,
|
|
"loss": 5.9503,
|
|
"mean_token_accuracy": 0.1496099278330803,
|
|
"num_tokens": 5365167.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 6.114148902893066,
|
|
"epoch": 0.23230525899565047,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004997985768184488,
|
|
"loss": 5.9658,
|
|
"mean_token_accuracy": 0.14984080642461778,
|
|
"num_tokens": 5375809.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 6.076729488372803,
|
|
"epoch": 0.23279952550415184,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004997970811466443,
|
|
"loss": 5.8811,
|
|
"mean_token_accuracy": 0.15644187480211258,
|
|
"num_tokens": 5387372.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 6.341212034225464,
|
|
"epoch": 0.2332937920126532,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004997955799447969,
|
|
"loss": 6.0607,
|
|
"mean_token_accuracy": 0.14222865626215936,
|
|
"num_tokens": 5399210.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 6.121503829956055,
|
|
"epoch": 0.2337880585211546,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004997940732129432,
|
|
"loss": 5.9777,
|
|
"mean_token_accuracy": 0.14460844695568084,
|
|
"num_tokens": 5410969.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 6.139842796325683,
|
|
"epoch": 0.23428232502965599,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004997925609511205,
|
|
"loss": 5.9347,
|
|
"mean_token_accuracy": 0.15759174525737762,
|
|
"num_tokens": 5422569.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 6.168006563186646,
|
|
"epoch": 0.2347765915381574,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004997910431593659,
|
|
"loss": 5.9449,
|
|
"mean_token_accuracy": 0.15459632575511933,
|
|
"num_tokens": 5434592.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 6.1388246536254885,
|
|
"epoch": 0.23527085804665876,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004997895198377168,
|
|
"loss": 6.0155,
|
|
"mean_token_accuracy": 0.15070839375257492,
|
|
"num_tokens": 5446525.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 6.147398662567139,
|
|
"epoch": 0.23576512455516013,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004997879909862105,
|
|
"loss": 6.0029,
|
|
"mean_token_accuracy": 0.14832279682159424,
|
|
"num_tokens": 5457327.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 6.24454779624939,
|
|
"epoch": 0.23625939106366153,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000499786456604885,
|
|
"loss": 6.0312,
|
|
"mean_token_accuracy": 0.14286630377173423,
|
|
"num_tokens": 5468533.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 6.117534255981445,
|
|
"epoch": 0.2367536575721629,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.0004997849166937776,
|
|
"loss": 5.9146,
|
|
"mean_token_accuracy": 0.14934935867786409,
|
|
"num_tokens": 5480652.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 6.200796270370484,
|
|
"epoch": 0.2372479240806643,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004997833712529265,
|
|
"loss": 5.9595,
|
|
"mean_token_accuracy": 0.14828556329011916,
|
|
"num_tokens": 5491446.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 6.11882472038269,
|
|
"epoch": 0.23774219058916568,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004997818202823695,
|
|
"loss": 5.9543,
|
|
"mean_token_accuracy": 0.1541999250650406,
|
|
"num_tokens": 5502886.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 6.157952308654785,
|
|
"epoch": 0.23823645709766705,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004997802637821449,
|
|
"loss": 5.9933,
|
|
"mean_token_accuracy": 0.14664816558361055,
|
|
"num_tokens": 5514305.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 6.153297376632691,
|
|
"epoch": 0.23873072360616845,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004997787017522909,
|
|
"loss": 5.9195,
|
|
"mean_token_accuracy": 0.16202860027551652,
|
|
"num_tokens": 5525913.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 6.031432867050171,
|
|
"epoch": 0.23922499011466983,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004997771341928459,
|
|
"loss": 5.9028,
|
|
"mean_token_accuracy": 0.15993187874555587,
|
|
"num_tokens": 5537069.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 6.132914733886719,
|
|
"epoch": 0.2397192566231712,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004997755611038487,
|
|
"loss": 6.0155,
|
|
"mean_token_accuracy": 0.1475693963468075,
|
|
"num_tokens": 5547988.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 6.177093124389648,
|
|
"epoch": 0.2402135231316726,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004997739824853378,
|
|
"loss": 6.0784,
|
|
"mean_token_accuracy": 0.14507206082344054,
|
|
"num_tokens": 5560400.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 6.125483798980713,
|
|
"epoch": 0.24070778964017397,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004997723983373519,
|
|
"loss": 5.9408,
|
|
"mean_token_accuracy": 0.1547750160098076,
|
|
"num_tokens": 5572291.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 6.23362283706665,
|
|
"epoch": 0.24120205614867538,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004997708086599303,
|
|
"loss": 5.9657,
|
|
"mean_token_accuracy": 0.14544104561209678,
|
|
"num_tokens": 5583583.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 6.20045838356018,
|
|
"epoch": 0.24169632265717675,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0004997692134531119,
|
|
"loss": 5.9888,
|
|
"mean_token_accuracy": 0.1482198178768158,
|
|
"num_tokens": 5595499.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 6.0919482707977295,
|
|
"epoch": 0.24219058916567812,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000499767612716936,
|
|
"loss": 5.9138,
|
|
"mean_token_accuracy": 0.1550483450293541,
|
|
"num_tokens": 5606745.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 6.092108058929443,
|
|
"epoch": 0.24268485567417952,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004997660064514419,
|
|
"loss": 5.9935,
|
|
"mean_token_accuracy": 0.15098343044519424,
|
|
"num_tokens": 5618226.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 6.259077262878418,
|
|
"epoch": 0.2431791221826809,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004997643946566691,
|
|
"loss": 6.0146,
|
|
"mean_token_accuracy": 0.14846471548080445,
|
|
"num_tokens": 5628941.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 6.092586326599121,
|
|
"epoch": 0.2436733886911823,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004997627773326574,
|
|
"loss": 5.8118,
|
|
"mean_token_accuracy": 0.1619025096297264,
|
|
"num_tokens": 5641077.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 6.1263285160064695,
|
|
"epoch": 0.24416765519968367,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004997611544794465,
|
|
"loss": 6.1202,
|
|
"mean_token_accuracy": 0.14339257180690765,
|
|
"num_tokens": 5652156.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 6.257670974731445,
|
|
"epoch": 0.24466192170818504,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004997595260970764,
|
|
"loss": 6.0358,
|
|
"mean_token_accuracy": 0.14745921045541763,
|
|
"num_tokens": 5663988.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 6.131538438796997,
|
|
"epoch": 0.24515618821668644,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.000499757892185587,
|
|
"loss": 5.9627,
|
|
"mean_token_accuracy": 0.14847193360328675,
|
|
"num_tokens": 5676302.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 6.159386014938354,
|
|
"epoch": 0.24565045472518782,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004997562527450187,
|
|
"loss": 5.9976,
|
|
"mean_token_accuracy": 0.15123492628335952,
|
|
"num_tokens": 5686686.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 6.16356930732727,
|
|
"epoch": 0.24614472123368922,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004997546077754116,
|
|
"loss": 5.9342,
|
|
"mean_token_accuracy": 0.16507403403520585,
|
|
"num_tokens": 5698379.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 6.121092033386231,
|
|
"epoch": 0.2466389877421906,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004997529572768063,
|
|
"loss": 5.9167,
|
|
"mean_token_accuracy": 0.154787477850914,
|
|
"num_tokens": 5708891.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 6.0729203701019285,
|
|
"epoch": 0.24713325425069196,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004997513012492434,
|
|
"loss": 5.9648,
|
|
"mean_token_accuracy": 0.1536966696381569,
|
|
"num_tokens": 5720326.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 6.153505516052246,
|
|
"epoch": 0.24762752075919336,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004997496396927636,
|
|
"loss": 5.8959,
|
|
"mean_token_accuracy": 0.15583585798740388,
|
|
"num_tokens": 5731121.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 6.155553531646729,
|
|
"epoch": 0.24812178726769474,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004997479726074077,
|
|
"loss": 5.93,
|
|
"mean_token_accuracy": 0.15216445177793503,
|
|
"num_tokens": 5741637.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 6.1915891647338865,
|
|
"epoch": 0.24861605377619614,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000499746299993217,
|
|
"loss": 5.9735,
|
|
"mean_token_accuracy": 0.15114908963441848,
|
|
"num_tokens": 5752659.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 6.094469976425171,
|
|
"epoch": 0.2491103202846975,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004997446218502324,
|
|
"loss": 5.9525,
|
|
"mean_token_accuracy": 0.15155530497431755,
|
|
"num_tokens": 5763829.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 6.132277250289917,
|
|
"epoch": 0.24960458679319888,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004997429381784952,
|
|
"loss": 6.022,
|
|
"mean_token_accuracy": 0.14910133332014083,
|
|
"num_tokens": 5775848.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 6.120943069458008,
|
|
"epoch": 0.2500988533017003,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004997412489780469,
|
|
"loss": 5.9612,
|
|
"mean_token_accuracy": 0.15950672328472137,
|
|
"num_tokens": 5786595.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 6.140192317962646,
|
|
"epoch": 0.25059311981020166,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000499739554248929,
|
|
"loss": 5.8703,
|
|
"mean_token_accuracy": 0.16149618923664094,
|
|
"num_tokens": 5797841.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 6.0338222026824955,
|
|
"epoch": 0.25108738631870303,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004997378539911832,
|
|
"loss": 5.8765,
|
|
"mean_token_accuracy": 0.15623045712709427,
|
|
"num_tokens": 5809106.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 6.107133483886718,
|
|
"epoch": 0.2515816528272044,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004997361482048513,
|
|
"loss": 5.8885,
|
|
"mean_token_accuracy": 0.15565961226820946,
|
|
"num_tokens": 5820316.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 6.121990442276001,
|
|
"epoch": 0.25207591933570583,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004997344368899753,
|
|
"loss": 6.02,
|
|
"mean_token_accuracy": 0.14788583666086197,
|
|
"num_tokens": 5833137.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 6.175726079940796,
|
|
"epoch": 0.2525701858442072,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004997327200465972,
|
|
"loss": 5.9304,
|
|
"mean_token_accuracy": 0.15318210422992706,
|
|
"num_tokens": 5844088.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 6.133507585525512,
|
|
"epoch": 0.2530644523527086,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004997309976747594,
|
|
"loss": 5.9265,
|
|
"mean_token_accuracy": 0.15268272608518602,
|
|
"num_tokens": 5854546.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 6.149356508255005,
|
|
"epoch": 0.25355871886120995,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004997292697745042,
|
|
"loss": 5.9255,
|
|
"mean_token_accuracy": 0.1569249227643013,
|
|
"num_tokens": 5865326.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 6.136588287353516,
|
|
"epoch": 0.2540529853697113,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004997275363458741,
|
|
"loss": 5.8293,
|
|
"mean_token_accuracy": 0.15652241110801696,
|
|
"num_tokens": 5876487.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 6.0281952857971195,
|
|
"epoch": 0.25454725187821275,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004997257973889118,
|
|
"loss": 5.9355,
|
|
"mean_token_accuracy": 0.14910998791456223,
|
|
"num_tokens": 5887846.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 6.157446002960205,
|
|
"epoch": 0.25504151838671413,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004997240529036599,
|
|
"loss": 5.8919,
|
|
"mean_token_accuracy": 0.15892206728458405,
|
|
"num_tokens": 5898572.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 6.071728754043579,
|
|
"epoch": 0.2555357848952155,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004997223028901615,
|
|
"loss": 5.9449,
|
|
"mean_token_accuracy": 0.15322234779596328,
|
|
"num_tokens": 5908927.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 6.124873542785645,
|
|
"epoch": 0.2560300514037169,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004997205473484596,
|
|
"loss": 5.9037,
|
|
"mean_token_accuracy": 0.1580432742834091,
|
|
"num_tokens": 5919807.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 6.10838508605957,
|
|
"epoch": 0.25652431791221825,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004997187862785974,
|
|
"loss": 5.9259,
|
|
"mean_token_accuracy": 0.15537415742874144,
|
|
"num_tokens": 5931160.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 6.13609881401062,
|
|
"epoch": 0.2570185844207197,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004997170196806181,
|
|
"loss": 5.9987,
|
|
"mean_token_accuracy": 0.14711610227823257,
|
|
"num_tokens": 5943694.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 6.120221710205078,
|
|
"epoch": 0.25751285092922105,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004997152475545652,
|
|
"loss": 5.8863,
|
|
"mean_token_accuracy": 0.15604913979768753,
|
|
"num_tokens": 5954543.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 6.046738481521606,
|
|
"epoch": 0.2580071174377224,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004997134699004825,
|
|
"loss": 5.9358,
|
|
"mean_token_accuracy": 0.15628221184015273,
|
|
"num_tokens": 5965080.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 6.095473432540894,
|
|
"epoch": 0.2585013839462238,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004997116867184136,
|
|
"loss": 5.937,
|
|
"mean_token_accuracy": 0.14930862933397293,
|
|
"num_tokens": 5975314.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 6.180641603469849,
|
|
"epoch": 0.25899565045472517,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004997098980084022,
|
|
"loss": 5.9354,
|
|
"mean_token_accuracy": 0.15294832438230516,
|
|
"num_tokens": 5985927.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 6.087760400772095,
|
|
"epoch": 0.2594899169632266,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004997081037704926,
|
|
"loss": 5.9517,
|
|
"mean_token_accuracy": 0.15042607337236405,
|
|
"num_tokens": 5997311.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 6.086427164077759,
|
|
"epoch": 0.25998418347172797,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004997063040047288,
|
|
"loss": 5.9843,
|
|
"mean_token_accuracy": 0.1468744695186615,
|
|
"num_tokens": 6008595.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 6.178718137741089,
|
|
"epoch": 0.26047844998022934,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004997044987111548,
|
|
"loss": 5.9078,
|
|
"mean_token_accuracy": 0.1546178936958313,
|
|
"num_tokens": 6019055.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 6.0701744079589846,
|
|
"epoch": 0.2609727164887307,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004997026878898155,
|
|
"loss": 6.0342,
|
|
"mean_token_accuracy": 0.14548355862498283,
|
|
"num_tokens": 6030029.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 6.153164005279541,
|
|
"epoch": 0.2614669829972321,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004997008715407552,
|
|
"loss": 5.9902,
|
|
"mean_token_accuracy": 0.1496541254222393,
|
|
"num_tokens": 6041495.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 6.143697834014892,
|
|
"epoch": 0.2619612495057335,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004996990496640185,
|
|
"loss": 5.8818,
|
|
"mean_token_accuracy": 0.15100248157978058,
|
|
"num_tokens": 6052307.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 6.028233671188355,
|
|
"epoch": 0.2624555160142349,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004996972222596505,
|
|
"loss": 5.8423,
|
|
"mean_token_accuracy": 0.15583381801843643,
|
|
"num_tokens": 6062550.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 6.0477863311767575,
|
|
"epoch": 0.26294978252273626,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004996953893276958,
|
|
"loss": 5.8781,
|
|
"mean_token_accuracy": 0.14674556404352188,
|
|
"num_tokens": 6074450.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 6.078070211410522,
|
|
"epoch": 0.26344404903123764,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.0004996935508681997,
|
|
"loss": 5.9265,
|
|
"mean_token_accuracy": 0.1591618239879608,
|
|
"num_tokens": 6085749.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 6.09918999671936,
|
|
"epoch": 0.263938315539739,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004996917068812075,
|
|
"loss": 5.8488,
|
|
"mean_token_accuracy": 0.15992971658706664,
|
|
"num_tokens": 6096438.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 6.059639120101929,
|
|
"epoch": 0.26443258204824044,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004996898573667643,
|
|
"loss": 5.9086,
|
|
"mean_token_accuracy": 0.14664011746644973,
|
|
"num_tokens": 6108386.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 6.083099842071533,
|
|
"epoch": 0.2649268485567418,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004996880023249158,
|
|
"loss": 5.8096,
|
|
"mean_token_accuracy": 0.16396086513996125,
|
|
"num_tokens": 6118771.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 6.0931103229522705,
|
|
"epoch": 0.2654211150652432,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004996861417557076,
|
|
"loss": 5.9009,
|
|
"mean_token_accuracy": 0.15202439576387405,
|
|
"num_tokens": 6130718.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 6.067891788482666,
|
|
"epoch": 0.26591538157374456,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004996842756591855,
|
|
"loss": 5.9283,
|
|
"mean_token_accuracy": 0.15080306231975554,
|
|
"num_tokens": 6142086.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 6.042020416259765,
|
|
"epoch": 0.26640964808224593,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004996824040353953,
|
|
"loss": 5.904,
|
|
"mean_token_accuracy": 0.1560404196381569,
|
|
"num_tokens": 6153857.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 6.241527795791626,
|
|
"epoch": 0.2669039145907473,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004996805268843832,
|
|
"loss": 6.0421,
|
|
"mean_token_accuracy": 0.14912690445780755,
|
|
"num_tokens": 6166796.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 6.072644186019898,
|
|
"epoch": 0.26739818109924873,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004996786442061952,
|
|
"loss": 5.8597,
|
|
"mean_token_accuracy": 0.15616288185119628,
|
|
"num_tokens": 6178046.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 6.0524310111999515,
|
|
"epoch": 0.2678924476077501,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004996767560008777,
|
|
"loss": 5.9115,
|
|
"mean_token_accuracy": 0.1550481289625168,
|
|
"num_tokens": 6190455.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 6.06898980140686,
|
|
"epoch": 0.2683867141162515,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004996748622684771,
|
|
"loss": 5.8107,
|
|
"mean_token_accuracy": 0.1622350811958313,
|
|
"num_tokens": 6201659.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 5.995411539077759,
|
|
"epoch": 0.26888098062475285,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004996729630090401,
|
|
"loss": 5.9163,
|
|
"mean_token_accuracy": 0.14895097762346268,
|
|
"num_tokens": 6213870.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 6.167082738876343,
|
|
"epoch": 0.2693752471332542,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004996710582226133,
|
|
"loss": 5.8605,
|
|
"mean_token_accuracy": 0.1600447803735733,
|
|
"num_tokens": 6224237.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 6.082718801498413,
|
|
"epoch": 0.26986951364175565,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004996691479092437,
|
|
"loss": 5.9085,
|
|
"mean_token_accuracy": 0.15194753110408782,
|
|
"num_tokens": 6235868.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 6.102431201934815,
|
|
"epoch": 0.270363780150257,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004996672320689782,
|
|
"loss": 5.9515,
|
|
"mean_token_accuracy": 0.14958198070526124,
|
|
"num_tokens": 6247994.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 5.958840847015381,
|
|
"epoch": 0.2708580466587584,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004996653107018638,
|
|
"loss": 5.8104,
|
|
"mean_token_accuracy": 0.16383833289146424,
|
|
"num_tokens": 6259654.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 6.046971559524536,
|
|
"epoch": 0.2713523131672598,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.000499663383807948,
|
|
"loss": 5.8601,
|
|
"mean_token_accuracy": 0.1581733912229538,
|
|
"num_tokens": 6270582.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 6.1495075702667235,
|
|
"epoch": 0.27184657967576115,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004996614513872781,
|
|
"loss": 5.8536,
|
|
"mean_token_accuracy": 0.1616063266992569,
|
|
"num_tokens": 6281088.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 6.113773393630981,
|
|
"epoch": 0.2723408461842626,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004996595134399017,
|
|
"loss": 6.0061,
|
|
"mean_token_accuracy": 0.1472685419023037,
|
|
"num_tokens": 6292920.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 6.09270396232605,
|
|
"epoch": 0.27283511269276395,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004996575699658664,
|
|
"loss": 5.8642,
|
|
"mean_token_accuracy": 0.1566832423210144,
|
|
"num_tokens": 6304980.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 6.006596946716309,
|
|
"epoch": 0.2733293792012653,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004996556209652199,
|
|
"loss": 5.8635,
|
|
"mean_token_accuracy": 0.1554800748825073,
|
|
"num_tokens": 6316175.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 6.153335475921631,
|
|
"epoch": 0.2738236457097667,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004996536664380104,
|
|
"loss": 5.9069,
|
|
"mean_token_accuracy": 0.1567239671945572,
|
|
"num_tokens": 6327590.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 6.017695617675781,
|
|
"epoch": 0.27431791221826807,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004996517063842859,
|
|
"loss": 5.9336,
|
|
"mean_token_accuracy": 0.15392989069223403,
|
|
"num_tokens": 6339066.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 6.086386060714721,
|
|
"epoch": 0.2748121787267695,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004996497408040944,
|
|
"loss": 5.8555,
|
|
"mean_token_accuracy": 0.15775285810232162,
|
|
"num_tokens": 6349485.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 5.981375408172608,
|
|
"epoch": 0.27530644523527087,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004996477696974845,
|
|
"loss": 5.8113,
|
|
"mean_token_accuracy": 0.15953710824251174,
|
|
"num_tokens": 6360775.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 6.034738063812256,
|
|
"epoch": 0.27580071174377224,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004996457930645046,
|
|
"loss": 5.8281,
|
|
"mean_token_accuracy": 0.15122711658477783,
|
|
"num_tokens": 6371645.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 6.013451719284058,
|
|
"epoch": 0.2762949782522736,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004996438109052034,
|
|
"loss": 5.8573,
|
|
"mean_token_accuracy": 0.15728358179330826,
|
|
"num_tokens": 6383877.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 6.08137354850769,
|
|
"epoch": 0.276789244760775,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004996418232196296,
|
|
"loss": 5.9411,
|
|
"mean_token_accuracy": 0.1478685200214386,
|
|
"num_tokens": 6394488.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 6.082438802719116,
|
|
"epoch": 0.2772835112692764,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004996398300078321,
|
|
"loss": 5.8731,
|
|
"mean_token_accuracy": 0.1562285304069519,
|
|
"num_tokens": 6404948.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 6.112468290328979,
|
|
"epoch": 0.2777777777777778,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004996378312698598,
|
|
"loss": 5.8919,
|
|
"mean_token_accuracy": 0.15620481222867966,
|
|
"num_tokens": 6416273.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 6.089010572433471,
|
|
"epoch": 0.27827204428627916,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004996358270057622,
|
|
"loss": 5.9114,
|
|
"mean_token_accuracy": 0.15275675654411316,
|
|
"num_tokens": 6427091.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 6.05656418800354,
|
|
"epoch": 0.27876631079478054,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004996338172155883,
|
|
"loss": 5.8618,
|
|
"mean_token_accuracy": 0.15593931525945665,
|
|
"num_tokens": 6438254.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 6.018289232254029,
|
|
"epoch": 0.2792605773032819,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004996318018993876,
|
|
"loss": 5.7876,
|
|
"mean_token_accuracy": 0.15474870353937148,
|
|
"num_tokens": 6450618.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 6.027504634857178,
|
|
"epoch": 0.27975484381178334,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004996297810572098,
|
|
"loss": 5.8482,
|
|
"mean_token_accuracy": 0.1601539134979248,
|
|
"num_tokens": 6462467.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 6.017673635482788,
|
|
"epoch": 0.2802491103202847,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004996277546891045,
|
|
"loss": 5.819,
|
|
"mean_token_accuracy": 0.1652624100446701,
|
|
"num_tokens": 6474206.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 6.056855010986328,
|
|
"epoch": 0.2807433768287861,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004996257227951216,
|
|
"loss": 5.9084,
|
|
"mean_token_accuracy": 0.15035663545131683,
|
|
"num_tokens": 6486405.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 6.039477729797364,
|
|
"epoch": 0.28123764333728746,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000499623685375311,
|
|
"loss": 5.8714,
|
|
"mean_token_accuracy": 0.15469448566436766,
|
|
"num_tokens": 6496949.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 6.027335119247437,
|
|
"epoch": 0.28173190984578883,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.000499621642429723,
|
|
"loss": 5.812,
|
|
"mean_token_accuracy": 0.16200096756219864,
|
|
"num_tokens": 6508913.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 6.040955018997193,
|
|
"epoch": 0.28222617635429026,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004996195939584078,
|
|
"loss": 5.8639,
|
|
"mean_token_accuracy": 0.1561881922185421,
|
|
"num_tokens": 6519837.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 6.054210805892945,
|
|
"epoch": 0.28272044286279163,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004996175399614156,
|
|
"loss": 5.8126,
|
|
"mean_token_accuracy": 0.1579636812210083,
|
|
"num_tokens": 6531409.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 5.995057392120361,
|
|
"epoch": 0.283214709371293,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004996154804387972,
|
|
"loss": 5.8556,
|
|
"mean_token_accuracy": 0.15890285670757293,
|
|
"num_tokens": 6543433.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 6.053872203826904,
|
|
"epoch": 0.2837089758797944,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000499613415390603,
|
|
"loss": 5.9406,
|
|
"mean_token_accuracy": 0.151950404047966,
|
|
"num_tokens": 6554511.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 6.0932821273803714,
|
|
"epoch": 0.28420324238829575,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000499611344816884,
|
|
"loss": 5.8527,
|
|
"mean_token_accuracy": 0.1596566081047058,
|
|
"num_tokens": 6564591.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 6.049549293518067,
|
|
"epoch": 0.2846975088967972,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.000499609268717691,
|
|
"loss": 5.8486,
|
|
"mean_token_accuracy": 0.15844885110855103,
|
|
"num_tokens": 6575684.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 6.073301649093628,
|
|
"epoch": 0.28519177540529855,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004996071870930753,
|
|
"loss": 5.9619,
|
|
"mean_token_accuracy": 0.15451340526342391,
|
|
"num_tokens": 6588504.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 5.963142967224121,
|
|
"epoch": 0.2856860419137999,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004996050999430878,
|
|
"loss": 5.7836,
|
|
"mean_token_accuracy": 0.15867630690336226,
|
|
"num_tokens": 6599969.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 6.050960683822632,
|
|
"epoch": 0.2861803084223013,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004996030072677802,
|
|
"loss": 5.8486,
|
|
"mean_token_accuracy": 0.163725782930851,
|
|
"num_tokens": 6611430.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 6.036673974990845,
|
|
"epoch": 0.2866745749308027,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004996009090672037,
|
|
"loss": 5.8447,
|
|
"mean_token_accuracy": 0.15628752261400222,
|
|
"num_tokens": 6623247.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 6.034439992904663,
|
|
"epoch": 0.28716884143930405,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004995988053414098,
|
|
"loss": 5.9669,
|
|
"mean_token_accuracy": 0.15123033076524733,
|
|
"num_tokens": 6635508.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 6.039534950256348,
|
|
"epoch": 0.2876631079478055,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004995966960904506,
|
|
"loss": 5.8779,
|
|
"mean_token_accuracy": 0.15228607207536698,
|
|
"num_tokens": 6646020.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 6.089818906784058,
|
|
"epoch": 0.28815737445630685,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004995945813143778,
|
|
"loss": 5.9051,
|
|
"mean_token_accuracy": 0.14688372910022734,
|
|
"num_tokens": 6658194.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 6.052178907394409,
|
|
"epoch": 0.2886516409648082,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004995924610132435,
|
|
"loss": 5.8738,
|
|
"mean_token_accuracy": 0.15363554805517196,
|
|
"num_tokens": 6670582.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 6.036479187011719,
|
|
"epoch": 0.2891459074733096,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004995903351870998,
|
|
"loss": 5.8224,
|
|
"mean_token_accuracy": 0.1569667100906372,
|
|
"num_tokens": 6680836.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 6.112790536880493,
|
|
"epoch": 0.28964017398181097,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000499588203835999,
|
|
"loss": 5.9134,
|
|
"mean_token_accuracy": 0.14992318898439408,
|
|
"num_tokens": 6693045.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 6.067016649246216,
|
|
"epoch": 0.2901344404903124,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004995860669599935,
|
|
"loss": 5.8637,
|
|
"mean_token_accuracy": 0.1490854561328888,
|
|
"num_tokens": 6704006.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 5.941438627243042,
|
|
"epoch": 0.29062870699881377,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.000499583924559136,
|
|
"loss": 5.8258,
|
|
"mean_token_accuracy": 0.15538810789585114,
|
|
"num_tokens": 6717027.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 6.075238943099976,
|
|
"epoch": 0.29112297350731514,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000499581776633479,
|
|
"loss": 5.8174,
|
|
"mean_token_accuracy": 0.1584622010588646,
|
|
"num_tokens": 6728107.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 5.981333208084107,
|
|
"epoch": 0.2916172400158165,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004995796231830756,
|
|
"loss": 5.8421,
|
|
"mean_token_accuracy": 0.15783749520778656,
|
|
"num_tokens": 6738628.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 5.984621667861939,
|
|
"epoch": 0.2921115065243179,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004995774642079786,
|
|
"loss": 5.7884,
|
|
"mean_token_accuracy": 0.1646315023303032,
|
|
"num_tokens": 6749416.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 5.985525369644165,
|
|
"epoch": 0.2926057730328193,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.000499575299708241,
|
|
"loss": 5.8929,
|
|
"mean_token_accuracy": 0.1488107793033123,
|
|
"num_tokens": 6761663.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 6.097906303405762,
|
|
"epoch": 0.2931000395413207,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004995731296839163,
|
|
"loss": 5.8395,
|
|
"mean_token_accuracy": 0.15889886170625686,
|
|
"num_tokens": 6773557.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 5.906320524215698,
|
|
"epoch": 0.29359430604982206,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004995709541350578,
|
|
"loss": 5.7589,
|
|
"mean_token_accuracy": 0.16536967307329178,
|
|
"num_tokens": 6782832.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 6.064802503585815,
|
|
"epoch": 0.29408857255832344,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.000499568773061719,
|
|
"loss": 5.882,
|
|
"mean_token_accuracy": 0.15133094489574433,
|
|
"num_tokens": 6795165.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 6.07623519897461,
|
|
"epoch": 0.2945828390668248,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004995665864639534,
|
|
"loss": 5.9023,
|
|
"mean_token_accuracy": 0.15480878055095673,
|
|
"num_tokens": 6807054.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 5.915165376663208,
|
|
"epoch": 0.29507710557532624,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004995643943418149,
|
|
"loss": 5.6952,
|
|
"mean_token_accuracy": 0.16479281187057496,
|
|
"num_tokens": 6817681.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 6.030046367645264,
|
|
"epoch": 0.2955713720838276,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004995621966953576,
|
|
"loss": 5.8628,
|
|
"mean_token_accuracy": 0.15873999148607254,
|
|
"num_tokens": 6829387.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 5.994547605514526,
|
|
"epoch": 0.296065638592329,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 0.0004995599935246355,
|
|
"loss": 5.8458,
|
|
"mean_token_accuracy": 0.15944462567567824,
|
|
"num_tokens": 6842023.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 6.006680917739868,
|
|
"epoch": 0.29655990510083036,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004995577848297026,
|
|
"loss": 5.8521,
|
|
"mean_token_accuracy": 0.1528011903166771,
|
|
"num_tokens": 6853189.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 0.29655990510083036,
|
|
"eval_entropy": 5.838898362913988,
|
|
"eval_loss": 5.886573314666748,
|
|
"eval_mean_token_accuracy": 0.16203675489657332,
|
|
"eval_num_tokens": 6853189.0,
|
|
"eval_runtime": 20.7287,
|
|
"eval_samples_per_second": 1568.503,
|
|
"eval_steps_per_second": 196.105,
|
|
"step": 3000
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 101150,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 10,
|
|
"save_steps": 3000,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 1590050377728000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|