6046 lines
164 KiB
JSON
6046 lines
164 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.27213352685050796,
|
|
"eval_steps": 3000,
|
|
"global_step": 3000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.69149408340454,
|
|
"epoch": 0.00045355587808418,
|
|
"grad_norm": 13.0625,
|
|
"learning_rate": 2e-06,
|
|
"loss": 10.7699,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 9661.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.691533756256103,
|
|
"epoch": 0.00090711175616836,
|
|
"grad_norm": 12.8125,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 10.7409,
|
|
"mean_token_accuracy": 0.00011574074160307646,
|
|
"num_tokens": 21028.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.691256523132324,
|
|
"epoch": 0.00136066763425254,
|
|
"grad_norm": 9.6875,
|
|
"learning_rate": 7e-06,
|
|
"loss": 10.4719,
|
|
"mean_token_accuracy": 0.019357142923399805,
|
|
"num_tokens": 31794.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.684057140350342,
|
|
"epoch": 0.00181422351233672,
|
|
"grad_norm": 5.8125,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 10.1776,
|
|
"mean_token_accuracy": 0.04135416336357593,
|
|
"num_tokens": 42892.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.636651706695556,
|
|
"epoch": 0.0022677793904208998,
|
|
"grad_norm": 4.125,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 9.8988,
|
|
"mean_token_accuracy": 0.04433486089110374,
|
|
"num_tokens": 53015.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.585999393463135,
|
|
"epoch": 0.00272133526850508,
|
|
"grad_norm": 3.15625,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 9.7643,
|
|
"mean_token_accuracy": 0.04109931848943234,
|
|
"num_tokens": 63334.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.606278038024902,
|
|
"epoch": 0.00317489114658926,
|
|
"grad_norm": 2.8125,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 9.6589,
|
|
"mean_token_accuracy": 0.04485522694885731,
|
|
"num_tokens": 74256.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.576828956604004,
|
|
"epoch": 0.00362844702467344,
|
|
"grad_norm": 2.734375,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 9.6182,
|
|
"mean_token_accuracy": 0.046111271157860755,
|
|
"num_tokens": 84699.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.563911628723144,
|
|
"epoch": 0.00408200290275762,
|
|
"grad_norm": 2.46875,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 9.581,
|
|
"mean_token_accuracy": 0.04264976643025875,
|
|
"num_tokens": 95178.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.59095401763916,
|
|
"epoch": 0.0045355587808417995,
|
|
"grad_norm": 2.3125,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 9.5471,
|
|
"mean_token_accuracy": 0.04867156893014908,
|
|
"num_tokens": 105438.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.587072944641113,
|
|
"epoch": 0.00498911465892598,
|
|
"grad_norm": 2.46875,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.4759,
|
|
"mean_token_accuracy": 0.0453902505338192,
|
|
"num_tokens": 115518.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.536315631866454,
|
|
"epoch": 0.00544267053701016,
|
|
"grad_norm": 2.4375,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.377,
|
|
"mean_token_accuracy": 0.053391102328896524,
|
|
"num_tokens": 125636.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.496469593048095,
|
|
"epoch": 0.00589622641509434,
|
|
"grad_norm": 2.515625,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.3108,
|
|
"mean_token_accuracy": 0.06191379800438881,
|
|
"num_tokens": 135990.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.413052368164063,
|
|
"epoch": 0.00634978229317852,
|
|
"grad_norm": 2.328125,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.2425,
|
|
"mean_token_accuracy": 0.0613830640912056,
|
|
"num_tokens": 146069.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.440055561065673,
|
|
"epoch": 0.006803338171262699,
|
|
"grad_norm": 2.265625,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.1547,
|
|
"mean_token_accuracy": 0.07091366611421109,
|
|
"num_tokens": 156335.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.438591575622558,
|
|
"epoch": 0.00725689404934688,
|
|
"grad_norm": 2.4375,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.0463,
|
|
"mean_token_accuracy": 0.06796186193823814,
|
|
"num_tokens": 166214.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.390478229522705,
|
|
"epoch": 0.007710449927431059,
|
|
"grad_norm": 3.015625,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 8.9608,
|
|
"mean_token_accuracy": 0.06832841411232948,
|
|
"num_tokens": 177010.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.332666969299316,
|
|
"epoch": 0.00816400580551524,
|
|
"grad_norm": 2.671875,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 8.7359,
|
|
"mean_token_accuracy": 0.07582402378320693,
|
|
"num_tokens": 187829.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.192521572113037,
|
|
"epoch": 0.00861756168359942,
|
|
"grad_norm": 2.609375,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 8.6972,
|
|
"mean_token_accuracy": 0.0755643829703331,
|
|
"num_tokens": 198107.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.20487937927246,
|
|
"epoch": 0.009071117561683599,
|
|
"grad_norm": 2.15625,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 8.5722,
|
|
"mean_token_accuracy": 0.07500871792435646,
|
|
"num_tokens": 208607.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.160844707489014,
|
|
"epoch": 0.009524673439767779,
|
|
"grad_norm": 2.296875,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 8.5622,
|
|
"mean_token_accuracy": 0.07453125491738319,
|
|
"num_tokens": 219899.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.124257946014405,
|
|
"epoch": 0.00997822931785196,
|
|
"grad_norm": 1.890625,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 8.4797,
|
|
"mean_token_accuracy": 0.07378747165203095,
|
|
"num_tokens": 230625.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 10.023534488677978,
|
|
"epoch": 0.01043178519593614,
|
|
"grad_norm": 2.03125,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 8.332,
|
|
"mean_token_accuracy": 0.08365851938724518,
|
|
"num_tokens": 241156.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 9.936892032623291,
|
|
"epoch": 0.01088534107402032,
|
|
"grad_norm": 1.8828125,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.1824,
|
|
"mean_token_accuracy": 0.08069921731948852,
|
|
"num_tokens": 252297.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 9.793073558807373,
|
|
"epoch": 0.011338896952104499,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.1575,
|
|
"mean_token_accuracy": 0.07586940303444863,
|
|
"num_tokens": 263188.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 9.714948558807373,
|
|
"epoch": 0.01179245283018868,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 7.9988,
|
|
"mean_token_accuracy": 0.07976540848612786,
|
|
"num_tokens": 274064.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 9.50658073425293,
|
|
"epoch": 0.01224600870827286,
|
|
"grad_norm": 1.90625,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 7.8083,
|
|
"mean_token_accuracy": 0.08558063134551049,
|
|
"num_tokens": 283736.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 9.259089279174805,
|
|
"epoch": 0.01269956458635704,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 7.7077,
|
|
"mean_token_accuracy": 0.08600766956806183,
|
|
"num_tokens": 294473.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 9.041689109802245,
|
|
"epoch": 0.013153120464441219,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 7.6574,
|
|
"mean_token_accuracy": 0.08418083190917969,
|
|
"num_tokens": 303849.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 8.879548072814941,
|
|
"epoch": 0.013606676342525399,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 7.5414,
|
|
"mean_token_accuracy": 0.08226052820682525,
|
|
"num_tokens": 314521.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 8.658054065704345,
|
|
"epoch": 0.01406023222060958,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 7.4964,
|
|
"mean_token_accuracy": 0.08540133312344551,
|
|
"num_tokens": 325210.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 8.465118217468262,
|
|
"epoch": 0.01451378809869376,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 7.4289,
|
|
"mean_token_accuracy": 0.0857605628669262,
|
|
"num_tokens": 335035.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 8.27260684967041,
|
|
"epoch": 0.014967343976777939,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.4279,
|
|
"mean_token_accuracy": 0.08624853491783142,
|
|
"num_tokens": 345402.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 8.192702579498292,
|
|
"epoch": 0.015420899854862119,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.3865,
|
|
"mean_token_accuracy": 0.0875001959502697,
|
|
"num_tokens": 356161.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 8.096983528137207,
|
|
"epoch": 0.0158744557329463,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.3087,
|
|
"mean_token_accuracy": 0.08894997537136078,
|
|
"num_tokens": 366695.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.000302648544311,
|
|
"epoch": 0.01632801161103048,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.2459,
|
|
"mean_token_accuracy": 0.08826502189040183,
|
|
"num_tokens": 376354.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 7.892614269256592,
|
|
"epoch": 0.01678156748911466,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.2853,
|
|
"mean_token_accuracy": 0.08936820179224014,
|
|
"num_tokens": 387468.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 7.8675158500671385,
|
|
"epoch": 0.01723512336719884,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.2366,
|
|
"mean_token_accuracy": 0.08572344705462456,
|
|
"num_tokens": 397658.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 7.800578165054321,
|
|
"epoch": 0.01768867924528302,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.2295,
|
|
"mean_token_accuracy": 0.08996806442737579,
|
|
"num_tokens": 408009.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 7.81451735496521,
|
|
"epoch": 0.018142235123367198,
|
|
"grad_norm": 1.8515625,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.1252,
|
|
"mean_token_accuracy": 0.09177512675523758,
|
|
"num_tokens": 417781.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 7.679977512359619,
|
|
"epoch": 0.018595791001451378,
|
|
"grad_norm": 1.7421875,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.3176,
|
|
"mean_token_accuracy": 0.08960338979959488,
|
|
"num_tokens": 428009.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 7.7656045913696286,
|
|
"epoch": 0.019049346879535557,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.1482,
|
|
"mean_token_accuracy": 0.0907668188214302,
|
|
"num_tokens": 438369.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 7.7085036277771,
|
|
"epoch": 0.01950290275761974,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.1341,
|
|
"mean_token_accuracy": 0.09216333702206611,
|
|
"num_tokens": 449047.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 7.61162919998169,
|
|
"epoch": 0.01995645863570392,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.1272,
|
|
"mean_token_accuracy": 0.093924118578434,
|
|
"num_tokens": 459992.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 7.708517932891846,
|
|
"epoch": 0.0204100145137881,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.207,
|
|
"mean_token_accuracy": 0.09182708933949471,
|
|
"num_tokens": 470542.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 7.576424551010132,
|
|
"epoch": 0.02086357039187228,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.1577,
|
|
"mean_token_accuracy": 0.09969761073589326,
|
|
"num_tokens": 479983.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 7.695700073242188,
|
|
"epoch": 0.02131712626995646,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.1299,
|
|
"mean_token_accuracy": 0.09077642261981964,
|
|
"num_tokens": 489827.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 7.514981985092163,
|
|
"epoch": 0.02177068214804064,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.1267,
|
|
"mean_token_accuracy": 0.08924711495637894,
|
|
"num_tokens": 500257.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.638169956207276,
|
|
"epoch": 0.022224238026124818,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.1507,
|
|
"mean_token_accuracy": 0.08717882186174393,
|
|
"num_tokens": 510227.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.506703805923462,
|
|
"epoch": 0.022677793904208998,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.0492,
|
|
"mean_token_accuracy": 0.0963970109820366,
|
|
"num_tokens": 520442.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 7.448557662963867,
|
|
"epoch": 0.023131349782293177,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.0689,
|
|
"mean_token_accuracy": 0.0880157247185707,
|
|
"num_tokens": 531075.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.53116626739502,
|
|
"epoch": 0.02358490566037736,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.1308,
|
|
"mean_token_accuracy": 0.08964502289891244,
|
|
"num_tokens": 542576.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.471751165390015,
|
|
"epoch": 0.02403846153846154,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.0485,
|
|
"mean_token_accuracy": 0.09242086634039878,
|
|
"num_tokens": 552380.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.469887447357178,
|
|
"epoch": 0.02449201741654572,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 7.0608,
|
|
"mean_token_accuracy": 0.08920432329177856,
|
|
"num_tokens": 563310.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.405267572402954,
|
|
"epoch": 0.0249455732946299,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 6.9664,
|
|
"mean_token_accuracy": 0.09448141977190971,
|
|
"num_tokens": 574113.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.440743923187256,
|
|
"epoch": 0.02539912917271408,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 6.9802,
|
|
"mean_token_accuracy": 0.09543166309595108,
|
|
"num_tokens": 585198.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.435253047943116,
|
|
"epoch": 0.025852685050798258,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 6.9734,
|
|
"mean_token_accuracy": 0.0946137085556984,
|
|
"num_tokens": 594837.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.340703868865967,
|
|
"epoch": 0.026306240928882438,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.0063,
|
|
"mean_token_accuracy": 0.09453380405902863,
|
|
"num_tokens": 604000.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.364510869979858,
|
|
"epoch": 0.026759796806966617,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.0165,
|
|
"mean_token_accuracy": 0.09476850554347038,
|
|
"num_tokens": 614831.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.43275899887085,
|
|
"epoch": 0.027213352685050797,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 7.0328,
|
|
"mean_token_accuracy": 0.09851854294538498,
|
|
"num_tokens": 625584.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.33774585723877,
|
|
"epoch": 0.027666908563134977,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000152,
|
|
"loss": 6.9117,
|
|
"mean_token_accuracy": 0.09698478281497955,
|
|
"num_tokens": 636282.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.42357873916626,
|
|
"epoch": 0.02812046444121916,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 7.0307,
|
|
"mean_token_accuracy": 0.09482022523880004,
|
|
"num_tokens": 646556.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.372399139404297,
|
|
"epoch": 0.02857402031930334,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000157,
|
|
"loss": 6.9899,
|
|
"mean_token_accuracy": 0.09617659002542496,
|
|
"num_tokens": 656983.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.231950092315674,
|
|
"epoch": 0.02902757619738752,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 6.9155,
|
|
"mean_token_accuracy": 0.1019688904285431,
|
|
"num_tokens": 668148.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.4007758617401125,
|
|
"epoch": 0.0294811320754717,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000162,
|
|
"loss": 7.0107,
|
|
"mean_token_accuracy": 0.09772700667381287,
|
|
"num_tokens": 677400.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.2482020378112795,
|
|
"epoch": 0.029934687953555878,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 6.896,
|
|
"mean_token_accuracy": 0.09065234810113906,
|
|
"num_tokens": 687519.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.264542865753174,
|
|
"epoch": 0.030388243831640058,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 6.8005,
|
|
"mean_token_accuracy": 0.10230840593576432,
|
|
"num_tokens": 698402.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.134629440307617,
|
|
"epoch": 0.030841799709724237,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 6.8627,
|
|
"mean_token_accuracy": 0.10249442160129547,
|
|
"num_tokens": 710628.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.336190032958984,
|
|
"epoch": 0.03129535558780842,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 6.9194,
|
|
"mean_token_accuracy": 0.09371023699641227,
|
|
"num_tokens": 720511.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.196484518051148,
|
|
"epoch": 0.0317489114658926,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 6.9486,
|
|
"mean_token_accuracy": 0.10238764733076096,
|
|
"num_tokens": 731357.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.2075560092926025,
|
|
"epoch": 0.032202467343976776,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.000177,
|
|
"loss": 6.8483,
|
|
"mean_token_accuracy": 0.1011496514081955,
|
|
"num_tokens": 741569.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.342852926254272,
|
|
"epoch": 0.03265602322206096,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 6.8818,
|
|
"mean_token_accuracy": 0.09690657183527947,
|
|
"num_tokens": 751194.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.139012861251831,
|
|
"epoch": 0.033109579100145135,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000182,
|
|
"loss": 6.8674,
|
|
"mean_token_accuracy": 0.0964200034737587,
|
|
"num_tokens": 762554.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.168058109283447,
|
|
"epoch": 0.03356313497822932,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 6.7947,
|
|
"mean_token_accuracy": 0.1081994466483593,
|
|
"num_tokens": 772819.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.116481113433838,
|
|
"epoch": 0.034016690856313495,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000187,
|
|
"loss": 6.8457,
|
|
"mean_token_accuracy": 0.09905849322676659,
|
|
"num_tokens": 782717.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.275808191299438,
|
|
"epoch": 0.03447024673439768,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 6.9231,
|
|
"mean_token_accuracy": 0.10113033130764962,
|
|
"num_tokens": 792523.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.136093997955323,
|
|
"epoch": 0.03492380261248186,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000192,
|
|
"loss": 6.852,
|
|
"mean_token_accuracy": 0.10536141693592072,
|
|
"num_tokens": 802696.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.199086666107178,
|
|
"epoch": 0.03537735849056604,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 6.8945,
|
|
"mean_token_accuracy": 0.09893356338143348,
|
|
"num_tokens": 813041.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.261086368560791,
|
|
"epoch": 0.03583091436865022,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 6.8549,
|
|
"mean_token_accuracy": 0.09995201453566552,
|
|
"num_tokens": 823366.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.149146509170532,
|
|
"epoch": 0.036284470246734396,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 6.8201,
|
|
"mean_token_accuracy": 0.100285005569458,
|
|
"num_tokens": 835455.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.098216772079468,
|
|
"epoch": 0.03673802612481858,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000202,
|
|
"loss": 6.8581,
|
|
"mean_token_accuracy": 0.10088052004575729,
|
|
"num_tokens": 846190.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.215963459014892,
|
|
"epoch": 0.037191582002902755,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 6.8161,
|
|
"mean_token_accuracy": 0.10391620472073555,
|
|
"num_tokens": 856303.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.0431232929229735,
|
|
"epoch": 0.03764513788098694,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000207,
|
|
"loss": 6.7939,
|
|
"mean_token_accuracy": 0.09926707297563553,
|
|
"num_tokens": 866240.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.138209819793701,
|
|
"epoch": 0.038098693759071114,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 6.9358,
|
|
"mean_token_accuracy": 0.09860606789588929,
|
|
"num_tokens": 877035.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.092493534088135,
|
|
"epoch": 0.0385522496371553,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.000212,
|
|
"loss": 6.8158,
|
|
"mean_token_accuracy": 0.09759353399276734,
|
|
"num_tokens": 887344.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.141839790344238,
|
|
"epoch": 0.03900580551523948,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 6.8314,
|
|
"mean_token_accuracy": 0.10464526489377021,
|
|
"num_tokens": 897271.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.065910148620605,
|
|
"epoch": 0.03945936139332366,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 6.7285,
|
|
"mean_token_accuracy": 0.1087069682776928,
|
|
"num_tokens": 908700.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 6.984343385696411,
|
|
"epoch": 0.03991291727140784,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 6.7459,
|
|
"mean_token_accuracy": 0.10460741296410561,
|
|
"num_tokens": 918980.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.071007776260376,
|
|
"epoch": 0.040366473149492016,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000222,
|
|
"loss": 6.7444,
|
|
"mean_token_accuracy": 0.10659246519207954,
|
|
"num_tokens": 928616.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.033933305740357,
|
|
"epoch": 0.0408200290275762,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 6.897,
|
|
"mean_token_accuracy": 0.10052595734596252,
|
|
"num_tokens": 939066.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.086242055892944,
|
|
"epoch": 0.041273584905660375,
|
|
"grad_norm": 1.765625,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 6.7852,
|
|
"mean_token_accuracy": 0.1003576584160328,
|
|
"num_tokens": 949160.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.0361896514892575,
|
|
"epoch": 0.04172714078374456,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 6.8063,
|
|
"mean_token_accuracy": 0.10417447835206986,
|
|
"num_tokens": 959959.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.061341714859009,
|
|
"epoch": 0.042180696661828734,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 6.8092,
|
|
"mean_token_accuracy": 0.10948210060596467,
|
|
"num_tokens": 969752.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 6.854967498779297,
|
|
"epoch": 0.04263425253991292,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.8149,
|
|
"mean_token_accuracy": 0.10138045027852058,
|
|
"num_tokens": 979917.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.039763593673706,
|
|
"epoch": 0.0430878084179971,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.8134,
|
|
"mean_token_accuracy": 0.09902644529938698,
|
|
"num_tokens": 990613.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.048016119003296,
|
|
"epoch": 0.04354136429608128,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.8134,
|
|
"mean_token_accuracy": 0.10063022077083587,
|
|
"num_tokens": 1000753.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 6.978139591217041,
|
|
"epoch": 0.04399492017416546,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.7445,
|
|
"mean_token_accuracy": 0.10282852351665497,
|
|
"num_tokens": 1012141.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.031801795959472,
|
|
"epoch": 0.044448476052249636,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.7459,
|
|
"mean_token_accuracy": 0.10179351791739463,
|
|
"num_tokens": 1023322.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 6.850466966629028,
|
|
"epoch": 0.04490203193033382,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.6463,
|
|
"mean_token_accuracy": 0.09672334268689156,
|
|
"num_tokens": 1033949.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 6.988512468338013,
|
|
"epoch": 0.045355587808417995,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 6.7366,
|
|
"mean_token_accuracy": 0.10716779977083206,
|
|
"num_tokens": 1043580.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 6.902674722671509,
|
|
"epoch": 0.04580914368650218,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.656,
|
|
"mean_token_accuracy": 0.11081461235880852,
|
|
"num_tokens": 1052106.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 6.90415506362915,
|
|
"epoch": 0.046262699564586354,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.6971,
|
|
"mean_token_accuracy": 0.1064011238515377,
|
|
"num_tokens": 1061764.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 6.838364171981811,
|
|
"epoch": 0.04671625544267054,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.5697,
|
|
"mean_token_accuracy": 0.10854723304510117,
|
|
"num_tokens": 1071164.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 6.968482303619385,
|
|
"epoch": 0.04716981132075472,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.8149,
|
|
"mean_token_accuracy": 0.10352605357766151,
|
|
"num_tokens": 1082602.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 6.9721253395080565,
|
|
"epoch": 0.047623367198838897,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.7378,
|
|
"mean_token_accuracy": 0.1052236057817936,
|
|
"num_tokens": 1091871.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 6.919319772720337,
|
|
"epoch": 0.04807692307692308,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.7272,
|
|
"mean_token_accuracy": 0.10676647424697876,
|
|
"num_tokens": 1102568.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 6.941032934188843,
|
|
"epoch": 0.048530478955007256,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.6203,
|
|
"mean_token_accuracy": 0.10674264281988144,
|
|
"num_tokens": 1112777.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 6.716324281692505,
|
|
"epoch": 0.04898403483309144,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.6119,
|
|
"mean_token_accuracy": 0.10510820895433426,
|
|
"num_tokens": 1123809.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 6.8692456722259525,
|
|
"epoch": 0.049437590711175615,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.6644,
|
|
"mean_token_accuracy": 0.10762705281376839,
|
|
"num_tokens": 1133793.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 6.804552936553955,
|
|
"epoch": 0.0498911465892598,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.5543,
|
|
"mean_token_accuracy": 0.10652651414275169,
|
|
"num_tokens": 1144478.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 6.81122727394104,
|
|
"epoch": 0.050344702467343974,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.595,
|
|
"mean_token_accuracy": 0.10836011320352554,
|
|
"num_tokens": 1155518.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 6.848041963577271,
|
|
"epoch": 0.05079825834542816,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.6803,
|
|
"mean_token_accuracy": 0.10240246802568435,
|
|
"num_tokens": 1166494.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 6.864998054504395,
|
|
"epoch": 0.05125181422351233,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.6603,
|
|
"mean_token_accuracy": 0.10499662458896637,
|
|
"num_tokens": 1177052.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 6.7566328048706055,
|
|
"epoch": 0.051705370101596516,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.588,
|
|
"mean_token_accuracy": 0.1072057232260704,
|
|
"num_tokens": 1187382.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 6.84862585067749,
|
|
"epoch": 0.0521589259796807,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.6371,
|
|
"mean_token_accuracy": 0.10280174314975739,
|
|
"num_tokens": 1197535.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 6.854958152770996,
|
|
"epoch": 0.052612481857764876,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.6932,
|
|
"mean_token_accuracy": 0.11012415438890458,
|
|
"num_tokens": 1207626.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 6.8038904666900635,
|
|
"epoch": 0.05306603773584906,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.5812,
|
|
"mean_token_accuracy": 0.10877829939126968,
|
|
"num_tokens": 1218433.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 6.773093318939209,
|
|
"epoch": 0.053519593613933235,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.5272,
|
|
"mean_token_accuracy": 0.10834594070911407,
|
|
"num_tokens": 1228698.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 6.839606189727784,
|
|
"epoch": 0.05397314949201742,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.6608,
|
|
"mean_token_accuracy": 0.11245888248085975,
|
|
"num_tokens": 1239278.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 6.663392496109009,
|
|
"epoch": 0.054426705370101594,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.6276,
|
|
"mean_token_accuracy": 0.11012510806322098,
|
|
"num_tokens": 1248976.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 6.770804977416992,
|
|
"epoch": 0.05488026124818578,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.5805,
|
|
"mean_token_accuracy": 0.11286153569817543,
|
|
"num_tokens": 1259545.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 6.810527324676514,
|
|
"epoch": 0.05533381712626995,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.6029,
|
|
"mean_token_accuracy": 0.10805755332112313,
|
|
"num_tokens": 1269444.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 6.8169921875,
|
|
"epoch": 0.055787373004354136,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.6931,
|
|
"mean_token_accuracy": 0.10932271853089333,
|
|
"num_tokens": 1279821.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 6.654322671890259,
|
|
"epoch": 0.05624092888243832,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.5238,
|
|
"mean_token_accuracy": 0.10867622718214989,
|
|
"num_tokens": 1290398.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 6.814783906936645,
|
|
"epoch": 0.056694484760522496,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.6259,
|
|
"mean_token_accuracy": 0.10474204868078232,
|
|
"num_tokens": 1300920.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 6.690603113174438,
|
|
"epoch": 0.05714804063860668,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.6062,
|
|
"mean_token_accuracy": 0.10462497025728226,
|
|
"num_tokens": 1311928.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 6.83460955619812,
|
|
"epoch": 0.057601596516690855,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.6113,
|
|
"mean_token_accuracy": 0.10693813562393188,
|
|
"num_tokens": 1322554.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 6.699485635757446,
|
|
"epoch": 0.05805515239477504,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.5474,
|
|
"mean_token_accuracy": 0.11296007037162781,
|
|
"num_tokens": 1333195.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 6.70126633644104,
|
|
"epoch": 0.058508708272859214,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.6557,
|
|
"mean_token_accuracy": 0.10681698620319366,
|
|
"num_tokens": 1343655.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 6.784696960449219,
|
|
"epoch": 0.0589622641509434,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.5197,
|
|
"mean_token_accuracy": 0.11125476583838463,
|
|
"num_tokens": 1352528.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 6.562724494934082,
|
|
"epoch": 0.05941582002902757,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.5629,
|
|
"mean_token_accuracy": 0.11251123249530792,
|
|
"num_tokens": 1363143.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 6.857563257217407,
|
|
"epoch": 0.059869375907111756,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.578,
|
|
"mean_token_accuracy": 0.1062569722533226,
|
|
"num_tokens": 1373049.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 6.634031963348389,
|
|
"epoch": 0.06032293178519594,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.5634,
|
|
"mean_token_accuracy": 0.11504042670130729,
|
|
"num_tokens": 1384040.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 6.730848979949951,
|
|
"epoch": 0.060776487663280115,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.6586,
|
|
"mean_token_accuracy": 0.11258623376488686,
|
|
"num_tokens": 1394511.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 6.641570663452148,
|
|
"epoch": 0.0612300435413643,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.4682,
|
|
"mean_token_accuracy": 0.11615501791238785,
|
|
"num_tokens": 1405381.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 6.756093549728393,
|
|
"epoch": 0.061683599419448475,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.6404,
|
|
"mean_token_accuracy": 0.11007921248674393,
|
|
"num_tokens": 1415840.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 6.7638633251190186,
|
|
"epoch": 0.06213715529753266,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.6825,
|
|
"mean_token_accuracy": 0.10886021852493286,
|
|
"num_tokens": 1426634.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 6.609492397308349,
|
|
"epoch": 0.06259071117561683,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.5622,
|
|
"mean_token_accuracy": 0.11168285235762596,
|
|
"num_tokens": 1436794.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 6.853395700454712,
|
|
"epoch": 0.06304426705370102,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.6395,
|
|
"mean_token_accuracy": 0.11344821155071258,
|
|
"num_tokens": 1447114.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 6.582093334197998,
|
|
"epoch": 0.0634978229317852,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.4899,
|
|
"mean_token_accuracy": 0.11673147901892662,
|
|
"num_tokens": 1456494.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 6.599524879455567,
|
|
"epoch": 0.06395137880986937,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.4935,
|
|
"mean_token_accuracy": 0.11297031342983246,
|
|
"num_tokens": 1466580.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 6.699067687988281,
|
|
"epoch": 0.06440493468795355,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.5999,
|
|
"mean_token_accuracy": 0.10977515503764153,
|
|
"num_tokens": 1476854.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 6.683993911743164,
|
|
"epoch": 0.06485849056603774,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.5426,
|
|
"mean_token_accuracy": 0.10898664593696594,
|
|
"num_tokens": 1487133.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 6.6332762241363525,
|
|
"epoch": 0.06531204644412192,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.563,
|
|
"mean_token_accuracy": 0.11930395513772965,
|
|
"num_tokens": 1496974.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 6.615490865707398,
|
|
"epoch": 0.0657656023222061,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.5058,
|
|
"mean_token_accuracy": 0.11035025268793106,
|
|
"num_tokens": 1507665.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 6.698334789276123,
|
|
"epoch": 0.06621915820029027,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.4698,
|
|
"mean_token_accuracy": 0.11344846114516258,
|
|
"num_tokens": 1518625.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 6.690861082077026,
|
|
"epoch": 0.06667271407837445,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.608,
|
|
"mean_token_accuracy": 0.10640608072280884,
|
|
"num_tokens": 1529184.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.481381893157959,
|
|
"epoch": 0.06712626995645864,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.526,
|
|
"mean_token_accuracy": 0.10988623276352882,
|
|
"num_tokens": 1540130.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 6.694868326187134,
|
|
"epoch": 0.06757982583454282,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.5181,
|
|
"mean_token_accuracy": 0.11159003973007202,
|
|
"num_tokens": 1550161.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.574037361145019,
|
|
"epoch": 0.06803338171262699,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.4826,
|
|
"mean_token_accuracy": 0.11772774383425713,
|
|
"num_tokens": 1560076.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 6.67036771774292,
|
|
"epoch": 0.06848693759071117,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.4693,
|
|
"mean_token_accuracy": 0.11172557696700096,
|
|
"num_tokens": 1570882.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 6.481294822692871,
|
|
"epoch": 0.06894049346879536,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.4696,
|
|
"mean_token_accuracy": 0.11403056010603904,
|
|
"num_tokens": 1582227.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.62347674369812,
|
|
"epoch": 0.06939404934687954,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.5213,
|
|
"mean_token_accuracy": 0.11278930827975273,
|
|
"num_tokens": 1592691.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.617785978317261,
|
|
"epoch": 0.06984760522496372,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.4965,
|
|
"mean_token_accuracy": 0.11648303642868996,
|
|
"num_tokens": 1602258.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.573126649856567,
|
|
"epoch": 0.07030116110304789,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.5672,
|
|
"mean_token_accuracy": 0.11526603251695633,
|
|
"num_tokens": 1611953.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.6316383361816404,
|
|
"epoch": 0.07075471698113207,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.5492,
|
|
"mean_token_accuracy": 0.11260883659124374,
|
|
"num_tokens": 1621944.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.496122121810913,
|
|
"epoch": 0.07120827285921626,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.4659,
|
|
"mean_token_accuracy": 0.11800124645233154,
|
|
"num_tokens": 1632702.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.6150144100189205,
|
|
"epoch": 0.07166182873730044,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.5214,
|
|
"mean_token_accuracy": 0.11103227436542511,
|
|
"num_tokens": 1642622.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.574394226074219,
|
|
"epoch": 0.07211538461538461,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.4592,
|
|
"mean_token_accuracy": 0.11586317420005798,
|
|
"num_tokens": 1652442.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.473141813278199,
|
|
"epoch": 0.07256894049346879,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.4886,
|
|
"mean_token_accuracy": 0.11505223363637924,
|
|
"num_tokens": 1663022.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.656652927398682,
|
|
"epoch": 0.07302249637155298,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.5181,
|
|
"mean_token_accuracy": 0.11541026830673218,
|
|
"num_tokens": 1673825.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.590756893157959,
|
|
"epoch": 0.07347605224963716,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.4718,
|
|
"mean_token_accuracy": 0.11257318034768105,
|
|
"num_tokens": 1683783.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.504452848434449,
|
|
"epoch": 0.07392960812772134,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.5008,
|
|
"mean_token_accuracy": 0.1076662614941597,
|
|
"num_tokens": 1694456.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.454553031921387,
|
|
"epoch": 0.07438316400580551,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.5053,
|
|
"mean_token_accuracy": 0.11051234900951386,
|
|
"num_tokens": 1706108.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.532086324691773,
|
|
"epoch": 0.0748367198838897,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.3786,
|
|
"mean_token_accuracy": 0.1280963383615017,
|
|
"num_tokens": 1716273.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.5029052734375,
|
|
"epoch": 0.07529027576197388,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.4237,
|
|
"mean_token_accuracy": 0.11734267845749854,
|
|
"num_tokens": 1727271.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.529041481018067,
|
|
"epoch": 0.07574383164005806,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.4755,
|
|
"mean_token_accuracy": 0.1184304565191269,
|
|
"num_tokens": 1737198.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.576714324951172,
|
|
"epoch": 0.07619738751814223,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.5197,
|
|
"mean_token_accuracy": 0.11203703507781029,
|
|
"num_tokens": 1747077.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.594869232177734,
|
|
"epoch": 0.07665094339622641,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.5119,
|
|
"mean_token_accuracy": 0.1193366564810276,
|
|
"num_tokens": 1757626.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.527251434326172,
|
|
"epoch": 0.0771044992743106,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.5454,
|
|
"mean_token_accuracy": 0.11205081716179847,
|
|
"num_tokens": 1768431.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.592037868499756,
|
|
"epoch": 0.07755805515239478,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.4819,
|
|
"mean_token_accuracy": 0.11357785761356354,
|
|
"num_tokens": 1779071.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.481006050109864,
|
|
"epoch": 0.07801161103047896,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.504,
|
|
"mean_token_accuracy": 0.11291966959834099,
|
|
"num_tokens": 1788905.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.435082483291626,
|
|
"epoch": 0.07846516690856313,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.3973,
|
|
"mean_token_accuracy": 0.12037285566329955,
|
|
"num_tokens": 1799003.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.53296480178833,
|
|
"epoch": 0.07891872278664731,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.436,
|
|
"mean_token_accuracy": 0.11971265748143196,
|
|
"num_tokens": 1809262.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.4183008670806885,
|
|
"epoch": 0.0793722786647315,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.4654,
|
|
"mean_token_accuracy": 0.11073716729879379,
|
|
"num_tokens": 1821059.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.492483615875244,
|
|
"epoch": 0.07982583454281568,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.3791,
|
|
"mean_token_accuracy": 0.11755654066801072,
|
|
"num_tokens": 1830246.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.335828971862793,
|
|
"epoch": 0.08027939042089985,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.408,
|
|
"mean_token_accuracy": 0.12151500731706619,
|
|
"num_tokens": 1839950.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.464663171768189,
|
|
"epoch": 0.08073294629898403,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.3876,
|
|
"mean_token_accuracy": 0.11592675521969795,
|
|
"num_tokens": 1852026.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.560555934906006,
|
|
"epoch": 0.08118650217706821,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.4635,
|
|
"mean_token_accuracy": 0.11812743693590164,
|
|
"num_tokens": 1861672.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.551700782775879,
|
|
"epoch": 0.0816400580551524,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.4836,
|
|
"mean_token_accuracy": 0.11402211338281631,
|
|
"num_tokens": 1872294.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.45425329208374,
|
|
"epoch": 0.08209361393323658,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.4414,
|
|
"mean_token_accuracy": 0.11440050825476647,
|
|
"num_tokens": 1883049.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.345271444320678,
|
|
"epoch": 0.08254716981132075,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.3992,
|
|
"mean_token_accuracy": 0.12349174469709397,
|
|
"num_tokens": 1893026.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.415435361862182,
|
|
"epoch": 0.08300072568940493,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.4304,
|
|
"mean_token_accuracy": 0.11467023715376853,
|
|
"num_tokens": 1903987.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.59291787147522,
|
|
"epoch": 0.08345428156748912,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.4365,
|
|
"mean_token_accuracy": 0.11061537489295006,
|
|
"num_tokens": 1914285.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.461905241012573,
|
|
"epoch": 0.0839078374455733,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.4536,
|
|
"mean_token_accuracy": 0.1159026212990284,
|
|
"num_tokens": 1924841.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.357869672775268,
|
|
"epoch": 0.08436139332365747,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.3829,
|
|
"mean_token_accuracy": 0.11473607048392295,
|
|
"num_tokens": 1935536.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.538698625564575,
|
|
"epoch": 0.08481494920174165,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.4852,
|
|
"mean_token_accuracy": 0.11677240207791328,
|
|
"num_tokens": 1946179.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.407983589172363,
|
|
"epoch": 0.08526850507982583,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.383,
|
|
"mean_token_accuracy": 0.11190493553876876,
|
|
"num_tokens": 1956139.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.46036171913147,
|
|
"epoch": 0.08572206095791002,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.4659,
|
|
"mean_token_accuracy": 0.11674941927194596,
|
|
"num_tokens": 1966944.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.424592447280884,
|
|
"epoch": 0.0861756168359942,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.4265,
|
|
"mean_token_accuracy": 0.1166701041162014,
|
|
"num_tokens": 1976910.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.57973198890686,
|
|
"epoch": 0.08662917271407837,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.4712,
|
|
"mean_token_accuracy": 0.11671740487217903,
|
|
"num_tokens": 1987111.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.31265082359314,
|
|
"epoch": 0.08708272859216255,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.3582,
|
|
"mean_token_accuracy": 0.12392476126551628,
|
|
"num_tokens": 1996962.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.368039894104004,
|
|
"epoch": 0.08753628447024674,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.2983,
|
|
"mean_token_accuracy": 0.11986770704388619,
|
|
"num_tokens": 2006775.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.482371377944946,
|
|
"epoch": 0.08798984034833092,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.4064,
|
|
"mean_token_accuracy": 0.12042228356003762,
|
|
"num_tokens": 2016453.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.435760164260865,
|
|
"epoch": 0.08844339622641509,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.4685,
|
|
"mean_token_accuracy": 0.11267957463860512,
|
|
"num_tokens": 2029100.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.409471607208252,
|
|
"epoch": 0.08889695210449927,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.4124,
|
|
"mean_token_accuracy": 0.11172624826431274,
|
|
"num_tokens": 2040041.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.419856071472168,
|
|
"epoch": 0.08935050798258345,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.3962,
|
|
"mean_token_accuracy": 0.11573249772191048,
|
|
"num_tokens": 2050377.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.359882593154907,
|
|
"epoch": 0.08980406386066764,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.3743,
|
|
"mean_token_accuracy": 0.1218277484178543,
|
|
"num_tokens": 2060081.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.496881818771362,
|
|
"epoch": 0.09025761973875182,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.5431,
|
|
"mean_token_accuracy": 0.1156518079340458,
|
|
"num_tokens": 2070335.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.389731550216675,
|
|
"epoch": 0.09071117561683599,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.3597,
|
|
"mean_token_accuracy": 0.1212635301053524,
|
|
"num_tokens": 2079259.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.374619245529175,
|
|
"epoch": 0.09116473149492017,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999999985110217,
|
|
"loss": 6.4764,
|
|
"mean_token_accuracy": 0.11514603495597839,
|
|
"num_tokens": 2089256.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.3883942604064945,
|
|
"epoch": 0.09161828737300436,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004999999924620471,
|
|
"loss": 6.4064,
|
|
"mean_token_accuracy": 0.11673571988940239,
|
|
"num_tokens": 2099141.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.387277984619141,
|
|
"epoch": 0.09207184325108854,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004999999817600155,
|
|
"loss": 6.3246,
|
|
"mean_token_accuracy": 0.12680575177073478,
|
|
"num_tokens": 2108934.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.33391375541687,
|
|
"epoch": 0.09252539912917271,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004999999664049268,
|
|
"loss": 6.4441,
|
|
"mean_token_accuracy": 0.1187924824655056,
|
|
"num_tokens": 2120302.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.453229522705078,
|
|
"epoch": 0.09297895500725689,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004999999463967816,
|
|
"loss": 6.4247,
|
|
"mean_token_accuracy": 0.1195537157356739,
|
|
"num_tokens": 2131096.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.416885185241699,
|
|
"epoch": 0.09343251088534107,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00049999992173558,
|
|
"loss": 6.4041,
|
|
"mean_token_accuracy": 0.12701962441205977,
|
|
"num_tokens": 2140603.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.377317237854004,
|
|
"epoch": 0.09388606676342526,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999998924213228,
|
|
"loss": 6.4604,
|
|
"mean_token_accuracy": 0.11771498769521713,
|
|
"num_tokens": 2151150.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.381362867355347,
|
|
"epoch": 0.09433962264150944,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004999998584540105,
|
|
"loss": 6.3849,
|
|
"mean_token_accuracy": 0.11927906274795533,
|
|
"num_tokens": 2161237.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.435468864440918,
|
|
"epoch": 0.09479317851959361,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999998198336437,
|
|
"loss": 6.39,
|
|
"mean_token_accuracy": 0.11034503877162934,
|
|
"num_tokens": 2171861.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.281571960449218,
|
|
"epoch": 0.09524673439767779,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004999997765602233,
|
|
"loss": 6.3031,
|
|
"mean_token_accuracy": 0.12425279021263122,
|
|
"num_tokens": 2181415.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.323150634765625,
|
|
"epoch": 0.09570029027576198,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999997286337502,
|
|
"loss": 6.3244,
|
|
"mean_token_accuracy": 0.1208683580160141,
|
|
"num_tokens": 2192036.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.358468723297119,
|
|
"epoch": 0.09615384615384616,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999996760542254,
|
|
"loss": 6.4465,
|
|
"mean_token_accuracy": 0.1163475088775158,
|
|
"num_tokens": 2203001.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.384481906890869,
|
|
"epoch": 0.09660740203193033,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00049999961882165,
|
|
"loss": 6.2868,
|
|
"mean_token_accuracy": 0.1244203880429268,
|
|
"num_tokens": 2213310.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.250124216079712,
|
|
"epoch": 0.09706095791001451,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004999995569360251,
|
|
"loss": 6.2546,
|
|
"mean_token_accuracy": 0.12180497273802757,
|
|
"num_tokens": 2223909.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.361983823776245,
|
|
"epoch": 0.0975145137880987,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000499999490397352,
|
|
"loss": 6.3172,
|
|
"mean_token_accuracy": 0.11828529611229896,
|
|
"num_tokens": 2234547.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.273873519897461,
|
|
"epoch": 0.09796806966618288,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004999994192056321,
|
|
"loss": 6.3341,
|
|
"mean_token_accuracy": 0.12068448886275292,
|
|
"num_tokens": 2244493.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.257021093368531,
|
|
"epoch": 0.09842162554426705,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004999993433608669,
|
|
"loss": 6.236,
|
|
"mean_token_accuracy": 0.1235893614590168,
|
|
"num_tokens": 2254881.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.276112365722656,
|
|
"epoch": 0.09887518142235123,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004999992628630579,
|
|
"loss": 6.3323,
|
|
"mean_token_accuracy": 0.11811816319823265,
|
|
"num_tokens": 2264964.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.486545896530151,
|
|
"epoch": 0.09932873730043541,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999991777122069,
|
|
"loss": 6.4318,
|
|
"mean_token_accuracy": 0.11846130490303039,
|
|
"num_tokens": 2275507.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.27265567779541,
|
|
"epoch": 0.0997822931785196,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004999990879083156,
|
|
"loss": 6.2819,
|
|
"mean_token_accuracy": 0.12145883291959762,
|
|
"num_tokens": 2284405.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.259899759292603,
|
|
"epoch": 0.10023584905660378,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999989934513857,
|
|
"loss": 6.324,
|
|
"mean_token_accuracy": 0.1184744507074356,
|
|
"num_tokens": 2294082.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.317243003845215,
|
|
"epoch": 0.10068940493468795,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999988943414193,
|
|
"loss": 6.3617,
|
|
"mean_token_accuracy": 0.11865237578749657,
|
|
"num_tokens": 2303805.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.3640618324279785,
|
|
"epoch": 0.10114296081277213,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999987905784184,
|
|
"loss": 6.2915,
|
|
"mean_token_accuracy": 0.11877703964710236,
|
|
"num_tokens": 2314974.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.386012029647827,
|
|
"epoch": 0.10159651669085631,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999986821623853,
|
|
"loss": 6.4138,
|
|
"mean_token_accuracy": 0.1207954816520214,
|
|
"num_tokens": 2326719.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.265592050552368,
|
|
"epoch": 0.1020500725689405,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004999985690933219,
|
|
"loss": 6.3131,
|
|
"mean_token_accuracy": 0.11659505292773246,
|
|
"num_tokens": 2337379.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.320196676254272,
|
|
"epoch": 0.10250362844702467,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999984513712311,
|
|
"loss": 6.2983,
|
|
"mean_token_accuracy": 0.12314697355031967,
|
|
"num_tokens": 2347509.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.3434178829193115,
|
|
"epoch": 0.10295718432510885,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999983289961146,
|
|
"loss": 6.2882,
|
|
"mean_token_accuracy": 0.11846850141882896,
|
|
"num_tokens": 2357835.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.269227457046509,
|
|
"epoch": 0.10341074020319303,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004999982019679755,
|
|
"loss": 6.3579,
|
|
"mean_token_accuracy": 0.12629576474428178,
|
|
"num_tokens": 2368373.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.270691156387329,
|
|
"epoch": 0.10386429608127722,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004999980702868164,
|
|
"loss": 6.3287,
|
|
"mean_token_accuracy": 0.12598486691713334,
|
|
"num_tokens": 2379170.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.256811571121216,
|
|
"epoch": 0.1043178519593614,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004999979339526396,
|
|
"loss": 6.2471,
|
|
"mean_token_accuracy": 0.12656034380197526,
|
|
"num_tokens": 2389711.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.2689375400543215,
|
|
"epoch": 0.10477140783744557,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999977929654484,
|
|
"loss": 6.2285,
|
|
"mean_token_accuracy": 0.13038562908768653,
|
|
"num_tokens": 2399710.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.235695457458496,
|
|
"epoch": 0.10522496371552975,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004999976473252453,
|
|
"loss": 6.3514,
|
|
"mean_token_accuracy": 0.11452672109007836,
|
|
"num_tokens": 2410567.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.344023704528809,
|
|
"epoch": 0.10567851959361393,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999974970320338,
|
|
"loss": 6.307,
|
|
"mean_token_accuracy": 0.12359435632824897,
|
|
"num_tokens": 2420010.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.309238719940185,
|
|
"epoch": 0.10613207547169812,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999973420858165,
|
|
"loss": 6.2661,
|
|
"mean_token_accuracy": 0.1272447846829891,
|
|
"num_tokens": 2429706.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.296723413467407,
|
|
"epoch": 0.10658563134978229,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004999971824865968,
|
|
"loss": 6.4111,
|
|
"mean_token_accuracy": 0.11686691716313362,
|
|
"num_tokens": 2441482.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.422159862518311,
|
|
"epoch": 0.10703918722786647,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999970182343782,
|
|
"loss": 6.3682,
|
|
"mean_token_accuracy": 0.11550676226615905,
|
|
"num_tokens": 2452094.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.148933982849121,
|
|
"epoch": 0.10749274310595065,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999968493291638,
|
|
"loss": 6.1916,
|
|
"mean_token_accuracy": 0.12536179572343825,
|
|
"num_tokens": 2462174.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.300602436065674,
|
|
"epoch": 0.10794629898403484,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999966757709573,
|
|
"loss": 6.2581,
|
|
"mean_token_accuracy": 0.1321061834692955,
|
|
"num_tokens": 2473294.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.228463363647461,
|
|
"epoch": 0.10839985486211902,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004999964975597622,
|
|
"loss": 6.3172,
|
|
"mean_token_accuracy": 0.1278726689517498,
|
|
"num_tokens": 2482954.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.31824312210083,
|
|
"epoch": 0.10885341074020319,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999963146955821,
|
|
"loss": 6.3121,
|
|
"mean_token_accuracy": 0.12523070126771926,
|
|
"num_tokens": 2493894.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.202036809921265,
|
|
"epoch": 0.10930696661828737,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000499996127178421,
|
|
"loss": 6.2219,
|
|
"mean_token_accuracy": 0.12660781443119049,
|
|
"num_tokens": 2503983.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.2187714099884035,
|
|
"epoch": 0.10976052249637155,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004999959350082825,
|
|
"loss": 6.1808,
|
|
"mean_token_accuracy": 0.12330685779452324,
|
|
"num_tokens": 2513721.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.269692087173462,
|
|
"epoch": 0.11021407837445574,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999957381851709,
|
|
"loss": 6.2502,
|
|
"mean_token_accuracy": 0.12575069963932037,
|
|
"num_tokens": 2523312.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.2597119331359865,
|
|
"epoch": 0.1106676342525399,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999955367090901,
|
|
"loss": 6.1965,
|
|
"mean_token_accuracy": 0.12818494513630868,
|
|
"num_tokens": 2532685.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.278915119171143,
|
|
"epoch": 0.11112119013062409,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004999953305800441,
|
|
"loss": 6.3135,
|
|
"mean_token_accuracy": 0.12152358293533325,
|
|
"num_tokens": 2543324.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.2923393726348875,
|
|
"epoch": 0.11157474600870827,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999951197980374,
|
|
"loss": 6.2963,
|
|
"mean_token_accuracy": 0.12113744467496872,
|
|
"num_tokens": 2554242.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.273481845855713,
|
|
"epoch": 0.11202830188679246,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999949043630744,
|
|
"loss": 6.2467,
|
|
"mean_token_accuracy": 0.1278710223734379,
|
|
"num_tokens": 2563372.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.234022188186645,
|
|
"epoch": 0.11248185776487664,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999946842751593,
|
|
"loss": 6.3483,
|
|
"mean_token_accuracy": 0.12514663413167,
|
|
"num_tokens": 2574029.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.359460401535034,
|
|
"epoch": 0.11293541364296081,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004999944595342968,
|
|
"loss": 6.3691,
|
|
"mean_token_accuracy": 0.1290361575782299,
|
|
"num_tokens": 2583308.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.420801258087158,
|
|
"epoch": 0.11338896952104499,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004999942301404916,
|
|
"loss": 6.3833,
|
|
"mean_token_accuracy": 0.11551335826516151,
|
|
"num_tokens": 2595105.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.13080940246582,
|
|
"epoch": 0.11384252539912917,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004999939960937484,
|
|
"loss": 6.1801,
|
|
"mean_token_accuracy": 0.13008004054427147,
|
|
"num_tokens": 2604937.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.238644075393677,
|
|
"epoch": 0.11429608127721336,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004999937573940721,
|
|
"loss": 6.259,
|
|
"mean_token_accuracy": 0.12672679945826532,
|
|
"num_tokens": 2615218.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.200268936157227,
|
|
"epoch": 0.11474963715529753,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999935140414674,
|
|
"loss": 6.1979,
|
|
"mean_token_accuracy": 0.12415119707584381,
|
|
"num_tokens": 2625668.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.2111094951629635,
|
|
"epoch": 0.11520319303338171,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004999932660359395,
|
|
"loss": 6.2862,
|
|
"mean_token_accuracy": 0.12427769452333451,
|
|
"num_tokens": 2636233.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.263781785964966,
|
|
"epoch": 0.11565674891146589,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999930133774936,
|
|
"loss": 6.2352,
|
|
"mean_token_accuracy": 0.12104817926883697,
|
|
"num_tokens": 2647725.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.226747751235962,
|
|
"epoch": 0.11611030478955008,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999927560661348,
|
|
"loss": 6.2223,
|
|
"mean_token_accuracy": 0.13130831718444824,
|
|
"num_tokens": 2658670.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.139042472839355,
|
|
"epoch": 0.11656386066763426,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999924941018685,
|
|
"loss": 6.2077,
|
|
"mean_token_accuracy": 0.13245302811264992,
|
|
"num_tokens": 2668782.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.23336877822876,
|
|
"epoch": 0.11701741654571843,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999922274847,
|
|
"loss": 6.2498,
|
|
"mean_token_accuracy": 0.12909818440675735,
|
|
"num_tokens": 2677973.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.244420528411865,
|
|
"epoch": 0.11747097242380261,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000499991956214635,
|
|
"loss": 6.1934,
|
|
"mean_token_accuracy": 0.12651574686169625,
|
|
"num_tokens": 2687813.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.160448169708252,
|
|
"epoch": 0.1179245283018868,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000499991680291679,
|
|
"loss": 6.2207,
|
|
"mean_token_accuracy": 0.1258676081895828,
|
|
"num_tokens": 2698664.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.191532325744629,
|
|
"epoch": 0.11837808417997098,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999913997158378,
|
|
"loss": 6.1572,
|
|
"mean_token_accuracy": 0.13130440413951874,
|
|
"num_tokens": 2708936.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.242334794998169,
|
|
"epoch": 0.11883164005805515,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004999911144871169,
|
|
"loss": 6.2468,
|
|
"mean_token_accuracy": 0.1300816871225834,
|
|
"num_tokens": 2718711.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.0896015644073485,
|
|
"epoch": 0.11928519593613933,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999908246055226,
|
|
"loss": 6.1894,
|
|
"mean_token_accuracy": 0.13144304975867271,
|
|
"num_tokens": 2729956.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.332587003707886,
|
|
"epoch": 0.11973875181422351,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004999905300710607,
|
|
"loss": 6.3051,
|
|
"mean_token_accuracy": 0.12343701347708702,
|
|
"num_tokens": 2740932.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.29391508102417,
|
|
"epoch": 0.1201923076923077,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004999902308837373,
|
|
"loss": 6.3452,
|
|
"mean_token_accuracy": 0.12498589381575584,
|
|
"num_tokens": 2750781.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.286653852462768,
|
|
"epoch": 0.12064586357039188,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004999899270435584,
|
|
"loss": 6.2415,
|
|
"mean_token_accuracy": 0.12924491837620736,
|
|
"num_tokens": 2761542.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.169296550750732,
|
|
"epoch": 0.12109941944847605,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004999896185505307,
|
|
"loss": 6.1863,
|
|
"mean_token_accuracy": 0.12321913093328477,
|
|
"num_tokens": 2772507.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.211149549484253,
|
|
"epoch": 0.12155297532656023,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999893054046603,
|
|
"loss": 6.2009,
|
|
"mean_token_accuracy": 0.12625789046287536,
|
|
"num_tokens": 2782806.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.175951194763184,
|
|
"epoch": 0.12200653120464441,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004999889876059537,
|
|
"loss": 6.1991,
|
|
"mean_token_accuracy": 0.12693845182657243,
|
|
"num_tokens": 2793484.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.184288167953492,
|
|
"epoch": 0.1224600870827286,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999886651544176,
|
|
"loss": 6.2731,
|
|
"mean_token_accuracy": 0.1297612629830837,
|
|
"num_tokens": 2804443.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.163767099380493,
|
|
"epoch": 0.12291364296081277,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999883380500584,
|
|
"loss": 6.1485,
|
|
"mean_token_accuracy": 0.1262630842626095,
|
|
"num_tokens": 2814840.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.2059821605682375,
|
|
"epoch": 0.12336719883889695,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999880062928831,
|
|
"loss": 6.2207,
|
|
"mean_token_accuracy": 0.1247759722173214,
|
|
"num_tokens": 2824789.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.2396345138549805,
|
|
"epoch": 0.12382075471698113,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999876698828985,
|
|
"loss": 6.1775,
|
|
"mean_token_accuracy": 0.13116121515631676,
|
|
"num_tokens": 2834501.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.202531719207764,
|
|
"epoch": 0.12427431059506532,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004999873288201116,
|
|
"loss": 6.1893,
|
|
"mean_token_accuracy": 0.12950848862528802,
|
|
"num_tokens": 2845399.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.032520198822022,
|
|
"epoch": 0.1247278664731495,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999869831045294,
|
|
"loss": 6.0872,
|
|
"mean_token_accuracy": 0.14028632938861846,
|
|
"num_tokens": 2855146.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.3156661033630375,
|
|
"epoch": 0.12518142235123367,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004999866327361589,
|
|
"loss": 6.2597,
|
|
"mean_token_accuracy": 0.12432174906134605,
|
|
"num_tokens": 2866438.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.019028043746948,
|
|
"epoch": 0.12563497822931785,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999862777150077,
|
|
"loss": 6.1301,
|
|
"mean_token_accuracy": 0.12995802015066146,
|
|
"num_tokens": 2876802.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.315221118927002,
|
|
"epoch": 0.12608853410740203,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999859180410829,
|
|
"loss": 6.3285,
|
|
"mean_token_accuracy": 0.128648017346859,
|
|
"num_tokens": 2888657.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.177397584915161,
|
|
"epoch": 0.12654208998548622,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004999855537143919,
|
|
"loss": 6.1832,
|
|
"mean_token_accuracy": 0.1301404818892479,
|
|
"num_tokens": 2899605.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.116765117645263,
|
|
"epoch": 0.1269956458635704,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004999851847349425,
|
|
"loss": 6.1427,
|
|
"mean_token_accuracy": 0.1246509574353695,
|
|
"num_tokens": 2910332.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.126271677017212,
|
|
"epoch": 0.12744920174165458,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004999848111027419,
|
|
"loss": 6.1231,
|
|
"mean_token_accuracy": 0.1296907387673855,
|
|
"num_tokens": 2921226.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.1753684997558596,
|
|
"epoch": 0.12790275761973874,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999844328177983,
|
|
"loss": 6.165,
|
|
"mean_token_accuracy": 0.129497080296278,
|
|
"num_tokens": 2930941.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.1793512344360355,
|
|
"epoch": 0.12835631349782292,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004999840498801191,
|
|
"loss": 6.1917,
|
|
"mean_token_accuracy": 0.13117415681481362,
|
|
"num_tokens": 2941383.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.207342672348022,
|
|
"epoch": 0.1288098693759071,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004999836622897126,
|
|
"loss": 6.1587,
|
|
"mean_token_accuracy": 0.12762442380189895,
|
|
"num_tokens": 2951634.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.203984594345092,
|
|
"epoch": 0.1292634252539913,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999832700465865,
|
|
"loss": 6.2219,
|
|
"mean_token_accuracy": 0.12739467918872832,
|
|
"num_tokens": 2962240.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.187308359146118,
|
|
"epoch": 0.12971698113207547,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004999828731507491,
|
|
"loss": 6.1977,
|
|
"mean_token_accuracy": 0.12866248711943626,
|
|
"num_tokens": 2973466.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.248796129226685,
|
|
"epoch": 0.13017053701015965,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999824716022085,
|
|
"loss": 6.1989,
|
|
"mean_token_accuracy": 0.13207736387848854,
|
|
"num_tokens": 2982986.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.168663787841797,
|
|
"epoch": 0.13062409288824384,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004999820654009731,
|
|
"loss": 6.2264,
|
|
"mean_token_accuracy": 0.1301150493323803,
|
|
"num_tokens": 2994151.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.106994342803955,
|
|
"epoch": 0.13107764876632802,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999816545470513,
|
|
"loss": 6.0382,
|
|
"mean_token_accuracy": 0.1328746996819973,
|
|
"num_tokens": 3003481.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.062295961380005,
|
|
"epoch": 0.1315312046444122,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004999812390404514,
|
|
"loss": 6.1274,
|
|
"mean_token_accuracy": 0.1257198430597782,
|
|
"num_tokens": 3013835.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.146769189834595,
|
|
"epoch": 0.13198476052249636,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004999808188811823,
|
|
"loss": 6.1207,
|
|
"mean_token_accuracy": 0.14100732132792473,
|
|
"num_tokens": 3024292.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.171922969818115,
|
|
"epoch": 0.13243831640058054,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004999803940692524,
|
|
"loss": 6.2384,
|
|
"mean_token_accuracy": 0.12596325874328612,
|
|
"num_tokens": 3035114.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.1891436100006105,
|
|
"epoch": 0.13289187227866472,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004999799646046707,
|
|
"loss": 6.1387,
|
|
"mean_token_accuracy": 0.13315948471426964,
|
|
"num_tokens": 3045151.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.107039928436279,
|
|
"epoch": 0.1333454281567489,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999795304874461,
|
|
"loss": 6.1193,
|
|
"mean_token_accuracy": 0.13525033965706826,
|
|
"num_tokens": 3054552.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.211303329467773,
|
|
"epoch": 0.1337989840348331,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004999790917175873,
|
|
"loss": 6.2495,
|
|
"mean_token_accuracy": 0.12571337893605233,
|
|
"num_tokens": 3065386.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.1838174819946286,
|
|
"epoch": 0.13425253991291727,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004999786482951036,
|
|
"loss": 6.0979,
|
|
"mean_token_accuracy": 0.13161127269268036,
|
|
"num_tokens": 3075934.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.044588327407837,
|
|
"epoch": 0.13470609579100146,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004999782002200041,
|
|
"loss": 6.0649,
|
|
"mean_token_accuracy": 0.13151389360427856,
|
|
"num_tokens": 3086578.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.235484266281128,
|
|
"epoch": 0.13515965166908564,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004999777474922982,
|
|
"loss": 6.2253,
|
|
"mean_token_accuracy": 0.12512897700071335,
|
|
"num_tokens": 3098183.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.0513090133667,
|
|
"epoch": 0.13561320754716982,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999772901119951,
|
|
"loss": 6.0581,
|
|
"mean_token_accuracy": 0.1343822069466114,
|
|
"num_tokens": 3107649.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.221631050109863,
|
|
"epoch": 0.13606676342525398,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999768280791043,
|
|
"loss": 6.1646,
|
|
"mean_token_accuracy": 0.13448878824710847,
|
|
"num_tokens": 3117652.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 6.061485052108765,
|
|
"epoch": 0.13652031930333816,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004999763613936352,
|
|
"loss": 6.1065,
|
|
"mean_token_accuracy": 0.13679977506399155,
|
|
"num_tokens": 3127785.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 6.083025693893433,
|
|
"epoch": 0.13697387518142234,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999758900555979,
|
|
"loss": 6.0635,
|
|
"mean_token_accuracy": 0.13337264955043793,
|
|
"num_tokens": 3138657.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 6.140411472320556,
|
|
"epoch": 0.13742743105950653,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004999754140650017,
|
|
"loss": 6.1911,
|
|
"mean_token_accuracy": 0.12004235461354255,
|
|
"num_tokens": 3150692.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 6.23555703163147,
|
|
"epoch": 0.1378809869375907,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999749334218567,
|
|
"loss": 6.2595,
|
|
"mean_token_accuracy": 0.12430977448821068,
|
|
"num_tokens": 3161807.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 6.107507514953613,
|
|
"epoch": 0.1383345428156749,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999744481261726,
|
|
"loss": 6.0703,
|
|
"mean_token_accuracy": 0.1393376462161541,
|
|
"num_tokens": 3172446.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 6.145687675476074,
|
|
"epoch": 0.13878809869375908,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999739581779597,
|
|
"loss": 6.2071,
|
|
"mean_token_accuracy": 0.1238051250576973,
|
|
"num_tokens": 3183476.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 6.0691694736480715,
|
|
"epoch": 0.13924165457184326,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000499973463577228,
|
|
"loss": 6.0885,
|
|
"mean_token_accuracy": 0.13867690786719322,
|
|
"num_tokens": 3193586.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 6.1251280307769775,
|
|
"epoch": 0.13969521044992744,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999729643239877,
|
|
"loss": 6.1195,
|
|
"mean_token_accuracy": 0.13339588195085525,
|
|
"num_tokens": 3203341.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 6.136832141876221,
|
|
"epoch": 0.1401487663280116,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999724604182492,
|
|
"loss": 6.1088,
|
|
"mean_token_accuracy": 0.12823742404580116,
|
|
"num_tokens": 3214177.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 6.141563987731933,
|
|
"epoch": 0.14060232220609578,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999719518600229,
|
|
"loss": 6.182,
|
|
"mean_token_accuracy": 0.1269642286002636,
|
|
"num_tokens": 3224516.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 6.098951148986816,
|
|
"epoch": 0.14105587808417996,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999714386493192,
|
|
"loss": 6.0717,
|
|
"mean_token_accuracy": 0.14152515679597855,
|
|
"num_tokens": 3234001.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 6.196884202957153,
|
|
"epoch": 0.14150943396226415,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004999709207861489,
|
|
"loss": 6.1286,
|
|
"mean_token_accuracy": 0.13318521454930304,
|
|
"num_tokens": 3246095.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 6.135161066055298,
|
|
"epoch": 0.14196298984034833,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999703982705227,
|
|
"loss": 6.177,
|
|
"mean_token_accuracy": 0.12648626491427423,
|
|
"num_tokens": 3255828.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 6.101465272903442,
|
|
"epoch": 0.1424165457184325,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999698711024513,
|
|
"loss": 6.1378,
|
|
"mean_token_accuracy": 0.13857534304261207,
|
|
"num_tokens": 3265127.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 6.23589973449707,
|
|
"epoch": 0.1428701015965167,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999693392819456,
|
|
"loss": 6.2083,
|
|
"mean_token_accuracy": 0.12722571119666098,
|
|
"num_tokens": 3277205.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 6.039449787139892,
|
|
"epoch": 0.14332365747460088,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999688028090166,
|
|
"loss": 6.1109,
|
|
"mean_token_accuracy": 0.13461936116218567,
|
|
"num_tokens": 3287119.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 6.237564992904663,
|
|
"epoch": 0.14377721335268506,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999682616836755,
|
|
"loss": 6.2073,
|
|
"mean_token_accuracy": 0.12796173691749574,
|
|
"num_tokens": 3297324.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 6.166965675354004,
|
|
"epoch": 0.14423076923076922,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004999677159059334,
|
|
"loss": 6.1568,
|
|
"mean_token_accuracy": 0.12957976162433624,
|
|
"num_tokens": 3308162.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 6.156198501586914,
|
|
"epoch": 0.1446843251088534,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004999671654758016,
|
|
"loss": 6.1587,
|
|
"mean_token_accuracy": 0.12751780077815056,
|
|
"num_tokens": 3319422.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 6.074300146102905,
|
|
"epoch": 0.14513788098693758,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004999666103932915,
|
|
"loss": 6.1303,
|
|
"mean_token_accuracy": 0.13079726696014404,
|
|
"num_tokens": 3328668.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 6.196269941329956,
|
|
"epoch": 0.14559143686502177,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999660506584145,
|
|
"loss": 6.1789,
|
|
"mean_token_accuracy": 0.12973756417632104,
|
|
"num_tokens": 3338563.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 6.123033428192139,
|
|
"epoch": 0.14604499274310595,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999654862711823,
|
|
"loss": 6.1866,
|
|
"mean_token_accuracy": 0.13245568349957465,
|
|
"num_tokens": 3348025.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 6.125161266326904,
|
|
"epoch": 0.14649854862119013,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004999649172316065,
|
|
"loss": 6.0621,
|
|
"mean_token_accuracy": 0.14053588956594468,
|
|
"num_tokens": 3359094.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 6.113393926620484,
|
|
"epoch": 0.14695210449927432,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000499964343539699,
|
|
"loss": 6.1022,
|
|
"mean_token_accuracy": 0.13681961521506308,
|
|
"num_tokens": 3368965.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 6.058830976486206,
|
|
"epoch": 0.1474056603773585,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004999637651954714,
|
|
"loss": 6.1126,
|
|
"mean_token_accuracy": 0.13101110011339187,
|
|
"num_tokens": 3378612.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 6.140189456939697,
|
|
"epoch": 0.14785921625544268,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004999631821989358,
|
|
"loss": 6.0731,
|
|
"mean_token_accuracy": 0.13366867899894713,
|
|
"num_tokens": 3389634.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 6.045487594604492,
|
|
"epoch": 0.14831277213352684,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004999625945501043,
|
|
"loss": 6.1454,
|
|
"mean_token_accuracy": 0.13442882299423217,
|
|
"num_tokens": 3399604.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 6.223248147964478,
|
|
"epoch": 0.14876632801161102,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000499962002248989,
|
|
"loss": 6.0949,
|
|
"mean_token_accuracy": 0.13319409787654876,
|
|
"num_tokens": 3410314.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 6.071327877044678,
|
|
"epoch": 0.1492198838896952,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004999614052956024,
|
|
"loss": 6.1164,
|
|
"mean_token_accuracy": 0.13651233538985252,
|
|
"num_tokens": 3420773.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 6.234537887573242,
|
|
"epoch": 0.1496734397677794,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999608036899563,
|
|
"loss": 6.2406,
|
|
"mean_token_accuracy": 0.1285805732011795,
|
|
"num_tokens": 3431702.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 6.026836967468261,
|
|
"epoch": 0.15012699564586357,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999601974320636,
|
|
"loss": 6.1376,
|
|
"mean_token_accuracy": 0.133916974067688,
|
|
"num_tokens": 3441925.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 6.225948667526245,
|
|
"epoch": 0.15058055152394775,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004999595865219367,
|
|
"loss": 6.1489,
|
|
"mean_token_accuracy": 0.13318530544638635,
|
|
"num_tokens": 3452391.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 6.018878602981568,
|
|
"epoch": 0.15103410740203194,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004999589709595881,
|
|
"loss": 6.0672,
|
|
"mean_token_accuracy": 0.12874112129211426,
|
|
"num_tokens": 3463028.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 6.063703966140747,
|
|
"epoch": 0.15148766328011612,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999583507450308,
|
|
"loss": 6.0641,
|
|
"mean_token_accuracy": 0.13683240562677385,
|
|
"num_tokens": 3473470.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 6.126550817489624,
|
|
"epoch": 0.1519412191582003,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999577258782774,
|
|
"loss": 6.1066,
|
|
"mean_token_accuracy": 0.13182729333639145,
|
|
"num_tokens": 3484486.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 6.084259557723999,
|
|
"epoch": 0.15239477503628446,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000499957096359341,
|
|
"loss": 6.0679,
|
|
"mean_token_accuracy": 0.13736046478152275,
|
|
"num_tokens": 3494885.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 6.07341947555542,
|
|
"epoch": 0.15284833091436864,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004999564621882343,
|
|
"loss": 6.079,
|
|
"mean_token_accuracy": 0.13572860211133958,
|
|
"num_tokens": 3506088.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 6.015376949310303,
|
|
"epoch": 0.15330188679245282,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999558233649708,
|
|
"loss": 6.0114,
|
|
"mean_token_accuracy": 0.13501859828829765,
|
|
"num_tokens": 3516380.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 6.094719457626343,
|
|
"epoch": 0.153755442670537,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999551798895636,
|
|
"loss": 6.0229,
|
|
"mean_token_accuracy": 0.14153220504522324,
|
|
"num_tokens": 3525965.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 6.060795879364013,
|
|
"epoch": 0.1542089985486212,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999545317620258,
|
|
"loss": 6.04,
|
|
"mean_token_accuracy": 0.13283166736364366,
|
|
"num_tokens": 3536592.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 6.117245006561279,
|
|
"epoch": 0.15466255442670537,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999538789823711,
|
|
"loss": 6.1651,
|
|
"mean_token_accuracy": 0.13683197945356368,
|
|
"num_tokens": 3546001.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 6.123035430908203,
|
|
"epoch": 0.15511611030478956,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999532215506127,
|
|
"loss": 6.0487,
|
|
"mean_token_accuracy": 0.12888889610767365,
|
|
"num_tokens": 3555575.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 6.06559419631958,
|
|
"epoch": 0.15556966618287374,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004999525594667645,
|
|
"loss": 6.0429,
|
|
"mean_token_accuracy": 0.14311017096042633,
|
|
"num_tokens": 3565685.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 6.119355535507202,
|
|
"epoch": 0.15602322206095792,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00049995189273084,
|
|
"loss": 6.0872,
|
|
"mean_token_accuracy": 0.14001662507653237,
|
|
"num_tokens": 3575119.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 6.048756504058838,
|
|
"epoch": 0.15647677793904208,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000499951221342853,
|
|
"loss": 6.0393,
|
|
"mean_token_accuracy": 0.13779830187559128,
|
|
"num_tokens": 3584620.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 6.0873788356781,
|
|
"epoch": 0.15693033381712626,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004999505453028174,
|
|
"loss": 6.0783,
|
|
"mean_token_accuracy": 0.13565813079476358,
|
|
"num_tokens": 3595403.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 6.042232465744019,
|
|
"epoch": 0.15738388969521044,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999498646107472,
|
|
"loss": 6.0665,
|
|
"mean_token_accuracy": 0.135944102704525,
|
|
"num_tokens": 3606355.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 6.039039087295532,
|
|
"epoch": 0.15783744557329463,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999491792666566,
|
|
"loss": 6.0526,
|
|
"mean_token_accuracy": 0.1334886446595192,
|
|
"num_tokens": 3616222.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 6.117766857147217,
|
|
"epoch": 0.1582910014513788,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004999484892705597,
|
|
"loss": 6.0978,
|
|
"mean_token_accuracy": 0.1427782490849495,
|
|
"num_tokens": 3625179.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 6.066883754730225,
|
|
"epoch": 0.158744557329463,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004999477946224705,
|
|
"loss": 6.057,
|
|
"mean_token_accuracy": 0.13318037912249564,
|
|
"num_tokens": 3636256.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 5.9570455074310305,
|
|
"epoch": 0.15919811320754718,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999470953224037,
|
|
"loss": 5.9024,
|
|
"mean_token_accuracy": 0.14086213558912278,
|
|
"num_tokens": 3645919.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 6.067557430267334,
|
|
"epoch": 0.15965166908563136,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999463913703735,
|
|
"loss": 6.1052,
|
|
"mean_token_accuracy": 0.13390027210116387,
|
|
"num_tokens": 3657008.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 6.014396524429321,
|
|
"epoch": 0.16010522496371554,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004999456827663948,
|
|
"loss": 5.9915,
|
|
"mean_token_accuracy": 0.13444581776857376,
|
|
"num_tokens": 3667686.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 6.022483873367309,
|
|
"epoch": 0.1605587808417997,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000499944969510482,
|
|
"loss": 5.9907,
|
|
"mean_token_accuracy": 0.13489970788359643,
|
|
"num_tokens": 3678972.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 6.032540559768677,
|
|
"epoch": 0.16101233671988388,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004999442516026498,
|
|
"loss": 5.9973,
|
|
"mean_token_accuracy": 0.13953388929367067,
|
|
"num_tokens": 3688903.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 6.049791049957276,
|
|
"epoch": 0.16146589259796806,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999435290429133,
|
|
"loss": 6.0542,
|
|
"mean_token_accuracy": 0.13899827301502227,
|
|
"num_tokens": 3697880.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 6.113507795333862,
|
|
"epoch": 0.16191944847605225,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999428018312872,
|
|
"loss": 6.1146,
|
|
"mean_token_accuracy": 0.13196710646152496,
|
|
"num_tokens": 3708500.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 5.967461633682251,
|
|
"epoch": 0.16237300435413643,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999420699677867,
|
|
"loss": 5.9456,
|
|
"mean_token_accuracy": 0.14808355420827865,
|
|
"num_tokens": 3718711.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 6.003653526306152,
|
|
"epoch": 0.1628265602322206,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004999413334524269,
|
|
"loss": 6.0784,
|
|
"mean_token_accuracy": 0.13483019024133683,
|
|
"num_tokens": 3729431.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 6.118337678909302,
|
|
"epoch": 0.1632801161103048,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999405922852229,
|
|
"loss": 6.0157,
|
|
"mean_token_accuracy": 0.13496499881148338,
|
|
"num_tokens": 3741061.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 5.951924753189087,
|
|
"epoch": 0.16373367198838898,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999398464661901,
|
|
"loss": 6.0049,
|
|
"mean_token_accuracy": 0.14013575315475463,
|
|
"num_tokens": 3751878.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 6.038108873367309,
|
|
"epoch": 0.16418722786647316,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999390959953441,
|
|
"loss": 6.0101,
|
|
"mean_token_accuracy": 0.13939072266221048,
|
|
"num_tokens": 3762039.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 5.974385404586792,
|
|
"epoch": 0.16464078374455732,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004999383408727002,
|
|
"loss": 5.9441,
|
|
"mean_token_accuracy": 0.1419306755065918,
|
|
"num_tokens": 3770755.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 6.02751522064209,
|
|
"epoch": 0.1650943396226415,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004999375810982741,
|
|
"loss": 6.1083,
|
|
"mean_token_accuracy": 0.13299722522497176,
|
|
"num_tokens": 3781698.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 6.067613887786865,
|
|
"epoch": 0.16554789550072568,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004999368166720815,
|
|
"loss": 6.1488,
|
|
"mean_token_accuracy": 0.13327884972095488,
|
|
"num_tokens": 3792508.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 6.086944532394409,
|
|
"epoch": 0.16600145137880987,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004999360475941382,
|
|
"loss": 6.0028,
|
|
"mean_token_accuracy": 0.14385750517249107,
|
|
"num_tokens": 3803065.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 6.022203254699707,
|
|
"epoch": 0.16645500725689405,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00049993527386446,
|
|
"loss": 6.0675,
|
|
"mean_token_accuracy": 0.1315029554069042,
|
|
"num_tokens": 3813696.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 6.0408408641815186,
|
|
"epoch": 0.16690856313497823,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004999344954830633,
|
|
"loss": 6.022,
|
|
"mean_token_accuracy": 0.14210820943117142,
|
|
"num_tokens": 3823786.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 6.0095936298370365,
|
|
"epoch": 0.16736211901306242,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999337124499638,
|
|
"loss": 5.9917,
|
|
"mean_token_accuracy": 0.13492688685655593,
|
|
"num_tokens": 3835011.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 6.074155664443969,
|
|
"epoch": 0.1678156748911466,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004999329247651777,
|
|
"loss": 6.0408,
|
|
"mean_token_accuracy": 0.13303120210766792,
|
|
"num_tokens": 3846161.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 5.95374641418457,
|
|
"epoch": 0.16826923076923078,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999321324287213,
|
|
"loss": 5.972,
|
|
"mean_token_accuracy": 0.14118631407618523,
|
|
"num_tokens": 3856013.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 6.1738697528839115,
|
|
"epoch": 0.16872278664731494,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004999313354406114,
|
|
"loss": 6.1002,
|
|
"mean_token_accuracy": 0.1409381054341793,
|
|
"num_tokens": 3865521.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 5.912585163116455,
|
|
"epoch": 0.16917634252539912,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000499930533800864,
|
|
"loss": 5.9034,
|
|
"mean_token_accuracy": 0.14283333346247673,
|
|
"num_tokens": 3876290.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 5.980841970443725,
|
|
"epoch": 0.1696298984034833,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999297275094958,
|
|
"loss": 6.1048,
|
|
"mean_token_accuracy": 0.13230404928326606,
|
|
"num_tokens": 3886437.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 6.112230062484741,
|
|
"epoch": 0.1700834542815675,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999289165665235,
|
|
"loss": 6.1435,
|
|
"mean_token_accuracy": 0.13440538123250007,
|
|
"num_tokens": 3896501.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 6.15153021812439,
|
|
"epoch": 0.17053701015965167,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004999281009719639,
|
|
"loss": 6.0671,
|
|
"mean_token_accuracy": 0.13559964820742607,
|
|
"num_tokens": 3906725.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 6.02051043510437,
|
|
"epoch": 0.17099056603773585,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004999272807258339,
|
|
"loss": 5.9351,
|
|
"mean_token_accuracy": 0.1449750393629074,
|
|
"num_tokens": 3917428.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 5.859882068634033,
|
|
"epoch": 0.17144412191582004,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004999264558281504,
|
|
"loss": 6.0304,
|
|
"mean_token_accuracy": 0.1393364980816841,
|
|
"num_tokens": 3927298.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 6.047381448745727,
|
|
"epoch": 0.17189767779390422,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999256262789305,
|
|
"loss": 5.9394,
|
|
"mean_token_accuracy": 0.1372090183198452,
|
|
"num_tokens": 3936727.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 6.062630701065063,
|
|
"epoch": 0.1723512336719884,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004999247920781913,
|
|
"loss": 6.0596,
|
|
"mean_token_accuracy": 0.13730003088712692,
|
|
"num_tokens": 3947246.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 5.838617277145386,
|
|
"epoch": 0.17280478955007256,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999239532259502,
|
|
"loss": 5.9592,
|
|
"mean_token_accuracy": 0.14350975602865218,
|
|
"num_tokens": 3958535.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 6.161811971664429,
|
|
"epoch": 0.17325834542815674,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999231097222244,
|
|
"loss": 6.0521,
|
|
"mean_token_accuracy": 0.1353868529200554,
|
|
"num_tokens": 3969635.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 6.025433874130249,
|
|
"epoch": 0.17371190130624092,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999222615670312,
|
|
"loss": 6.024,
|
|
"mean_token_accuracy": 0.1359100967645645,
|
|
"num_tokens": 3980227.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 6.06690092086792,
|
|
"epoch": 0.1741654571843251,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004999214087603885,
|
|
"loss": 6.0263,
|
|
"mean_token_accuracy": 0.14165394008159637,
|
|
"num_tokens": 3990915.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 5.958260536193848,
|
|
"epoch": 0.1746190130624093,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999205513023135,
|
|
"loss": 5.9216,
|
|
"mean_token_accuracy": 0.13917687460780143,
|
|
"num_tokens": 4001273.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 6.072769498825073,
|
|
"epoch": 0.17507256894049347,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999196891928245,
|
|
"loss": 5.994,
|
|
"mean_token_accuracy": 0.141386017203331,
|
|
"num_tokens": 4011596.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 5.960225534439087,
|
|
"epoch": 0.17552612481857766,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004999188224319388,
|
|
"loss": 6.0366,
|
|
"mean_token_accuracy": 0.13730789497494697,
|
|
"num_tokens": 4022638.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 6.080376815795899,
|
|
"epoch": 0.17597968069666184,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004999179510196746,
|
|
"loss": 6.0156,
|
|
"mean_token_accuracy": 0.13227792233228683,
|
|
"num_tokens": 4032398.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 5.898893404006958,
|
|
"epoch": 0.17643323657474602,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999170749560498,
|
|
"loss": 5.9724,
|
|
"mean_token_accuracy": 0.1371435798704624,
|
|
"num_tokens": 4042852.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 6.040456962585449,
|
|
"epoch": 0.17688679245283018,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004999161942410827,
|
|
"loss": 6.0056,
|
|
"mean_token_accuracy": 0.1349559858441353,
|
|
"num_tokens": 4053110.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 6.002525901794433,
|
|
"epoch": 0.17734034833091436,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999153088747913,
|
|
"loss": 6.0212,
|
|
"mean_token_accuracy": 0.1401715062558651,
|
|
"num_tokens": 4064571.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 5.953612756729126,
|
|
"epoch": 0.17779390420899854,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000499914418857194,
|
|
"loss": 5.9906,
|
|
"mean_token_accuracy": 0.13749739825725554,
|
|
"num_tokens": 4075400.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 6.028836965560913,
|
|
"epoch": 0.17824746008708273,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999135241883092,
|
|
"loss": 5.9439,
|
|
"mean_token_accuracy": 0.14185010865330697,
|
|
"num_tokens": 4085763.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 5.986079740524292,
|
|
"epoch": 0.1787010159651669,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999126248681555,
|
|
"loss": 6.0109,
|
|
"mean_token_accuracy": 0.13681597337126733,
|
|
"num_tokens": 4095993.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 5.986701774597168,
|
|
"epoch": 0.1791545718432511,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004999117208967513,
|
|
"loss": 5.9225,
|
|
"mean_token_accuracy": 0.1380699433386326,
|
|
"num_tokens": 4105012.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 5.942403030395508,
|
|
"epoch": 0.17960812772133528,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999108122741154,
|
|
"loss": 5.9692,
|
|
"mean_token_accuracy": 0.1353602610528469,
|
|
"num_tokens": 4115142.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 6.031651258468628,
|
|
"epoch": 0.18006168359941946,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999098990002666,
|
|
"loss": 5.957,
|
|
"mean_token_accuracy": 0.13746799603104592,
|
|
"num_tokens": 4125297.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 5.977378511428833,
|
|
"epoch": 0.18051523947750364,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004999089810752237,
|
|
"loss": 6.0917,
|
|
"mean_token_accuracy": 0.132157414406538,
|
|
"num_tokens": 4136404.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 6.080759954452515,
|
|
"epoch": 0.1809687953555878,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000499908058499006,
|
|
"loss": 5.9681,
|
|
"mean_token_accuracy": 0.14288827329874038,
|
|
"num_tokens": 4146228.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 5.95260910987854,
|
|
"epoch": 0.18142235123367198,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999071312716321,
|
|
"loss": 5.8708,
|
|
"mean_token_accuracy": 0.1427669905126095,
|
|
"num_tokens": 4156435.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 5.975965404510498,
|
|
"epoch": 0.18187590711175616,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999061993931215,
|
|
"loss": 6.0042,
|
|
"mean_token_accuracy": 0.1480802610516548,
|
|
"num_tokens": 4166525.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 6.0720906257629395,
|
|
"epoch": 0.18232946298984035,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004999052628634934,
|
|
"loss": 6.0833,
|
|
"mean_token_accuracy": 0.13352881222963334,
|
|
"num_tokens": 4176831.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 6.019681024551391,
|
|
"epoch": 0.18278301886792453,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004999043216827671,
|
|
"loss": 6.06,
|
|
"mean_token_accuracy": 0.13369136229157447,
|
|
"num_tokens": 4187830.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 6.006418657302857,
|
|
"epoch": 0.1832365747460087,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004999033758509621,
|
|
"loss": 5.8688,
|
|
"mean_token_accuracy": 0.1393364667892456,
|
|
"num_tokens": 4198898.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 5.97959532737732,
|
|
"epoch": 0.1836901306240929,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000499902425368098,
|
|
"loss": 5.9709,
|
|
"mean_token_accuracy": 0.1392282821238041,
|
|
"num_tokens": 4210297.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 6.037605619430542,
|
|
"epoch": 0.18414368650217708,
|
|
"grad_norm": 2.5625,
|
|
"learning_rate": 0.0004999014702341944,
|
|
"loss": 6.1283,
|
|
"mean_token_accuracy": 0.13221636712551116,
|
|
"num_tokens": 4221469.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 5.987315082550049,
|
|
"epoch": 0.18459724238026126,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999005104492712,
|
|
"loss": 5.9446,
|
|
"mean_token_accuracy": 0.14053445905447007,
|
|
"num_tokens": 4233046.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 6.048150110244751,
|
|
"epoch": 0.18505079825834542,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004998995460133481,
|
|
"loss": 5.9829,
|
|
"mean_token_accuracy": 0.13854565769433974,
|
|
"num_tokens": 4243233.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 6.046663284301758,
|
|
"epoch": 0.1855043541364296,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004998985769264451,
|
|
"loss": 6.0973,
|
|
"mean_token_accuracy": 0.13540371656417846,
|
|
"num_tokens": 4254020.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 5.959017562866211,
|
|
"epoch": 0.18595791001451378,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004998976031885821,
|
|
"loss": 5.8928,
|
|
"mean_token_accuracy": 0.14246483519673347,
|
|
"num_tokens": 4263901.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 5.890605878829956,
|
|
"epoch": 0.18641146589259797,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004998966247997794,
|
|
"loss": 5.9255,
|
|
"mean_token_accuracy": 0.13970548510551453,
|
|
"num_tokens": 4274814.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 6.063959264755249,
|
|
"epoch": 0.18686502177068215,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004998956417600572,
|
|
"loss": 6.0704,
|
|
"mean_token_accuracy": 0.1417410969734192,
|
|
"num_tokens": 4284846.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 6.033328819274902,
|
|
"epoch": 0.18731857764876633,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004998946540694358,
|
|
"loss": 5.9731,
|
|
"mean_token_accuracy": 0.14464445784687996,
|
|
"num_tokens": 4294862.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 6.068034410476685,
|
|
"epoch": 0.18777213352685052,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004998936617279357,
|
|
"loss": 6.0175,
|
|
"mean_token_accuracy": 0.1416530877351761,
|
|
"num_tokens": 4305708.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 5.995953607559204,
|
|
"epoch": 0.1882256894049347,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004998926647355773,
|
|
"loss": 5.9943,
|
|
"mean_token_accuracy": 0.13258325606584548,
|
|
"num_tokens": 4315853.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 5.9227760314941404,
|
|
"epoch": 0.18867924528301888,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004998916630923813,
|
|
"loss": 5.9565,
|
|
"mean_token_accuracy": 0.13823306187987328,
|
|
"num_tokens": 4326607.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 5.959938955307007,
|
|
"epoch": 0.18913280116110304,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004998906567983685,
|
|
"loss": 5.991,
|
|
"mean_token_accuracy": 0.13902745693922042,
|
|
"num_tokens": 4336050.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 6.056903553009033,
|
|
"epoch": 0.18958635703918722,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004998896458535595,
|
|
"loss": 5.9604,
|
|
"mean_token_accuracy": 0.14332208260893822,
|
|
"num_tokens": 4345453.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 5.951250743865967,
|
|
"epoch": 0.1900399129172714,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004998886302579752,
|
|
"loss": 5.9444,
|
|
"mean_token_accuracy": 0.1378558337688446,
|
|
"num_tokens": 4356372.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 5.958211088180542,
|
|
"epoch": 0.19049346879535559,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000499887610011637,
|
|
"loss": 6.0014,
|
|
"mean_token_accuracy": 0.14464358389377593,
|
|
"num_tokens": 4365500.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 6.067464399337768,
|
|
"epoch": 0.19094702467343977,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004998865851145656,
|
|
"loss": 5.9311,
|
|
"mean_token_accuracy": 0.14481838345527648,
|
|
"num_tokens": 4375031.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 5.901741409301758,
|
|
"epoch": 0.19140058055152395,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004998855555667822,
|
|
"loss": 5.972,
|
|
"mean_token_accuracy": 0.1389401637017727,
|
|
"num_tokens": 4386409.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 6.059998941421509,
|
|
"epoch": 0.19185413642960814,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004998845213683084,
|
|
"loss": 5.9942,
|
|
"mean_token_accuracy": 0.13713788390159606,
|
|
"num_tokens": 4397040.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 5.938760089874267,
|
|
"epoch": 0.19230769230769232,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004998834825191652,
|
|
"loss": 5.9508,
|
|
"mean_token_accuracy": 0.14106623604893684,
|
|
"num_tokens": 4406723.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 6.02097454071045,
|
|
"epoch": 0.1927612481857765,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004998824390193743,
|
|
"loss": 5.9982,
|
|
"mean_token_accuracy": 0.14898034185171127,
|
|
"num_tokens": 4416013.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 5.96734733581543,
|
|
"epoch": 0.19321480406386066,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004998813908689573,
|
|
"loss": 5.9793,
|
|
"mean_token_accuracy": 0.14236808717250823,
|
|
"num_tokens": 4426441.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 6.063247156143189,
|
|
"epoch": 0.19366835994194484,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004998803380679359,
|
|
"loss": 6.0948,
|
|
"mean_token_accuracy": 0.1332187458872795,
|
|
"num_tokens": 4437328.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 6.089685106277466,
|
|
"epoch": 0.19412191582002902,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004998792806163317,
|
|
"loss": 5.9181,
|
|
"mean_token_accuracy": 0.14083896055817605,
|
|
"num_tokens": 4447546.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 5.88359489440918,
|
|
"epoch": 0.1945754716981132,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004998782185141667,
|
|
"loss": 5.9561,
|
|
"mean_token_accuracy": 0.14243533834815025,
|
|
"num_tokens": 4457319.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 6.066173934936524,
|
|
"epoch": 0.1950290275761974,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004998771517614629,
|
|
"loss": 6.0624,
|
|
"mean_token_accuracy": 0.13416438773274422,
|
|
"num_tokens": 4468348.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 6.009496736526489,
|
|
"epoch": 0.19548258345428157,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004998760803582422,
|
|
"loss": 5.9612,
|
|
"mean_token_accuracy": 0.14601074978709222,
|
|
"num_tokens": 4478263.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 5.905720138549805,
|
|
"epoch": 0.19593613933236576,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004998750043045269,
|
|
"loss": 5.8817,
|
|
"mean_token_accuracy": 0.14295333474874497,
|
|
"num_tokens": 4488772.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 6.011665964126587,
|
|
"epoch": 0.19638969521044994,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004998739236003392,
|
|
"loss": 5.9905,
|
|
"mean_token_accuracy": 0.13327775821089743,
|
|
"num_tokens": 4499443.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 5.983055496215821,
|
|
"epoch": 0.1968432510885341,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004998728382457014,
|
|
"loss": 5.9876,
|
|
"mean_token_accuracy": 0.13847402930259706,
|
|
"num_tokens": 4509969.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 6.022305631637574,
|
|
"epoch": 0.19729680696661828,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004998717482406362,
|
|
"loss": 5.979,
|
|
"mean_token_accuracy": 0.14845634549856185,
|
|
"num_tokens": 4519668.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 5.9130627632141115,
|
|
"epoch": 0.19775036284470246,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004998706535851657,
|
|
"loss": 5.9578,
|
|
"mean_token_accuracy": 0.14601308181881906,
|
|
"num_tokens": 4530303.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 5.992128801345825,
|
|
"epoch": 0.19820391872278664,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004998695542793129,
|
|
"loss": 6.0187,
|
|
"mean_token_accuracy": 0.14158663749694825,
|
|
"num_tokens": 4541801.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 5.938469648361206,
|
|
"epoch": 0.19865747460087083,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004998684503231003,
|
|
"loss": 5.8341,
|
|
"mean_token_accuracy": 0.1426684334874153,
|
|
"num_tokens": 4551346.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 5.922765111923217,
|
|
"epoch": 0.199111030478955,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000499867341716551,
|
|
"loss": 5.915,
|
|
"mean_token_accuracy": 0.14817361980676652,
|
|
"num_tokens": 4561835.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 5.904284858703614,
|
|
"epoch": 0.1995645863570392,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004998662284596878,
|
|
"loss": 5.8667,
|
|
"mean_token_accuracy": 0.145841383934021,
|
|
"num_tokens": 4572760.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 5.936977243423462,
|
|
"epoch": 0.20001814223512338,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004998651105525335,
|
|
"loss": 5.9473,
|
|
"mean_token_accuracy": 0.14006382599473,
|
|
"num_tokens": 4582884.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 5.961916112899781,
|
|
"epoch": 0.20047169811320756,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004998639879951115,
|
|
"loss": 5.9315,
|
|
"mean_token_accuracy": 0.14331716150045395,
|
|
"num_tokens": 4594297.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 5.988901615142822,
|
|
"epoch": 0.2009252539912917,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000499862860787445,
|
|
"loss": 5.9907,
|
|
"mean_token_accuracy": 0.13747334852814674,
|
|
"num_tokens": 4605072.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 5.911744117736816,
|
|
"epoch": 0.2013788098693759,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004998617289295571,
|
|
"loss": 5.8751,
|
|
"mean_token_accuracy": 0.15403711050748825,
|
|
"num_tokens": 4615207.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 5.916146278381348,
|
|
"epoch": 0.20183236574746008,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004998605924214714,
|
|
"loss": 5.9039,
|
|
"mean_token_accuracy": 0.14326297491788864,
|
|
"num_tokens": 4625552.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 6.068982744216919,
|
|
"epoch": 0.20228592162554426,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004998594512632114,
|
|
"loss": 6.0052,
|
|
"mean_token_accuracy": 0.13947765678167343,
|
|
"num_tokens": 4635864.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 5.9241475582122805,
|
|
"epoch": 0.20273947750362845,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004998583054548006,
|
|
"loss": 5.9602,
|
|
"mean_token_accuracy": 0.13143666759133338,
|
|
"num_tokens": 4646978.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 5.949285650253296,
|
|
"epoch": 0.20319303338171263,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004998571549962628,
|
|
"loss": 5.8858,
|
|
"mean_token_accuracy": 0.14171508550643921,
|
|
"num_tokens": 4656737.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 5.886570167541504,
|
|
"epoch": 0.2036465892597968,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004998559998876216,
|
|
"loss": 6.0072,
|
|
"mean_token_accuracy": 0.14202161729335785,
|
|
"num_tokens": 4667663.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 5.985239171981812,
|
|
"epoch": 0.204100145137881,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004998548401289012,
|
|
"loss": 5.9673,
|
|
"mean_token_accuracy": 0.13756298944354056,
|
|
"num_tokens": 4678814.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 6.042722129821778,
|
|
"epoch": 0.20455370101596518,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004998536757201253,
|
|
"loss": 5.9821,
|
|
"mean_token_accuracy": 0.13774357065558435,
|
|
"num_tokens": 4689446.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 5.971980857849121,
|
|
"epoch": 0.20500725689404933,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004998525066613181,
|
|
"loss": 5.993,
|
|
"mean_token_accuracy": 0.13813588321208953,
|
|
"num_tokens": 4700207.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 6.003995275497436,
|
|
"epoch": 0.20546081277213352,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004998513329525038,
|
|
"loss": 5.9595,
|
|
"mean_token_accuracy": 0.1428474046289921,
|
|
"num_tokens": 4710743.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 5.981139850616455,
|
|
"epoch": 0.2059143686502177,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004998501545937066,
|
|
"loss": 5.9624,
|
|
"mean_token_accuracy": 0.14789393842220305,
|
|
"num_tokens": 4720964.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 5.965071296691894,
|
|
"epoch": 0.20636792452830188,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000499848971584951,
|
|
"loss": 6.001,
|
|
"mean_token_accuracy": 0.13834339007735252,
|
|
"num_tokens": 4732022.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 5.995220470428467,
|
|
"epoch": 0.20682148040638607,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004998477839262611,
|
|
"loss": 5.9789,
|
|
"mean_token_accuracy": 0.13753922730684282,
|
|
"num_tokens": 4743428.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 5.996651983261108,
|
|
"epoch": 0.20727503628447025,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000499846591617662,
|
|
"loss": 5.9059,
|
|
"mean_token_accuracy": 0.14165481328964233,
|
|
"num_tokens": 4754530.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 5.848381853103637,
|
|
"epoch": 0.20772859216255443,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004998453946591779,
|
|
"loss": 5.8387,
|
|
"mean_token_accuracy": 0.14721979647874833,
|
|
"num_tokens": 4764453.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 6.030099821090698,
|
|
"epoch": 0.20818214804063861,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004998441930508339,
|
|
"loss": 6.0122,
|
|
"mean_token_accuracy": 0.14176427572965622,
|
|
"num_tokens": 4774788.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 5.991954565048218,
|
|
"epoch": 0.2086357039187228,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004998429867926547,
|
|
"loss": 5.9593,
|
|
"mean_token_accuracy": 0.1358835816383362,
|
|
"num_tokens": 4785827.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 5.9251960754394535,
|
|
"epoch": 0.20908925979680695,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004998417758846651,
|
|
"loss": 5.836,
|
|
"mean_token_accuracy": 0.15004699677228928,
|
|
"num_tokens": 4795683.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 5.960970211029053,
|
|
"epoch": 0.20954281567489114,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004998405603268903,
|
|
"loss": 5.9322,
|
|
"mean_token_accuracy": 0.14288516640663146,
|
|
"num_tokens": 4805929.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 5.915129613876343,
|
|
"epoch": 0.20999637155297532,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004998393401193554,
|
|
"loss": 5.9297,
|
|
"mean_token_accuracy": 0.14064979255199433,
|
|
"num_tokens": 4816067.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 5.940646457672119,
|
|
"epoch": 0.2104499274310595,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004998381152620857,
|
|
"loss": 5.9153,
|
|
"mean_token_accuracy": 0.1419698029756546,
|
|
"num_tokens": 4826757.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 5.9893388748168945,
|
|
"epoch": 0.21090348330914369,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004998368857551067,
|
|
"loss": 5.9248,
|
|
"mean_token_accuracy": 0.14424414560198784,
|
|
"num_tokens": 4836548.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 6.0586730480194095,
|
|
"epoch": 0.21135703918722787,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004998356515984433,
|
|
"loss": 5.9785,
|
|
"mean_token_accuracy": 0.13741380125284194,
|
|
"num_tokens": 4847274.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 5.959125995635986,
|
|
"epoch": 0.21181059506531205,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004998344127921213,
|
|
"loss": 5.9483,
|
|
"mean_token_accuracy": 0.1400488071143627,
|
|
"num_tokens": 4857506.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 5.960813760757446,
|
|
"epoch": 0.21226415094339623,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004998331693361666,
|
|
"loss": 5.9773,
|
|
"mean_token_accuracy": 0.1386716030538082,
|
|
"num_tokens": 4868069.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 5.946921539306641,
|
|
"epoch": 0.21271770682148042,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004998319212306046,
|
|
"loss": 5.8775,
|
|
"mean_token_accuracy": 0.14960167706012725,
|
|
"num_tokens": 4877731.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 5.923819208145142,
|
|
"epoch": 0.21317126269956457,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000499830668475461,
|
|
"loss": 5.9369,
|
|
"mean_token_accuracy": 0.13987276405096055,
|
|
"num_tokens": 4887780.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 5.958174419403076,
|
|
"epoch": 0.21362481857764876,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000499829411070762,
|
|
"loss": 5.9577,
|
|
"mean_token_accuracy": 0.14128173291683196,
|
|
"num_tokens": 4897578.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 5.871128082275391,
|
|
"epoch": 0.21407837445573294,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004998281490165335,
|
|
"loss": 5.9053,
|
|
"mean_token_accuracy": 0.13984225764870645,
|
|
"num_tokens": 4908187.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 6.030574989318848,
|
|
"epoch": 0.21453193033381712,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004998268823128016,
|
|
"loss": 5.9289,
|
|
"mean_token_accuracy": 0.14696599096059798,
|
|
"num_tokens": 4918352.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 5.877168655395508,
|
|
"epoch": 0.2149854862119013,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004998256109595923,
|
|
"loss": 5.8947,
|
|
"mean_token_accuracy": 0.14657547250390052,
|
|
"num_tokens": 4928065.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 5.952869129180908,
|
|
"epoch": 0.2154390420899855,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004998243349569323,
|
|
"loss": 5.9084,
|
|
"mean_token_accuracy": 0.14537300616502763,
|
|
"num_tokens": 4939492.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 5.872677612304687,
|
|
"epoch": 0.21589259796806967,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004998230543048477,
|
|
"loss": 5.8546,
|
|
"mean_token_accuracy": 0.14155039489269255,
|
|
"num_tokens": 4949304.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 5.894457006454468,
|
|
"epoch": 0.21634615384615385,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000499821769003365,
|
|
"loss": 5.8853,
|
|
"mean_token_accuracy": 0.1352695107460022,
|
|
"num_tokens": 4959649.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 5.9166566848754885,
|
|
"epoch": 0.21679970972423804,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004998204790525109,
|
|
"loss": 5.9306,
|
|
"mean_token_accuracy": 0.13895038962364198,
|
|
"num_tokens": 4970029.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 5.966705751419068,
|
|
"epoch": 0.2172532656023222,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004998191844523119,
|
|
"loss": 5.8501,
|
|
"mean_token_accuracy": 0.14436866864562034,
|
|
"num_tokens": 4980431.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 5.862818002700806,
|
|
"epoch": 0.21770682148040638,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000499817885202795,
|
|
"loss": 5.9549,
|
|
"mean_token_accuracy": 0.14049403220415116,
|
|
"num_tokens": 4991448.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 5.938876914978027,
|
|
"epoch": 0.21816037735849056,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004998165813039869,
|
|
"loss": 5.8976,
|
|
"mean_token_accuracy": 0.1356552518904209,
|
|
"num_tokens": 5001225.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 5.924232530593872,
|
|
"epoch": 0.21861393323657474,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004998152727559145,
|
|
"loss": 5.8808,
|
|
"mean_token_accuracy": 0.14155457615852357,
|
|
"num_tokens": 5012096.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 5.954684495925903,
|
|
"epoch": 0.21906748911465893,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000499813959558605,
|
|
"loss": 5.9579,
|
|
"mean_token_accuracy": 0.13705262392759324,
|
|
"num_tokens": 5021675.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 5.930799150466919,
|
|
"epoch": 0.2195210449927431,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004998126417120856,
|
|
"loss": 5.9114,
|
|
"mean_token_accuracy": 0.1381124421954155,
|
|
"num_tokens": 5032250.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 5.906846952438355,
|
|
"epoch": 0.2199746008708273,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004998113192163835,
|
|
"loss": 5.9758,
|
|
"mean_token_accuracy": 0.13671192824840545,
|
|
"num_tokens": 5042508.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 6.031311511993408,
|
|
"epoch": 0.22042815674891147,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000499809992071526,
|
|
"loss": 5.8645,
|
|
"mean_token_accuracy": 0.14089929684996605,
|
|
"num_tokens": 5052765.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 5.894246578216553,
|
|
"epoch": 0.22088171262699566,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004998086602775406,
|
|
"loss": 5.8884,
|
|
"mean_token_accuracy": 0.14432070553302764,
|
|
"num_tokens": 5062379.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 5.954690265655517,
|
|
"epoch": 0.2213352685050798,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004998073238344547,
|
|
"loss": 5.9271,
|
|
"mean_token_accuracy": 0.14122223407030104,
|
|
"num_tokens": 5072267.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 5.857289552688599,
|
|
"epoch": 0.221788824383164,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004998059827422962,
|
|
"loss": 5.8285,
|
|
"mean_token_accuracy": 0.1439412847161293,
|
|
"num_tokens": 5082214.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 5.875279474258423,
|
|
"epoch": 0.22224238026124818,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004998046370010926,
|
|
"loss": 5.8849,
|
|
"mean_token_accuracy": 0.14643458500504494,
|
|
"num_tokens": 5092129.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 5.914616823196411,
|
|
"epoch": 0.22269593613933236,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004998032866108718,
|
|
"loss": 5.903,
|
|
"mean_token_accuracy": 0.13629107251763345,
|
|
"num_tokens": 5102924.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 5.9880273818969725,
|
|
"epoch": 0.22314949201741655,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004998019315716618,
|
|
"loss": 5.8787,
|
|
"mean_token_accuracy": 0.1420026421546936,
|
|
"num_tokens": 5113216.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 5.792202377319336,
|
|
"epoch": 0.22360304789550073,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004998005718834905,
|
|
"loss": 5.8578,
|
|
"mean_token_accuracy": 0.1403028056025505,
|
|
"num_tokens": 5122978.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 5.964217710494995,
|
|
"epoch": 0.2240566037735849,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004997992075463862,
|
|
"loss": 5.9519,
|
|
"mean_token_accuracy": 0.14088138267397882,
|
|
"num_tokens": 5133167.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 5.898852062225342,
|
|
"epoch": 0.2245101596516691,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004997978385603769,
|
|
"loss": 5.8241,
|
|
"mean_token_accuracy": 0.14743937328457832,
|
|
"num_tokens": 5143848.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 5.9200560569763185,
|
|
"epoch": 0.22496371552975328,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004997964649254911,
|
|
"loss": 5.8611,
|
|
"mean_token_accuracy": 0.15123707801103592,
|
|
"num_tokens": 5154172.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 5.81275691986084,
|
|
"epoch": 0.22541727140783743,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004997950866417572,
|
|
"loss": 5.8186,
|
|
"mean_token_accuracy": 0.1486186221241951,
|
|
"num_tokens": 5164459.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 5.979988098144531,
|
|
"epoch": 0.22587082728592162,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004997937037092034,
|
|
"loss": 5.9206,
|
|
"mean_token_accuracy": 0.13727832362055778,
|
|
"num_tokens": 5175751.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 5.904782056808472,
|
|
"epoch": 0.2263243831640058,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004997923161278587,
|
|
"loss": 5.836,
|
|
"mean_token_accuracy": 0.1466525003314018,
|
|
"num_tokens": 5186004.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 5.8142805099487305,
|
|
"epoch": 0.22677793904208998,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004997909238977514,
|
|
"loss": 5.7535,
|
|
"mean_token_accuracy": 0.1469535768032074,
|
|
"num_tokens": 5195610.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 5.915460872650146,
|
|
"epoch": 0.22723149492017417,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004997895270189108,
|
|
"loss": 5.9106,
|
|
"mean_token_accuracy": 0.1398490011692047,
|
|
"num_tokens": 5205607.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 5.935828447341919,
|
|
"epoch": 0.22768505079825835,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004997881254913653,
|
|
"loss": 5.8928,
|
|
"mean_token_accuracy": 0.1415001295506954,
|
|
"num_tokens": 5217372.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 5.895864343643188,
|
|
"epoch": 0.22813860667634253,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004997867193151441,
|
|
"loss": 5.8553,
|
|
"mean_token_accuracy": 0.14375398457050323,
|
|
"num_tokens": 5226782.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 5.875649499893188,
|
|
"epoch": 0.22859216255442671,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004997853084902763,
|
|
"loss": 5.7915,
|
|
"mean_token_accuracy": 0.1496797136962414,
|
|
"num_tokens": 5236209.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 5.870350313186646,
|
|
"epoch": 0.2290457184325109,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004997838930167911,
|
|
"loss": 5.8964,
|
|
"mean_token_accuracy": 0.14222536683082582,
|
|
"num_tokens": 5246041.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 5.970703125,
|
|
"epoch": 0.22949927431059505,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004997824728947176,
|
|
"loss": 5.8557,
|
|
"mean_token_accuracy": 0.14660589545965194,
|
|
"num_tokens": 5256902.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 5.928110027313233,
|
|
"epoch": 0.22995283018867924,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004997810481240853,
|
|
"loss": 5.8547,
|
|
"mean_token_accuracy": 0.14946367517113684,
|
|
"num_tokens": 5266180.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 5.936565351486206,
|
|
"epoch": 0.23040638606676342,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004997796187049238,
|
|
"loss": 6.0483,
|
|
"mean_token_accuracy": 0.13646556437015533,
|
|
"num_tokens": 5277475.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 6.029916381835937,
|
|
"epoch": 0.2308599419448476,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004997781846372623,
|
|
"loss": 5.8693,
|
|
"mean_token_accuracy": 0.14439473375678064,
|
|
"num_tokens": 5287108.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 5.854767751693726,
|
|
"epoch": 0.23131349782293179,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004997767459211307,
|
|
"loss": 5.9654,
|
|
"mean_token_accuracy": 0.13685449138283728,
|
|
"num_tokens": 5297598.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 6.062006139755249,
|
|
"epoch": 0.23176705370101597,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004997753025565587,
|
|
"loss": 5.9442,
|
|
"mean_token_accuracy": 0.1358692966401577,
|
|
"num_tokens": 5306953.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 5.887425899505615,
|
|
"epoch": 0.23222060957910015,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004997738545435763,
|
|
"loss": 5.9399,
|
|
"mean_token_accuracy": 0.15026361793279647,
|
|
"num_tokens": 5317608.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 5.862730932235718,
|
|
"epoch": 0.23267416545718433,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004997724018822133,
|
|
"loss": 5.8058,
|
|
"mean_token_accuracy": 0.13242100477218627,
|
|
"num_tokens": 5328503.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 5.914726114273071,
|
|
"epoch": 0.23312772133526852,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004997709445724996,
|
|
"loss": 5.883,
|
|
"mean_token_accuracy": 0.14427884221076964,
|
|
"num_tokens": 5338108.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 5.95159878730774,
|
|
"epoch": 0.23358127721335267,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004997694826144656,
|
|
"loss": 6.04,
|
|
"mean_token_accuracy": 0.13466350510716438,
|
|
"num_tokens": 5349890.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 6.051620054244995,
|
|
"epoch": 0.23403483309143686,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004997680160081414,
|
|
"loss": 6.0118,
|
|
"mean_token_accuracy": 0.1407109409570694,
|
|
"num_tokens": 5361319.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 5.919955205917359,
|
|
"epoch": 0.23448838896952104,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004997665447535573,
|
|
"loss": 5.9219,
|
|
"mean_token_accuracy": 0.14855802059173584,
|
|
"num_tokens": 5370932.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 5.9046718120574955,
|
|
"epoch": 0.23494194484760522,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004997650688507438,
|
|
"loss": 5.843,
|
|
"mean_token_accuracy": 0.14306004494428634,
|
|
"num_tokens": 5381661.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 5.818372058868408,
|
|
"epoch": 0.2353955007256894,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004997635882997314,
|
|
"loss": 5.8143,
|
|
"mean_token_accuracy": 0.14963757395744323,
|
|
"num_tokens": 5390999.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 5.943971872329712,
|
|
"epoch": 0.2358490566037736,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004997621031005507,
|
|
"loss": 5.9085,
|
|
"mean_token_accuracy": 0.14275056049227713,
|
|
"num_tokens": 5401175.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 5.954121685028076,
|
|
"epoch": 0.23630261248185777,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004997606132532325,
|
|
"loss": 5.8747,
|
|
"mean_token_accuracy": 0.14469726160168647,
|
|
"num_tokens": 5411644.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 5.812447595596313,
|
|
"epoch": 0.23675616835994195,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004997591187578075,
|
|
"loss": 5.8496,
|
|
"mean_token_accuracy": 0.14814821630716324,
|
|
"num_tokens": 5420923.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 5.941586923599243,
|
|
"epoch": 0.23720972423802614,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004997576196143067,
|
|
"loss": 5.8974,
|
|
"mean_token_accuracy": 0.14263342320919037,
|
|
"num_tokens": 5431475.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 5.9011571407318115,
|
|
"epoch": 0.2376632801161103,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000499756115822761,
|
|
"loss": 5.8093,
|
|
"mean_token_accuracy": 0.14849388152360915,
|
|
"num_tokens": 5441822.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 5.8005060195922855,
|
|
"epoch": 0.23811683599419448,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004997546073832015,
|
|
"loss": 5.8219,
|
|
"mean_token_accuracy": 0.13821850046515466,
|
|
"num_tokens": 5453266.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 5.917859411239624,
|
|
"epoch": 0.23857039187227866,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004997530942956595,
|
|
"loss": 5.8634,
|
|
"mean_token_accuracy": 0.14885092675685882,
|
|
"num_tokens": 5463023.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 5.934016132354737,
|
|
"epoch": 0.23902394775036284,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004997515765601662,
|
|
"loss": 5.8566,
|
|
"mean_token_accuracy": 0.14655044227838515,
|
|
"num_tokens": 5472345.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 5.923844289779663,
|
|
"epoch": 0.23947750362844702,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004997500541767531,
|
|
"loss": 5.9059,
|
|
"mean_token_accuracy": 0.1498401679098606,
|
|
"num_tokens": 5481468.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 5.773250436782837,
|
|
"epoch": 0.2399310595065312,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004997485271454516,
|
|
"loss": 5.8461,
|
|
"mean_token_accuracy": 0.14630547761917115,
|
|
"num_tokens": 5492107.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 5.871908712387085,
|
|
"epoch": 0.2403846153846154,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004997469954662933,
|
|
"loss": 5.8771,
|
|
"mean_token_accuracy": 0.1383388437330723,
|
|
"num_tokens": 5502196.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 5.904657554626465,
|
|
"epoch": 0.24083817126269957,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004997454591393097,
|
|
"loss": 5.8693,
|
|
"mean_token_accuracy": 0.1385428823530674,
|
|
"num_tokens": 5512859.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 5.975746059417725,
|
|
"epoch": 0.24129172714078376,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004997439181645329,
|
|
"loss": 5.8498,
|
|
"mean_token_accuracy": 0.14605885297060012,
|
|
"num_tokens": 5523041.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 5.764354801177978,
|
|
"epoch": 0.2417452830188679,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004997423725419945,
|
|
"loss": 5.761,
|
|
"mean_token_accuracy": 0.15253512784838677,
|
|
"num_tokens": 5533967.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 5.947396564483642,
|
|
"epoch": 0.2421988388969521,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004997408222717267,
|
|
"loss": 5.8664,
|
|
"mean_token_accuracy": 0.14548483043909072,
|
|
"num_tokens": 5543460.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 5.93350305557251,
|
|
"epoch": 0.24265239477503628,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004997392673537612,
|
|
"loss": 5.8592,
|
|
"mean_token_accuracy": 0.1520221307873726,
|
|
"num_tokens": 5553139.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 5.793046188354492,
|
|
"epoch": 0.24310595065312046,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004997377077881306,
|
|
"loss": 5.8541,
|
|
"mean_token_accuracy": 0.1441502094268799,
|
|
"num_tokens": 5564550.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 5.9918012619018555,
|
|
"epoch": 0.24355950653120464,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004997361435748668,
|
|
"loss": 5.9546,
|
|
"mean_token_accuracy": 0.1459794595837593,
|
|
"num_tokens": 5574393.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 5.940083932876587,
|
|
"epoch": 0.24401306240928883,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004997345747140024,
|
|
"loss": 5.8813,
|
|
"mean_token_accuracy": 0.14294293597340585,
|
|
"num_tokens": 5584206.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 5.877155923843384,
|
|
"epoch": 0.244466618287373,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004997330012055695,
|
|
"loss": 5.8873,
|
|
"mean_token_accuracy": 0.1417349547147751,
|
|
"num_tokens": 5594946.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 5.921493816375732,
|
|
"epoch": 0.2449201741654572,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004997314230496011,
|
|
"loss": 5.7744,
|
|
"mean_token_accuracy": 0.14902272820472717,
|
|
"num_tokens": 5605984.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 5.776959466934204,
|
|
"epoch": 0.24537373004354138,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004997298402461296,
|
|
"loss": 5.8131,
|
|
"mean_token_accuracy": 0.1488720089197159,
|
|
"num_tokens": 5616227.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 5.935484266281128,
|
|
"epoch": 0.24582728592162553,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004997282527951876,
|
|
"loss": 5.9032,
|
|
"mean_token_accuracy": 0.14310448318719865,
|
|
"num_tokens": 5626592.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 6.0017767429351805,
|
|
"epoch": 0.24628084179970972,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0004997266606968081,
|
|
"loss": 5.8379,
|
|
"mean_token_accuracy": 0.14523325935006143,
|
|
"num_tokens": 5636486.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 5.846785545349121,
|
|
"epoch": 0.2467343976777939,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004997250639510241,
|
|
"loss": 5.9128,
|
|
"mean_token_accuracy": 0.14552611336112023,
|
|
"num_tokens": 5646025.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 5.828639936447144,
|
|
"epoch": 0.24718795355587808,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004997234625578684,
|
|
"loss": 5.7968,
|
|
"mean_token_accuracy": 0.1471302717924118,
|
|
"num_tokens": 5656523.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 5.8145918369293215,
|
|
"epoch": 0.24764150943396226,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004997218565173742,
|
|
"loss": 5.8303,
|
|
"mean_token_accuracy": 0.14437200129032135,
|
|
"num_tokens": 5667598.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 5.928363561630249,
|
|
"epoch": 0.24809506531204645,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004997202458295748,
|
|
"loss": 5.8606,
|
|
"mean_token_accuracy": 0.14459095299243926,
|
|
"num_tokens": 5677721.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 5.847347164154053,
|
|
"epoch": 0.24854862119013063,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004997186304945034,
|
|
"loss": 5.8416,
|
|
"mean_token_accuracy": 0.1417038306593895,
|
|
"num_tokens": 5688450.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 5.865340757369995,
|
|
"epoch": 0.24900217706821481,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004997170105121935,
|
|
"loss": 5.8208,
|
|
"mean_token_accuracy": 0.1495169907808304,
|
|
"num_tokens": 5698142.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 5.833406114578247,
|
|
"epoch": 0.249455732946299,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004997153858826786,
|
|
"loss": 5.8201,
|
|
"mean_token_accuracy": 0.1465844228863716,
|
|
"num_tokens": 5708122.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 5.837479496002198,
|
|
"epoch": 0.24990928882438315,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004997137566059922,
|
|
"loss": 5.8107,
|
|
"mean_token_accuracy": 0.14600400999188423,
|
|
"num_tokens": 5718671.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 5.8692303657531735,
|
|
"epoch": 0.25036284470246734,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004997121226821681,
|
|
"loss": 5.8171,
|
|
"mean_token_accuracy": 0.14405162036418914,
|
|
"num_tokens": 5729463.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 5.922516870498657,
|
|
"epoch": 0.25081640058055155,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00049971048411124,
|
|
"loss": 5.8874,
|
|
"mean_token_accuracy": 0.14321258664131165,
|
|
"num_tokens": 5739212.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 5.901452112197876,
|
|
"epoch": 0.2512699564586357,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004997088408932418,
|
|
"loss": 5.8029,
|
|
"mean_token_accuracy": 0.14497288316488266,
|
|
"num_tokens": 5749159.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 5.932657766342163,
|
|
"epoch": 0.25172351233671986,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004997071930282076,
|
|
"loss": 5.993,
|
|
"mean_token_accuracy": 0.13827754259109498,
|
|
"num_tokens": 5759711.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 6.124020576477051,
|
|
"epoch": 0.25217706821480407,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004997055405161712,
|
|
"loss": 6.0043,
|
|
"mean_token_accuracy": 0.1372769333422184,
|
|
"num_tokens": 5771210.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 5.816952276229858,
|
|
"epoch": 0.2526306240928882,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004997038833571671,
|
|
"loss": 5.8426,
|
|
"mean_token_accuracy": 0.14726067781448365,
|
|
"num_tokens": 5780960.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 5.935285472869873,
|
|
"epoch": 0.25308417997097243,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004997022215512295,
|
|
"loss": 5.885,
|
|
"mean_token_accuracy": 0.14095496237277985,
|
|
"num_tokens": 5790891.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 5.894222116470337,
|
|
"epoch": 0.2535377358490566,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004997005550983925,
|
|
"loss": 5.8396,
|
|
"mean_token_accuracy": 0.14395667985081673,
|
|
"num_tokens": 5801078.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 5.885415887832641,
|
|
"epoch": 0.2539912917271408,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.000499698883998691,
|
|
"loss": 5.9003,
|
|
"mean_token_accuracy": 0.1393453024327755,
|
|
"num_tokens": 5811025.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 5.909905290603637,
|
|
"epoch": 0.25444484760522496,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004996972082521591,
|
|
"loss": 5.8086,
|
|
"mean_token_accuracy": 0.1448357567191124,
|
|
"num_tokens": 5821242.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 5.845452737808228,
|
|
"epoch": 0.25489840348330917,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004996955278588319,
|
|
"loss": 5.8592,
|
|
"mean_token_accuracy": 0.14133516177535058,
|
|
"num_tokens": 5832039.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 5.914500284194946,
|
|
"epoch": 0.2553519593613933,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004996938428187437,
|
|
"loss": 5.8397,
|
|
"mean_token_accuracy": 0.1460021823644638,
|
|
"num_tokens": 5841723.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 5.9476793766021725,
|
|
"epoch": 0.2558055152394775,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004996921531319297,
|
|
"loss": 5.8873,
|
|
"mean_token_accuracy": 0.14610021114349364,
|
|
"num_tokens": 5852814.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 5.826781702041626,
|
|
"epoch": 0.2562590711175617,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004996904587984246,
|
|
"loss": 5.8323,
|
|
"mean_token_accuracy": 0.14333641827106475,
|
|
"num_tokens": 5863174.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 5.853058385848999,
|
|
"epoch": 0.25671262699564584,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004996887598182636,
|
|
"loss": 5.7837,
|
|
"mean_token_accuracy": 0.14737775102257727,
|
|
"num_tokens": 5873698.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 5.760656118392944,
|
|
"epoch": 0.25716618287373005,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004996870561914819,
|
|
"loss": 5.8211,
|
|
"mean_token_accuracy": 0.1456883355975151,
|
|
"num_tokens": 5883982.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 5.894684791564941,
|
|
"epoch": 0.2576197387518142,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004996853479181145,
|
|
"loss": 5.7891,
|
|
"mean_token_accuracy": 0.14917163103818892,
|
|
"num_tokens": 5894454.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 5.844020223617553,
|
|
"epoch": 0.2580732946298984,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004996836349981969,
|
|
"loss": 5.8806,
|
|
"mean_token_accuracy": 0.1445062406361103,
|
|
"num_tokens": 5904739.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 5.931279706954956,
|
|
"epoch": 0.2585268505079826,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004996819174317645,
|
|
"loss": 5.8601,
|
|
"mean_token_accuracy": 0.14103500097990035,
|
|
"num_tokens": 5914578.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 5.924049043655396,
|
|
"epoch": 0.2589804063860668,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004996801952188528,
|
|
"loss": 5.8417,
|
|
"mean_token_accuracy": 0.1458780586719513,
|
|
"num_tokens": 5923918.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 5.938410520553589,
|
|
"epoch": 0.25943396226415094,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004996784683594973,
|
|
"loss": 5.8868,
|
|
"mean_token_accuracy": 0.14107644706964492,
|
|
"num_tokens": 5934096.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 5.885492944717408,
|
|
"epoch": 0.2598875181422351,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004996767368537338,
|
|
"loss": 5.9094,
|
|
"mean_token_accuracy": 0.1389304779469967,
|
|
"num_tokens": 5945416.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 5.875432109832763,
|
|
"epoch": 0.2603410740203193,
|
|
"grad_norm": 3.046875,
|
|
"learning_rate": 0.0004996750007015983,
|
|
"loss": 5.8406,
|
|
"mean_token_accuracy": 0.14503862857818603,
|
|
"num_tokens": 5956296.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 5.864486980438232,
|
|
"epoch": 0.26079462989840346,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004996732599031264,
|
|
"loss": 5.733,
|
|
"mean_token_accuracy": 0.15628641992807388,
|
|
"num_tokens": 5966147.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 5.841119432449341,
|
|
"epoch": 0.2612481857764877,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004996715144583543,
|
|
"loss": 5.8385,
|
|
"mean_token_accuracy": 0.15209657400846482,
|
|
"num_tokens": 5976650.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 5.918705177307129,
|
|
"epoch": 0.26170174165457183,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000499669764367318,
|
|
"loss": 5.9027,
|
|
"mean_token_accuracy": 0.14683974608778955,
|
|
"num_tokens": 5986643.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 5.891564416885376,
|
|
"epoch": 0.26215529753265604,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004996680096300537,
|
|
"loss": 5.8697,
|
|
"mean_token_accuracy": 0.14015998244285582,
|
|
"num_tokens": 5997991.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 5.783881521224975,
|
|
"epoch": 0.2626088534107402,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004996662502465978,
|
|
"loss": 5.6945,
|
|
"mean_token_accuracy": 0.15556951612234116,
|
|
"num_tokens": 6007179.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 5.919160652160644,
|
|
"epoch": 0.2630624092888244,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004996644862169864,
|
|
"loss": 5.8789,
|
|
"mean_token_accuracy": 0.14572433680295943,
|
|
"num_tokens": 6017786.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 5.915647411346436,
|
|
"epoch": 0.26351596516690856,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0004996627175412563,
|
|
"loss": 5.9004,
|
|
"mean_token_accuracy": 0.14329932779073715,
|
|
"num_tokens": 6027751.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 5.8271812915802,
|
|
"epoch": 0.2639695210449927,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.000499660944219444,
|
|
"loss": 5.7774,
|
|
"mean_token_accuracy": 0.14923276603221894,
|
|
"num_tokens": 6037136.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 5.790592765808105,
|
|
"epoch": 0.2644230769230769,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004996591662515861,
|
|
"loss": 5.7695,
|
|
"mean_token_accuracy": 0.14568121135234832,
|
|
"num_tokens": 6047341.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 5.903992128372193,
|
|
"epoch": 0.2648766328011611,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004996573836377193,
|
|
"loss": 5.9207,
|
|
"mean_token_accuracy": 0.14674188792705536,
|
|
"num_tokens": 6058337.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 5.892431545257568,
|
|
"epoch": 0.2653301886792453,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004996555963778806,
|
|
"loss": 5.8044,
|
|
"mean_token_accuracy": 0.14313996583223343,
|
|
"num_tokens": 6068566.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 5.918859958648682,
|
|
"epoch": 0.26578374455732945,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000499653804472107,
|
|
"loss": 5.8215,
|
|
"mean_token_accuracy": 0.15070558488368987,
|
|
"num_tokens": 6078813.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 5.816320133209229,
|
|
"epoch": 0.26623730043541366,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004996520079204353,
|
|
"loss": 5.8376,
|
|
"mean_token_accuracy": 0.15071689635515212,
|
|
"num_tokens": 6090007.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 5.924297523498535,
|
|
"epoch": 0.2666908563134978,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000499650206722903,
|
|
"loss": 5.9844,
|
|
"mean_token_accuracy": 0.1373060829937458,
|
|
"num_tokens": 6100917.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 6.015955924987793,
|
|
"epoch": 0.267144412191582,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004996484008795471,
|
|
"loss": 5.93,
|
|
"mean_token_accuracy": 0.147200308740139,
|
|
"num_tokens": 6110958.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 5.904809999465942,
|
|
"epoch": 0.2675979680696662,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000499646590390405,
|
|
"loss": 5.8316,
|
|
"mean_token_accuracy": 0.14588915333151817,
|
|
"num_tokens": 6120819.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 5.836217546463013,
|
|
"epoch": 0.26805152394775034,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004996447752555142,
|
|
"loss": 5.8205,
|
|
"mean_token_accuracy": 0.14316449761390687,
|
|
"num_tokens": 6131253.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 5.898743295669556,
|
|
"epoch": 0.26850507982583455,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004996429554749122,
|
|
"loss": 5.8484,
|
|
"mean_token_accuracy": 0.148350540548563,
|
|
"num_tokens": 6141338.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 5.846032381057739,
|
|
"epoch": 0.2689586357039187,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004996411310486367,
|
|
"loss": 5.8693,
|
|
"mean_token_accuracy": 0.14200861528515815,
|
|
"num_tokens": 6151081.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 5.9468964576721195,
|
|
"epoch": 0.2694121915820029,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004996393019767253,
|
|
"loss": 5.8606,
|
|
"mean_token_accuracy": 0.13914820924401283,
|
|
"num_tokens": 6160997.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 5.878990554809571,
|
|
"epoch": 0.26986574746008707,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004996374682592158,
|
|
"loss": 5.7723,
|
|
"mean_token_accuracy": 0.14771230965852739,
|
|
"num_tokens": 6171131.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 5.830885028839111,
|
|
"epoch": 0.2703193033381713,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004996356298961463,
|
|
"loss": 5.8911,
|
|
"mean_token_accuracy": 0.14644451886415483,
|
|
"num_tokens": 6181307.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 5.935413074493408,
|
|
"epoch": 0.27077285921625543,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004996337868875549,
|
|
"loss": 5.7345,
|
|
"mean_token_accuracy": 0.15274717211723327,
|
|
"num_tokens": 6191689.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 5.780738258361817,
|
|
"epoch": 0.27122641509433965,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004996319392334792,
|
|
"loss": 5.8039,
|
|
"mean_token_accuracy": 0.14987841248512268,
|
|
"num_tokens": 6201183.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 5.876131439208985,
|
|
"epoch": 0.2716799709724238,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.000499630086933958,
|
|
"loss": 5.897,
|
|
"mean_token_accuracy": 0.14560322165489198,
|
|
"num_tokens": 6211507.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 5.882185268402099,
|
|
"epoch": 0.27213352685050796,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004996282299890292,
|
|
"loss": 5.758,
|
|
"mean_token_accuracy": 0.15807864367961882,
|
|
"num_tokens": 6220899.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 0.27213352685050796,
|
|
"eval_entropy": 5.707925649789663,
|
|
"eval_loss": 5.863901615142822,
|
|
"eval_mean_token_accuracy": 0.15150013840356968,
|
|
"eval_num_tokens": 6220899.0,
|
|
"eval_runtime": 26.8067,
|
|
"eval_samples_per_second": 1319.074,
|
|
"eval_steps_per_second": 164.884,
|
|
"step": 3000
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 110230,
|
|
"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": 9148397451264000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|