506b135ec4
Model: fpadovani/rus-cyrl-10mb-after-ppt-Dp-10mb-ckpt500_seed10 Source: Original Platform
4079 lines
110 KiB
JSON
4079 lines
110 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 1.7749667110519307,
|
|
"eval_steps": 500,
|
|
"global_step": 2000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 7.78931941986084,
|
|
"epoch": 0.004438526409232135,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 2e-06,
|
|
"loss": 7.4769,
|
|
"mean_token_accuracy": 0.08522442281246186,
|
|
"num_tokens": 10699.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 7.787262725830078,
|
|
"epoch": 0.00887705281846427,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 7.4824,
|
|
"mean_token_accuracy": 0.08313449285924435,
|
|
"num_tokens": 20868.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 7.764151668548584,
|
|
"epoch": 0.013315579227696404,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 7e-06,
|
|
"loss": 7.4352,
|
|
"mean_token_accuracy": 0.08592776954174042,
|
|
"num_tokens": 31565.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 7.8020752429962155,
|
|
"epoch": 0.01775410563692854,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 7.5052,
|
|
"mean_token_accuracy": 0.08421143889427185,
|
|
"num_tokens": 42396.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 7.809156513214111,
|
|
"epoch": 0.022192632046160676,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 7.5029,
|
|
"mean_token_accuracy": 0.08358340412378311,
|
|
"num_tokens": 52422.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 7.804666757583618,
|
|
"epoch": 0.02663115845539281,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 7.4415,
|
|
"mean_token_accuracy": 0.0846289798617363,
|
|
"num_tokens": 63167.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 7.820871639251709,
|
|
"epoch": 0.031069684864624945,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 7.4654,
|
|
"mean_token_accuracy": 0.08765448182821274,
|
|
"num_tokens": 73620.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 7.827526521682739,
|
|
"epoch": 0.03550821127385708,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 7.3795,
|
|
"mean_token_accuracy": 0.08707336336374283,
|
|
"num_tokens": 83176.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 7.789561748504639,
|
|
"epoch": 0.03994673768308921,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 7.435,
|
|
"mean_token_accuracy": 0.0886497862637043,
|
|
"num_tokens": 94238.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 7.7829310417175295,
|
|
"epoch": 0.04438526409232135,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 7.4332,
|
|
"mean_token_accuracy": 0.08750618845224381,
|
|
"num_tokens": 104855.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 7.799046611785888,
|
|
"epoch": 0.048823790501553485,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 7.4366,
|
|
"mean_token_accuracy": 0.09203671216964722,
|
|
"num_tokens": 113812.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 7.795975160598755,
|
|
"epoch": 0.05326231691078562,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 7.4903,
|
|
"mean_token_accuracy": 0.08637641668319702,
|
|
"num_tokens": 123392.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 7.796218681335449,
|
|
"epoch": 0.05770084332001776,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 7.4831,
|
|
"mean_token_accuracy": 0.08530961498618125,
|
|
"num_tokens": 132764.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 7.810373401641845,
|
|
"epoch": 0.06213936972924989,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 7.434,
|
|
"mean_token_accuracy": 0.08844730108976365,
|
|
"num_tokens": 142965.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 7.803947973251343,
|
|
"epoch": 0.06657789613848203,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 7.4512,
|
|
"mean_token_accuracy": 0.08455823883414268,
|
|
"num_tokens": 152939.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 7.8467789649963375,
|
|
"epoch": 0.07101642254771416,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 7.5052,
|
|
"mean_token_accuracy": 0.08405138589441777,
|
|
"num_tokens": 164455.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 7.778741502761841,
|
|
"epoch": 0.0754549489569463,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 7.4599,
|
|
"mean_token_accuracy": 0.08571180626749993,
|
|
"num_tokens": 174110.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 7.789217472076416,
|
|
"epoch": 0.07989347536617843,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 7.4293,
|
|
"mean_token_accuracy": 0.0909264013171196,
|
|
"num_tokens": 184160.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 7.8063740730285645,
|
|
"epoch": 0.08433200177541056,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 7.4067,
|
|
"mean_token_accuracy": 0.08633613213896751,
|
|
"num_tokens": 194797.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 7.8098077297210695,
|
|
"epoch": 0.0887705281846427,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 7.4494,
|
|
"mean_token_accuracy": 0.08306959643959999,
|
|
"num_tokens": 204108.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 7.793222093582154,
|
|
"epoch": 0.09320905459387484,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 7.4157,
|
|
"mean_token_accuracy": 0.08446534052491188,
|
|
"num_tokens": 214945.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 7.810083961486816,
|
|
"epoch": 0.09764758100310697,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 7.5058,
|
|
"mean_token_accuracy": 0.08473781980574131,
|
|
"num_tokens": 226787.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 7.764282941818237,
|
|
"epoch": 0.1020861074123391,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 7.3897,
|
|
"mean_token_accuracy": 0.08418168947100639,
|
|
"num_tokens": 237273.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 7.864174175262451,
|
|
"epoch": 0.10652463382157124,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 7.4156,
|
|
"mean_token_accuracy": 0.09291890189051628,
|
|
"num_tokens": 246793.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 7.779165315628052,
|
|
"epoch": 0.11096316023080337,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 7.4275,
|
|
"mean_token_accuracy": 0.0859360195696354,
|
|
"num_tokens": 258022.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 7.7436638355255125,
|
|
"epoch": 0.11540168664003551,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 7.4169,
|
|
"mean_token_accuracy": 0.08928811252117157,
|
|
"num_tokens": 267670.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 7.7615663528442385,
|
|
"epoch": 0.11984021304926765,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 7.4734,
|
|
"mean_token_accuracy": 0.08662445694208146,
|
|
"num_tokens": 278441.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 7.868949317932129,
|
|
"epoch": 0.12427873945849978,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 7.44,
|
|
"mean_token_accuracy": 0.0834168739616871,
|
|
"num_tokens": 288501.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 7.918918085098267,
|
|
"epoch": 0.1287172658677319,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 7.5111,
|
|
"mean_token_accuracy": 0.08065465465188026,
|
|
"num_tokens": 299557.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 7.7995758056640625,
|
|
"epoch": 0.13315579227696406,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 7.4013,
|
|
"mean_token_accuracy": 0.09215340986847878,
|
|
"num_tokens": 309221.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 7.789921569824219,
|
|
"epoch": 0.13759431868619618,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 7.3926,
|
|
"mean_token_accuracy": 0.0882001981139183,
|
|
"num_tokens": 318890.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 7.7831714153289795,
|
|
"epoch": 0.14203284509542832,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 7.3731,
|
|
"mean_token_accuracy": 0.08587820529937744,
|
|
"num_tokens": 329547.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 7.732769966125488,
|
|
"epoch": 0.14647137150466044,
|
|
"grad_norm": 0.8203125,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.4036,
|
|
"mean_token_accuracy": 0.09170155450701714,
|
|
"num_tokens": 339819.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 7.838970375061035,
|
|
"epoch": 0.1509098979138926,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.4284,
|
|
"mean_token_accuracy": 0.09364147260785102,
|
|
"num_tokens": 349132.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 7.720353078842163,
|
|
"epoch": 0.15534842432312473,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.4269,
|
|
"mean_token_accuracy": 0.08701886683702469,
|
|
"num_tokens": 358950.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 7.802455282211303,
|
|
"epoch": 0.15978695073235685,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.3911,
|
|
"mean_token_accuracy": 0.08798973783850669,
|
|
"num_tokens": 368710.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 7.7396303653717045,
|
|
"epoch": 0.164225477141589,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.4086,
|
|
"mean_token_accuracy": 0.09115821123123169,
|
|
"num_tokens": 377388.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 7.784192132949829,
|
|
"epoch": 0.16866400355082112,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.368,
|
|
"mean_token_accuracy": 0.08906726986169815,
|
|
"num_tokens": 387860.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 7.78899359703064,
|
|
"epoch": 0.17310252996005326,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.3988,
|
|
"mean_token_accuracy": 0.09347473680973054,
|
|
"num_tokens": 397737.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 7.7772763729095455,
|
|
"epoch": 0.1775410563692854,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.4405,
|
|
"mean_token_accuracy": 0.08559705466032028,
|
|
"num_tokens": 408016.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 7.831363105773926,
|
|
"epoch": 0.18197958277851753,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.4124,
|
|
"mean_token_accuracy": 0.08368377536535263,
|
|
"num_tokens": 417578.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 7.737787818908691,
|
|
"epoch": 0.18641810918774968,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.3308,
|
|
"mean_token_accuracy": 0.08822357207536698,
|
|
"num_tokens": 428746.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 7.837935161590576,
|
|
"epoch": 0.1908566355969818,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.39,
|
|
"mean_token_accuracy": 0.08428716734051704,
|
|
"num_tokens": 439500.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 7.771363592147827,
|
|
"epoch": 0.19529516200621394,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.4119,
|
|
"mean_token_accuracy": 0.08383373767137528,
|
|
"num_tokens": 450105.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 7.8077551364898685,
|
|
"epoch": 0.19973368841544606,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.3646,
|
|
"mean_token_accuracy": 0.09255110621452331,
|
|
"num_tokens": 459800.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 7.745184326171875,
|
|
"epoch": 0.2041722148246782,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.2759,
|
|
"mean_token_accuracy": 0.09842955991625786,
|
|
"num_tokens": 470371.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 7.753671360015869,
|
|
"epoch": 0.20861074123391035,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.3007,
|
|
"mean_token_accuracy": 0.09646429568529129,
|
|
"num_tokens": 480347.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 7.81769061088562,
|
|
"epoch": 0.21304926764314247,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.3843,
|
|
"mean_token_accuracy": 0.08836668208241463,
|
|
"num_tokens": 491058.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.67605414390564,
|
|
"epoch": 0.21748779405237462,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.3048,
|
|
"mean_token_accuracy": 0.08849572688341141,
|
|
"num_tokens": 500826.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.790542507171631,
|
|
"epoch": 0.22192632046160674,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.284,
|
|
"mean_token_accuracy": 0.08906695693731308,
|
|
"num_tokens": 510952.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 7.7136742115020756,
|
|
"epoch": 0.22636484687083888,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.2967,
|
|
"mean_token_accuracy": 0.09439613595604897,
|
|
"num_tokens": 520680.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.686872863769532,
|
|
"epoch": 0.23080337328007103,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.378,
|
|
"mean_token_accuracy": 0.0834910437464714,
|
|
"num_tokens": 531400.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.752029180526733,
|
|
"epoch": 0.23524189968930315,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.4008,
|
|
"mean_token_accuracy": 0.08466090187430382,
|
|
"num_tokens": 541657.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.805071783065796,
|
|
"epoch": 0.2396804260985353,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 7.3264,
|
|
"mean_token_accuracy": 0.09501602202653885,
|
|
"num_tokens": 550609.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.560798215866089,
|
|
"epoch": 0.2441189525077674,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.292,
|
|
"mean_token_accuracy": 0.09585189595818519,
|
|
"num_tokens": 561482.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.8325278759002686,
|
|
"epoch": 0.24855747891699956,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 7.2829,
|
|
"mean_token_accuracy": 0.09389769136905671,
|
|
"num_tokens": 570582.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.7413294315338135,
|
|
"epoch": 0.2529960053262317,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.392,
|
|
"mean_token_accuracy": 0.0857204593718052,
|
|
"num_tokens": 581206.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.825753402709961,
|
|
"epoch": 0.2574345317354638,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.193,
|
|
"mean_token_accuracy": 0.09314141124486923,
|
|
"num_tokens": 592122.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.66306438446045,
|
|
"epoch": 0.26187305814469597,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.3151,
|
|
"mean_token_accuracy": 0.08986739963293075,
|
|
"num_tokens": 602919.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.771809720993042,
|
|
"epoch": 0.2663115845539281,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 7.2954,
|
|
"mean_token_accuracy": 0.09381957724690437,
|
|
"num_tokens": 612864.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.627486038208008,
|
|
"epoch": 0.2707501109631602,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000152,
|
|
"loss": 7.2278,
|
|
"mean_token_accuracy": 0.09549270346760749,
|
|
"num_tokens": 622682.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.6965526103973385,
|
|
"epoch": 0.27518863737239235,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 7.3782,
|
|
"mean_token_accuracy": 0.09121644049882889,
|
|
"num_tokens": 634017.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.837769556045532,
|
|
"epoch": 0.2796271637816245,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000157,
|
|
"loss": 7.4513,
|
|
"mean_token_accuracy": 0.08749650418758392,
|
|
"num_tokens": 644132.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.789153337478638,
|
|
"epoch": 0.28406569019085665,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 7.3349,
|
|
"mean_token_accuracy": 0.09567861631512642,
|
|
"num_tokens": 655246.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.75645751953125,
|
|
"epoch": 0.2885042166000888,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000162,
|
|
"loss": 7.3554,
|
|
"mean_token_accuracy": 0.08929399326443672,
|
|
"num_tokens": 664899.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.748797941207886,
|
|
"epoch": 0.2929427430093209,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 7.375,
|
|
"mean_token_accuracy": 0.09080832377076149,
|
|
"num_tokens": 675989.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.664048528671264,
|
|
"epoch": 0.29738126941855303,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 7.3221,
|
|
"mean_token_accuracy": 0.09584744423627853,
|
|
"num_tokens": 687759.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.7706445217132565,
|
|
"epoch": 0.3018197958277852,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 7.2649,
|
|
"mean_token_accuracy": 0.09178439974784851,
|
|
"num_tokens": 696943.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.699291706085205,
|
|
"epoch": 0.3062583222370173,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 7.2908,
|
|
"mean_token_accuracy": 0.0902982622385025,
|
|
"num_tokens": 708139.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.69988489151001,
|
|
"epoch": 0.31069684864624947,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 7.2561,
|
|
"mean_token_accuracy": 0.09475043341517449,
|
|
"num_tokens": 718000.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.659535074234009,
|
|
"epoch": 0.31513537505548156,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000177,
|
|
"loss": 7.2602,
|
|
"mean_token_accuracy": 0.09070580154657364,
|
|
"num_tokens": 728188.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.705199956893921,
|
|
"epoch": 0.3195739014647137,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 7.291,
|
|
"mean_token_accuracy": 0.09317795038223267,
|
|
"num_tokens": 738590.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.739383554458618,
|
|
"epoch": 0.32401242787394585,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000182,
|
|
"loss": 7.3736,
|
|
"mean_token_accuracy": 0.09111551642417907,
|
|
"num_tokens": 748628.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.654271173477173,
|
|
"epoch": 0.328450954283178,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 7.2501,
|
|
"mean_token_accuracy": 0.09355314895510673,
|
|
"num_tokens": 760126.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.799070549011231,
|
|
"epoch": 0.33288948069241014,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000187,
|
|
"loss": 7.3846,
|
|
"mean_token_accuracy": 0.08661651834845543,
|
|
"num_tokens": 770255.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.63041615486145,
|
|
"epoch": 0.33732800710164224,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 7.3517,
|
|
"mean_token_accuracy": 0.08633785173296929,
|
|
"num_tokens": 780744.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.73739104270935,
|
|
"epoch": 0.3417665335108744,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000192,
|
|
"loss": 7.2372,
|
|
"mean_token_accuracy": 0.09704277291893959,
|
|
"num_tokens": 790617.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.6727227687835695,
|
|
"epoch": 0.34620505992010653,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 7.2751,
|
|
"mean_token_accuracy": 0.09424345120787621,
|
|
"num_tokens": 800751.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.7567243576049805,
|
|
"epoch": 0.3506435863293387,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 7.3627,
|
|
"mean_token_accuracy": 0.09301683828234672,
|
|
"num_tokens": 810596.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.74293794631958,
|
|
"epoch": 0.3550821127385708,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 7.3311,
|
|
"mean_token_accuracy": 0.08610923066735268,
|
|
"num_tokens": 821001.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.658763313293457,
|
|
"epoch": 0.3595206391478029,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000202,
|
|
"loss": 7.2313,
|
|
"mean_token_accuracy": 0.09779016748070717,
|
|
"num_tokens": 831653.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.616789484024048,
|
|
"epoch": 0.36395916555703506,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 7.1687,
|
|
"mean_token_accuracy": 0.10131467953324318,
|
|
"num_tokens": 842953.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.563587284088134,
|
|
"epoch": 0.3683976919662672,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000207,
|
|
"loss": 7.2395,
|
|
"mean_token_accuracy": 0.0917266458272934,
|
|
"num_tokens": 853521.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.62612771987915,
|
|
"epoch": 0.37283621837549935,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 7.1909,
|
|
"mean_token_accuracy": 0.09840409010648728,
|
|
"num_tokens": 863525.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.7186483383178714,
|
|
"epoch": 0.37727474478473144,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000212,
|
|
"loss": 7.2794,
|
|
"mean_token_accuracy": 0.09545301347970962,
|
|
"num_tokens": 874101.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.579249715805053,
|
|
"epoch": 0.3817132711939636,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 7.1828,
|
|
"mean_token_accuracy": 0.1006378948688507,
|
|
"num_tokens": 884422.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.561653470993042,
|
|
"epoch": 0.38615179760319573,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 7.1764,
|
|
"mean_token_accuracy": 0.09805170893669128,
|
|
"num_tokens": 894787.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.73306303024292,
|
|
"epoch": 0.3905903240124279,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 7.2928,
|
|
"mean_token_accuracy": 0.08978268429636956,
|
|
"num_tokens": 906295.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.6030584335327145,
|
|
"epoch": 0.39502885042166,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000222,
|
|
"loss": 7.2112,
|
|
"mean_token_accuracy": 0.09668504521250725,
|
|
"num_tokens": 916791.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.55758113861084,
|
|
"epoch": 0.3994673768308921,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 7.1147,
|
|
"mean_token_accuracy": 0.09961161985993386,
|
|
"num_tokens": 926383.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.644591426849365,
|
|
"epoch": 0.40390590324012426,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 7.1943,
|
|
"mean_token_accuracy": 0.10151847004890442,
|
|
"num_tokens": 937319.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.615101528167725,
|
|
"epoch": 0.4083444296493564,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 7.2615,
|
|
"mean_token_accuracy": 0.09779548048973083,
|
|
"num_tokens": 947641.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.583732938766479,
|
|
"epoch": 0.41278295605858856,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 7.1568,
|
|
"mean_token_accuracy": 0.09658500477671624,
|
|
"num_tokens": 957778.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.627662897109985,
|
|
"epoch": 0.4172214824678207,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 7.1428,
|
|
"mean_token_accuracy": 0.09743335545063019,
|
|
"num_tokens": 967619.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.599508905410767,
|
|
"epoch": 0.4216600088770528,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000237,
|
|
"loss": 7.2186,
|
|
"mean_token_accuracy": 0.09613262861967087,
|
|
"num_tokens": 978301.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.5192223072052,
|
|
"epoch": 0.42609853528628494,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 7.2032,
|
|
"mean_token_accuracy": 0.09714524820446968,
|
|
"num_tokens": 989783.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.552160596847534,
|
|
"epoch": 0.4305370616955171,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000242,
|
|
"loss": 7.0988,
|
|
"mean_token_accuracy": 0.10628278627991676,
|
|
"num_tokens": 999827.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.5833902835845945,
|
|
"epoch": 0.43497558810474923,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 7.2168,
|
|
"mean_token_accuracy": 0.09577535167336464,
|
|
"num_tokens": 1011294.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.581695985794068,
|
|
"epoch": 0.4394141145139814,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000247,
|
|
"loss": 7.1425,
|
|
"mean_token_accuracy": 0.09786203876137733,
|
|
"num_tokens": 1021672.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.592150259017944,
|
|
"epoch": 0.44385264092321347,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 7.1239,
|
|
"mean_token_accuracy": 0.10068506821990013,
|
|
"num_tokens": 1031281.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 0.44385264092321347,
|
|
"eval_entropy": 7.398865165166787,
|
|
"eval_loss": 7.180932521820068,
|
|
"eval_mean_token_accuracy": 0.09428256430669806,
|
|
"eval_num_tokens": 1031281.0,
|
|
"eval_runtime": 2.1123,
|
|
"eval_samples_per_second": 1593.963,
|
|
"eval_steps_per_second": 199.305,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.5125542163848875,
|
|
"epoch": 0.4482911673324456,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000252,
|
|
"loss": 7.1865,
|
|
"mean_token_accuracy": 0.09644722416996956,
|
|
"num_tokens": 1042458.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.581930828094483,
|
|
"epoch": 0.45272969374167776,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 7.1654,
|
|
"mean_token_accuracy": 0.10110429748892784,
|
|
"num_tokens": 1051645.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.556396198272705,
|
|
"epoch": 0.4571682201509099,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000257,
|
|
"loss": 7.1574,
|
|
"mean_token_accuracy": 0.10311417281627655,
|
|
"num_tokens": 1061634.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.615333938598633,
|
|
"epoch": 0.46160674656014206,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 7.1778,
|
|
"mean_token_accuracy": 0.10142057165503501,
|
|
"num_tokens": 1071666.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.468676328659058,
|
|
"epoch": 0.46604527296937415,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000262,
|
|
"loss": 7.1345,
|
|
"mean_token_accuracy": 0.09515785649418831,
|
|
"num_tokens": 1082652.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 7.533456087112427,
|
|
"epoch": 0.4704837993786063,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 7.0998,
|
|
"mean_token_accuracy": 0.0970863476395607,
|
|
"num_tokens": 1092396.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.495568656921387,
|
|
"epoch": 0.47492232578783844,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 7.0605,
|
|
"mean_token_accuracy": 0.10546803250908851,
|
|
"num_tokens": 1102504.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 7.440021228790283,
|
|
"epoch": 0.4793608521970706,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 7.112,
|
|
"mean_token_accuracy": 0.09773821383714676,
|
|
"num_tokens": 1112505.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 7.558693218231201,
|
|
"epoch": 0.48379937860630273,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 7.1158,
|
|
"mean_token_accuracy": 0.0986550621688366,
|
|
"num_tokens": 1122603.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 7.461544179916382,
|
|
"epoch": 0.4882379050155348,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 7.1284,
|
|
"mean_token_accuracy": 0.09657116010785102,
|
|
"num_tokens": 1132178.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 7.562966060638428,
|
|
"epoch": 0.49267643142476697,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000277,
|
|
"loss": 7.0946,
|
|
"mean_token_accuracy": 0.09613885954022408,
|
|
"num_tokens": 1141712.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 7.508441638946533,
|
|
"epoch": 0.4971149578339991,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 7.1157,
|
|
"mean_token_accuracy": 0.0992264598608017,
|
|
"num_tokens": 1151748.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 7.558133125305176,
|
|
"epoch": 0.5015534842432312,
|
|
"grad_norm": 2.265625,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 7.0116,
|
|
"mean_token_accuracy": 0.10735590234398842,
|
|
"num_tokens": 1161234.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 7.4882372379302975,
|
|
"epoch": 0.5059920106524634,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 7.0915,
|
|
"mean_token_accuracy": 0.10239496752619744,
|
|
"num_tokens": 1171656.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 7.4804298877716064,
|
|
"epoch": 0.5104305370616955,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000287,
|
|
"loss": 7.1359,
|
|
"mean_token_accuracy": 0.09924655333161354,
|
|
"num_tokens": 1182983.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 7.447319459915161,
|
|
"epoch": 0.5148690634709276,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 7.1005,
|
|
"mean_token_accuracy": 0.09991766214370727,
|
|
"num_tokens": 1193720.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 7.4760712623596195,
|
|
"epoch": 0.5193075898801598,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000292,
|
|
"loss": 7.0304,
|
|
"mean_token_accuracy": 0.10363117456436158,
|
|
"num_tokens": 1203286.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 7.436632061004639,
|
|
"epoch": 0.5237461162893919,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 7.0193,
|
|
"mean_token_accuracy": 0.10448477938771247,
|
|
"num_tokens": 1212875.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 7.479945039749145,
|
|
"epoch": 0.5281846426986241,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000297,
|
|
"loss": 7.0455,
|
|
"mean_token_accuracy": 0.09875646382570266,
|
|
"num_tokens": 1223629.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 7.455782127380371,
|
|
"epoch": 0.5326231691078562,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 7.1663,
|
|
"mean_token_accuracy": 0.10330844521522523,
|
|
"num_tokens": 1233670.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 7.540210819244384,
|
|
"epoch": 0.5370616955170884,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.000302,
|
|
"loss": 7.0882,
|
|
"mean_token_accuracy": 0.10033463165163994,
|
|
"num_tokens": 1244140.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 7.42637152671814,
|
|
"epoch": 0.5415002219263204,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 7.0617,
|
|
"mean_token_accuracy": 0.10370045378804207,
|
|
"num_tokens": 1254496.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 7.397940731048584,
|
|
"epoch": 0.5459387483355526,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000307,
|
|
"loss": 7.0262,
|
|
"mean_token_accuracy": 0.10960142239928246,
|
|
"num_tokens": 1265391.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 7.426065683364868,
|
|
"epoch": 0.5503772747447847,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.9633,
|
|
"mean_token_accuracy": 0.10688221156597137,
|
|
"num_tokens": 1275857.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 7.255760049819946,
|
|
"epoch": 0.5548158011540169,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.9785,
|
|
"mean_token_accuracy": 0.10750289410352706,
|
|
"num_tokens": 1285122.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 7.525947952270508,
|
|
"epoch": 0.559254327563249,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 7.0365,
|
|
"mean_token_accuracy": 0.10156842395663261,
|
|
"num_tokens": 1294476.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 7.383149433135986,
|
|
"epoch": 0.5636928539724811,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000317,
|
|
"loss": 7.0386,
|
|
"mean_token_accuracy": 0.10288441628217697,
|
|
"num_tokens": 1304959.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 7.389831638336181,
|
|
"epoch": 0.5681313803817133,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.9818,
|
|
"mean_token_accuracy": 0.10231715813279152,
|
|
"num_tokens": 1316121.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 7.392814636230469,
|
|
"epoch": 0.5725699067909454,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000322,
|
|
"loss": 7.0793,
|
|
"mean_token_accuracy": 0.10169301480054856,
|
|
"num_tokens": 1326750.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 7.383424234390259,
|
|
"epoch": 0.5770084332001776,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.9712,
|
|
"mean_token_accuracy": 0.10646747648715973,
|
|
"num_tokens": 1337044.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 7.4170256614685055,
|
|
"epoch": 0.5814469596094097,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 7.0334,
|
|
"mean_token_accuracy": 0.09991655647754669,
|
|
"num_tokens": 1347610.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 7.2603747844696045,
|
|
"epoch": 0.5858854860186418,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.9609,
|
|
"mean_token_accuracy": 0.10539844930171967,
|
|
"num_tokens": 1356521.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 7.422995662689209,
|
|
"epoch": 0.5903240124278739,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.9662,
|
|
"mean_token_accuracy": 0.10543927997350692,
|
|
"num_tokens": 1366741.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 7.369775009155274,
|
|
"epoch": 0.5947625388371061,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 7.0411,
|
|
"mean_token_accuracy": 0.10072910860180855,
|
|
"num_tokens": 1377018.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 7.43230504989624,
|
|
"epoch": 0.5992010652463382,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.959,
|
|
"mean_token_accuracy": 0.10484865829348564,
|
|
"num_tokens": 1386387.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 7.3505573749542235,
|
|
"epoch": 0.6036395916555704,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.961,
|
|
"mean_token_accuracy": 0.10673004984855652,
|
|
"num_tokens": 1396049.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 7.355041980743408,
|
|
"epoch": 0.6080781180648025,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000342,
|
|
"loss": 7.0803,
|
|
"mean_token_accuracy": 0.0983244001865387,
|
|
"num_tokens": 1406952.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 7.410180234909058,
|
|
"epoch": 0.6125166444740346,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.9874,
|
|
"mean_token_accuracy": 0.10184509009122848,
|
|
"num_tokens": 1417146.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 7.331110858917237,
|
|
"epoch": 0.6169551708832668,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.9654,
|
|
"mean_token_accuracy": 0.11014403700828553,
|
|
"num_tokens": 1426146.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 7.430603981018066,
|
|
"epoch": 0.6213936972924989,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 7.0644,
|
|
"mean_token_accuracy": 0.10093853399157524,
|
|
"num_tokens": 1437913.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 7.3031325340271,
|
|
"epoch": 0.625832223701731,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.9308,
|
|
"mean_token_accuracy": 0.11181655824184418,
|
|
"num_tokens": 1448329.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 7.382747077941895,
|
|
"epoch": 0.6302707501109631,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.9559,
|
|
"mean_token_accuracy": 0.10284734219312668,
|
|
"num_tokens": 1457896.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 7.310059356689453,
|
|
"epoch": 0.6347092765201953,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.9601,
|
|
"mean_token_accuracy": 0.1071910947561264,
|
|
"num_tokens": 1467705.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 7.25341272354126,
|
|
"epoch": 0.6391478029294274,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.9512,
|
|
"mean_token_accuracy": 0.10645409822463989,
|
|
"num_tokens": 1478132.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 7.391849327087402,
|
|
"epoch": 0.6435863293386596,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000362,
|
|
"loss": 7.019,
|
|
"mean_token_accuracy": 0.10602135583758354,
|
|
"num_tokens": 1488424.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 7.310156202316284,
|
|
"epoch": 0.6480248557478917,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.8812,
|
|
"mean_token_accuracy": 0.11527719348669052,
|
|
"num_tokens": 1498197.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 7.234117460250855,
|
|
"epoch": 0.6524633821571239,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.9052,
|
|
"mean_token_accuracy": 0.10894913673400879,
|
|
"num_tokens": 1508242.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 7.232631635665894,
|
|
"epoch": 0.656901908566356,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.8983,
|
|
"mean_token_accuracy": 0.11159973964095116,
|
|
"num_tokens": 1517652.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 7.2952375411987305,
|
|
"epoch": 0.6613404349755881,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.9347,
|
|
"mean_token_accuracy": 0.10787282809615135,
|
|
"num_tokens": 1527501.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 7.291040468215942,
|
|
"epoch": 0.6657789613848203,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.9339,
|
|
"mean_token_accuracy": 0.11132785305380821,
|
|
"num_tokens": 1537640.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 7.20764536857605,
|
|
"epoch": 0.6702174877940523,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.8701,
|
|
"mean_token_accuracy": 0.11440131813287735,
|
|
"num_tokens": 1548049.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 7.340287876129151,
|
|
"epoch": 0.6746560142032845,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 7.0474,
|
|
"mean_token_accuracy": 0.10261607840657234,
|
|
"num_tokens": 1558389.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 7.222410678863525,
|
|
"epoch": 0.6790945406125166,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.8176,
|
|
"mean_token_accuracy": 0.11613213866949082,
|
|
"num_tokens": 1568489.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 7.2836723804473875,
|
|
"epoch": 0.6835330670217488,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.9313,
|
|
"mean_token_accuracy": 0.10658463388681412,
|
|
"num_tokens": 1578430.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 7.352204275131226,
|
|
"epoch": 0.6879715934309809,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 7.001,
|
|
"mean_token_accuracy": 0.10380481109023094,
|
|
"num_tokens": 1590307.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 7.313673448562622,
|
|
"epoch": 0.6924101198402131,
|
|
"grad_norm": 1.84375,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.8585,
|
|
"mean_token_accuracy": 0.11356580927968025,
|
|
"num_tokens": 1601719.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 7.149180507659912,
|
|
"epoch": 0.6968486462494452,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.8735,
|
|
"mean_token_accuracy": 0.11371317654848098,
|
|
"num_tokens": 1612047.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 7.195708179473877,
|
|
"epoch": 0.7012871726586773,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.8651,
|
|
"mean_token_accuracy": 0.11098539307713509,
|
|
"num_tokens": 1623141.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 7.241322708129883,
|
|
"epoch": 0.7057256990679095,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.8738,
|
|
"mean_token_accuracy": 0.1125923864543438,
|
|
"num_tokens": 1633249.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 7.270007610321045,
|
|
"epoch": 0.7101642254771416,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.9612,
|
|
"mean_token_accuracy": 0.10728915557265281,
|
|
"num_tokens": 1644451.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 7.245293235778808,
|
|
"epoch": 0.7146027518863737,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.8967,
|
|
"mean_token_accuracy": 0.11031619012355805,
|
|
"num_tokens": 1655196.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 7.137471008300781,
|
|
"epoch": 0.7190412782956058,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.8781,
|
|
"mean_token_accuracy": 0.11013623625040055,
|
|
"num_tokens": 1665613.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 7.282591247558594,
|
|
"epoch": 0.723479804704838,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.8188,
|
|
"mean_token_accuracy": 0.11500794291496277,
|
|
"num_tokens": 1675116.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 7.219833612442017,
|
|
"epoch": 0.7279183311140701,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.9236,
|
|
"mean_token_accuracy": 0.10631008669734002,
|
|
"num_tokens": 1684055.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 7.231753492355347,
|
|
"epoch": 0.7323568575233023,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.7797,
|
|
"mean_token_accuracy": 0.12125139310956001,
|
|
"num_tokens": 1694046.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 7.143084859848022,
|
|
"epoch": 0.7367953839325344,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.8325,
|
|
"mean_token_accuracy": 0.1153494082391262,
|
|
"num_tokens": 1704133.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 7.14149580001831,
|
|
"epoch": 0.7412339103417666,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.9275,
|
|
"mean_token_accuracy": 0.11007370427250862,
|
|
"num_tokens": 1714678.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 7.204297304153442,
|
|
"epoch": 0.7456724367509987,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.7453,
|
|
"mean_token_accuracy": 0.12288743630051613,
|
|
"num_tokens": 1725145.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 7.071165657043457,
|
|
"epoch": 0.7501109631602308,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.8729,
|
|
"mean_token_accuracy": 0.11628773137927055,
|
|
"num_tokens": 1735530.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 7.21942458152771,
|
|
"epoch": 0.7545494895694629,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.8964,
|
|
"mean_token_accuracy": 0.11181129217147827,
|
|
"num_tokens": 1746520.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 7.154268980026245,
|
|
"epoch": 0.758988015978695,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.8532,
|
|
"mean_token_accuracy": 0.1179992400109768,
|
|
"num_tokens": 1756091.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 7.089676809310913,
|
|
"epoch": 0.7634265423879272,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.8561,
|
|
"mean_token_accuracy": 0.10415932089090348,
|
|
"num_tokens": 1765968.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 7.192745923995972,
|
|
"epoch": 0.7678650687971593,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.8773,
|
|
"mean_token_accuracy": 0.11299539059400558,
|
|
"num_tokens": 1777409.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 7.130894994735717,
|
|
"epoch": 0.7723035952063915,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.8824,
|
|
"mean_token_accuracy": 0.10640934631228446,
|
|
"num_tokens": 1788440.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 7.20861964225769,
|
|
"epoch": 0.7767421216156236,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.8417,
|
|
"mean_token_accuracy": 0.11558186262845993,
|
|
"num_tokens": 1799145.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 7.05075421333313,
|
|
"epoch": 0.7811806480248558,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.808,
|
|
"mean_token_accuracy": 0.116386828571558,
|
|
"num_tokens": 1809366.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 7.129081392288208,
|
|
"epoch": 0.7856191744340879,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.8875,
|
|
"mean_token_accuracy": 0.10899080038070678,
|
|
"num_tokens": 1820878.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 7.155272579193115,
|
|
"epoch": 0.79005770084332,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.8349,
|
|
"mean_token_accuracy": 0.11343251317739486,
|
|
"num_tokens": 1831870.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 7.249894714355468,
|
|
"epoch": 0.7944962272525522,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.801,
|
|
"mean_token_accuracy": 0.11338023543357849,
|
|
"num_tokens": 1841482.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.969785833358765,
|
|
"epoch": 0.7989347536617842,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.8102,
|
|
"mean_token_accuracy": 0.11380291059613228,
|
|
"num_tokens": 1851748.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 7.100828695297241,
|
|
"epoch": 0.8033732800710164,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.7389,
|
|
"mean_token_accuracy": 0.1162735216319561,
|
|
"num_tokens": 1862292.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 7.249449729919434,
|
|
"epoch": 0.8078118064802485,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.959,
|
|
"mean_token_accuracy": 0.10582878962159156,
|
|
"num_tokens": 1873825.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 7.126041460037231,
|
|
"epoch": 0.8122503328894807,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.7795,
|
|
"mean_token_accuracy": 0.11718677878379821,
|
|
"num_tokens": 1883824.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 7.029882144927979,
|
|
"epoch": 0.8166888592987128,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.7984,
|
|
"mean_token_accuracy": 0.10695101991295815,
|
|
"num_tokens": 1894236.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 7.212645196914673,
|
|
"epoch": 0.821127385707945,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.779,
|
|
"mean_token_accuracy": 0.11693326756358147,
|
|
"num_tokens": 1904083.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 7.023255491256714,
|
|
"epoch": 0.8255659121171771,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.7907,
|
|
"mean_token_accuracy": 0.11693133264780045,
|
|
"num_tokens": 1914845.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 7.073426914215088,
|
|
"epoch": 0.8300044385264093,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.7945,
|
|
"mean_token_accuracy": 0.11135339140892028,
|
|
"num_tokens": 1925999.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 7.0789491653442385,
|
|
"epoch": 0.8344429649356414,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.694,
|
|
"mean_token_accuracy": 0.1280534103512764,
|
|
"num_tokens": 1935401.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.9891969680786135,
|
|
"epoch": 0.8388814913448736,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.7607,
|
|
"mean_token_accuracy": 0.12046314924955367,
|
|
"num_tokens": 1945510.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 7.16357650756836,
|
|
"epoch": 0.8433200177541056,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.7473,
|
|
"mean_token_accuracy": 0.11847912147641182,
|
|
"num_tokens": 1956934.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 7.054219198226929,
|
|
"epoch": 0.8477585441633377,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.7688,
|
|
"mean_token_accuracy": 0.11719952076673508,
|
|
"num_tokens": 1967032.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 7.080799579620361,
|
|
"epoch": 0.8521970705725699,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.8584,
|
|
"mean_token_accuracy": 0.11015522852540016,
|
|
"num_tokens": 1978895.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 7.09550929069519,
|
|
"epoch": 0.856635596981802,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.816,
|
|
"mean_token_accuracy": 0.11443090364336968,
|
|
"num_tokens": 1990156.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 7.139660930633545,
|
|
"epoch": 0.8610741233910342,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.7035,
|
|
"mean_token_accuracy": 0.11342604532837867,
|
|
"num_tokens": 1999643.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.957034969329834,
|
|
"epoch": 0.8655126498002663,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.7472,
|
|
"mean_token_accuracy": 0.11747978329658508,
|
|
"num_tokens": 2010328.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 7.062716341018676,
|
|
"epoch": 0.8699511762094985,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.8272,
|
|
"mean_token_accuracy": 0.11245640963315964,
|
|
"num_tokens": 2021353.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 7.081027889251709,
|
|
"epoch": 0.8743897026187306,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.7701,
|
|
"mean_token_accuracy": 0.11547347530722618,
|
|
"num_tokens": 2032331.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 7.023243951797485,
|
|
"epoch": 0.8788282290279628,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.7011,
|
|
"mean_token_accuracy": 0.1206856571137905,
|
|
"num_tokens": 2041786.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.975612211227417,
|
|
"epoch": 0.8832667554371949,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.6885,
|
|
"mean_token_accuracy": 0.11817316859960555,
|
|
"num_tokens": 2052512.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 7.02327561378479,
|
|
"epoch": 0.8877052818464269,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.6295,
|
|
"mean_token_accuracy": 0.12657489329576493,
|
|
"num_tokens": 2062785.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 0.8877052818464269,
|
|
"eval_entropy": 6.811686369698857,
|
|
"eval_loss": 6.767911434173584,
|
|
"eval_mean_token_accuracy": 0.11571601520772501,
|
|
"eval_num_tokens": 2062785.0,
|
|
"eval_runtime": 2.1114,
|
|
"eval_samples_per_second": 1594.66,
|
|
"eval_steps_per_second": 199.392,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 7.073720645904541,
|
|
"epoch": 0.8921438082556591,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004999998312369076,
|
|
"loss": 6.7709,
|
|
"mean_token_accuracy": 0.11314490139484405,
|
|
"num_tokens": 2072621.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 7.060971355438232,
|
|
"epoch": 0.8965823346648912,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999991456372788,
|
|
"loss": 6.8178,
|
|
"mean_token_accuracy": 0.10872112512588501,
|
|
"num_tokens": 2083356.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 7.0227889060974125,
|
|
"epoch": 0.9010208610741234,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000499997932655026,
|
|
"loss": 6.7777,
|
|
"mean_token_accuracy": 0.11077685430645942,
|
|
"num_tokens": 2092787.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.9889302253723145,
|
|
"epoch": 0.9054593874833555,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004999961922929925,
|
|
"loss": 6.6739,
|
|
"mean_token_accuracy": 0.11871889233589172,
|
|
"num_tokens": 2103721.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.950629568099975,
|
|
"epoch": 0.9098979138925877,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004999939245552573,
|
|
"loss": 6.6956,
|
|
"mean_token_accuracy": 0.12753814980387687,
|
|
"num_tokens": 2115016.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.951031637191773,
|
|
"epoch": 0.9143364403018198,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000499991129447136,
|
|
"loss": 6.6594,
|
|
"mean_token_accuracy": 0.1263865128159523,
|
|
"num_tokens": 2126183.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.928300905227661,
|
|
"epoch": 0.918774966711052,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004999878069751803,
|
|
"loss": 6.7069,
|
|
"mean_token_accuracy": 0.11957084909081458,
|
|
"num_tokens": 2135955.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.915622138977051,
|
|
"epoch": 0.9232134931202841,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004999839571471776,
|
|
"loss": 6.6919,
|
|
"mean_token_accuracy": 0.11919597014784813,
|
|
"num_tokens": 2146064.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.9707104682922365,
|
|
"epoch": 0.9276520195295161,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004999795799721517,
|
|
"loss": 6.6444,
|
|
"mean_token_accuracy": 0.1262632966041565,
|
|
"num_tokens": 2155707.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.99800386428833,
|
|
"epoch": 0.9320905459387483,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999746754603624,
|
|
"loss": 6.6928,
|
|
"mean_token_accuracy": 0.11961428299546242,
|
|
"num_tokens": 2166037.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.904142761230469,
|
|
"epoch": 0.9365290723479804,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004999692436233055,
|
|
"loss": 6.7519,
|
|
"mean_token_accuracy": 0.11929019168019295,
|
|
"num_tokens": 2175993.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 7.008247995376587,
|
|
"epoch": 0.9409675987572126,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000499963284473713,
|
|
"loss": 6.7099,
|
|
"mean_token_accuracy": 0.12117866948246955,
|
|
"num_tokens": 2186829.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.9378032207489015,
|
|
"epoch": 0.9454061251664447,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004999567980255526,
|
|
"loss": 6.665,
|
|
"mean_token_accuracy": 0.12237748578190803,
|
|
"num_tokens": 2195845.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 7.0276360511779785,
|
|
"epoch": 0.9498446515756769,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004999497842940282,
|
|
"loss": 6.6921,
|
|
"mean_token_accuracy": 0.12066913396120071,
|
|
"num_tokens": 2204937.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.900293159484863,
|
|
"epoch": 0.954283177984909,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999422432955794,
|
|
"loss": 6.6703,
|
|
"mean_token_accuracy": 0.12011573910713196,
|
|
"num_tokens": 2214582.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.922044038772583,
|
|
"epoch": 0.9587217043941412,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999341750478818,
|
|
"loss": 6.6229,
|
|
"mean_token_accuracy": 0.1269613318145275,
|
|
"num_tokens": 2225157.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.99785623550415,
|
|
"epoch": 0.9631602308033733,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000499925579569847,
|
|
"loss": 6.6932,
|
|
"mean_token_accuracy": 0.12152787148952485,
|
|
"num_tokens": 2234823.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.851722621917725,
|
|
"epoch": 0.9675987572126055,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999164568816219,
|
|
"loss": 6.6802,
|
|
"mean_token_accuracy": 0.119229806214571,
|
|
"num_tokens": 2244666.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 7.007480764389038,
|
|
"epoch": 0.9720372836218375,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999068070045897,
|
|
"loss": 6.6412,
|
|
"mean_token_accuracy": 0.117543176561594,
|
|
"num_tokens": 2254586.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.836320638656616,
|
|
"epoch": 0.9764758100310696,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000499896629961369,
|
|
"loss": 6.6026,
|
|
"mean_token_accuracy": 0.12370306029915809,
|
|
"num_tokens": 2264453.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.999147844314575,
|
|
"epoch": 0.9809143364403018,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.000499885925775814,
|
|
"loss": 6.6656,
|
|
"mean_token_accuracy": 0.11901217773556709,
|
|
"num_tokens": 2275170.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.810418939590454,
|
|
"epoch": 0.9853528628495339,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004998746944730148,
|
|
"loss": 6.6517,
|
|
"mean_token_accuracy": 0.12251685783267022,
|
|
"num_tokens": 2285448.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.978043413162231,
|
|
"epoch": 0.9897913892587661,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004998629360792965,
|
|
"loss": 6.6361,
|
|
"mean_token_accuracy": 0.1255613885819912,
|
|
"num_tokens": 2295673.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.860985517501831,
|
|
"epoch": 0.9942299156679982,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004998506506222202,
|
|
"loss": 6.644,
|
|
"mean_token_accuracy": 0.12026142254471779,
|
|
"num_tokens": 2306115.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.872910737991333,
|
|
"epoch": 0.9986684420772304,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004998378381305821,
|
|
"loss": 6.637,
|
|
"mean_token_accuracy": 0.12394563928246498,
|
|
"num_tokens": 2316495.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.93203698264228,
|
|
"epoch": 1.0026631158455392,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004998244986344138,
|
|
"loss": 6.5838,
|
|
"mean_token_accuracy": 0.1257213337553872,
|
|
"num_tokens": 2326409.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.777434873580932,
|
|
"epoch": 1.0071016422547714,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004998106321649822,
|
|
"loss": 6.439,
|
|
"mean_token_accuracy": 0.1350421704351902,
|
|
"num_tokens": 2337394.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.8583564281463625,
|
|
"epoch": 1.0115401686640035,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004997962387547893,
|
|
"loss": 6.5089,
|
|
"mean_token_accuracy": 0.12366606816649436,
|
|
"num_tokens": 2347768.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.965995025634766,
|
|
"epoch": 1.0159786950732357,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004997813184375723,
|
|
"loss": 6.5134,
|
|
"mean_token_accuracy": 0.12577569335699082,
|
|
"num_tokens": 2357969.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.872522068023682,
|
|
"epoch": 1.0204172214824678,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004997658712483034,
|
|
"loss": 6.4866,
|
|
"mean_token_accuracy": 0.1270396701991558,
|
|
"num_tokens": 2367878.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.814556646347046,
|
|
"epoch": 1.0248557478917,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004997498972231898,
|
|
"loss": 6.4781,
|
|
"mean_token_accuracy": 0.12578893974423408,
|
|
"num_tokens": 2377072.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.867713165283203,
|
|
"epoch": 1.0292942743009321,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004997333963996734,
|
|
"loss": 6.503,
|
|
"mean_token_accuracy": 0.13253186494112015,
|
|
"num_tokens": 2387348.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.912645387649536,
|
|
"epoch": 1.0337328007101643,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004997163688164313,
|
|
"loss": 6.4919,
|
|
"mean_token_accuracy": 0.12540172412991524,
|
|
"num_tokens": 2398113.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.738404178619385,
|
|
"epoch": 1.0381713271193964,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004996988145133745,
|
|
"loss": 6.3882,
|
|
"mean_token_accuracy": 0.13407822996377944,
|
|
"num_tokens": 2408074.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.699475193023682,
|
|
"epoch": 1.0426098535286286,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004996807335316496,
|
|
"loss": 6.3874,
|
|
"mean_token_accuracy": 0.1336548797786236,
|
|
"num_tokens": 2419752.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.843112182617188,
|
|
"epoch": 1.0470483799378607,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004996621259136368,
|
|
"loss": 6.4743,
|
|
"mean_token_accuracy": 0.1282840073108673,
|
|
"num_tokens": 2430853.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.741962194442749,
|
|
"epoch": 1.0514869063470929,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004996429917029513,
|
|
"loss": 6.4597,
|
|
"mean_token_accuracy": 0.1284678265452385,
|
|
"num_tokens": 2440686.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.87503981590271,
|
|
"epoch": 1.055925432756325,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004996233309444424,
|
|
"loss": 6.4862,
|
|
"mean_token_accuracy": 0.1273033231496811,
|
|
"num_tokens": 2450849.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.7747828483581545,
|
|
"epoch": 1.060363959165557,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004996031436841934,
|
|
"loss": 6.3842,
|
|
"mean_token_accuracy": 0.13247138634324074,
|
|
"num_tokens": 2460827.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.7094989776611325,
|
|
"epoch": 1.064802485574789,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004995824299695219,
|
|
"loss": 6.4723,
|
|
"mean_token_accuracy": 0.1337954096496105,
|
|
"num_tokens": 2471265.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.8431929588317875,
|
|
"epoch": 1.0692410119840212,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004995611898489796,
|
|
"loss": 6.3655,
|
|
"mean_token_accuracy": 0.13849280402064323,
|
|
"num_tokens": 2481346.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.77330379486084,
|
|
"epoch": 1.0736795383932534,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004995394233723516,
|
|
"loss": 6.4696,
|
|
"mean_token_accuracy": 0.13246190845966338,
|
|
"num_tokens": 2492281.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.771844577789307,
|
|
"epoch": 1.0781180648024855,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004995171305906574,
|
|
"loss": 6.397,
|
|
"mean_token_accuracy": 0.1389589823782444,
|
|
"num_tokens": 2502094.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.706840229034424,
|
|
"epoch": 1.0825565912117177,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004994943115561495,
|
|
"loss": 6.3975,
|
|
"mean_token_accuracy": 0.13392824158072472,
|
|
"num_tokens": 2512553.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.795430517196655,
|
|
"epoch": 1.0869951176209498,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004994709663223143,
|
|
"loss": 6.3947,
|
|
"mean_token_accuracy": 0.13575630336999894,
|
|
"num_tokens": 2522350.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.717914056777954,
|
|
"epoch": 1.091433644030182,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004994470949438712,
|
|
"loss": 6.4571,
|
|
"mean_token_accuracy": 0.12943646237254142,
|
|
"num_tokens": 2532884.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.821149110794067,
|
|
"epoch": 1.095872170439414,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004994226974767734,
|
|
"loss": 6.4947,
|
|
"mean_token_accuracy": 0.12449371442198753,
|
|
"num_tokens": 2543364.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.7451752662658695,
|
|
"epoch": 1.1003106968486462,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004993977739782066,
|
|
"loss": 6.3817,
|
|
"mean_token_accuracy": 0.13001500964164733,
|
|
"num_tokens": 2552281.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.721602582931519,
|
|
"epoch": 1.1047492232578784,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00049937232450659,
|
|
"loss": 6.446,
|
|
"mean_token_accuracy": 0.12717969268560408,
|
|
"num_tokens": 2561963.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.7753888130187985,
|
|
"epoch": 1.1091877496671105,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004993463491215753,
|
|
"loss": 6.4337,
|
|
"mean_token_accuracy": 0.13999854624271393,
|
|
"num_tokens": 2572951.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.816381406784058,
|
|
"epoch": 1.1136262760763427,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000499319847884047,
|
|
"loss": 6.4535,
|
|
"mean_token_accuracy": 0.1278873711824417,
|
|
"num_tokens": 2583019.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.7529043674469,
|
|
"epoch": 1.1180648024855748,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004992928208561224,
|
|
"loss": 6.3902,
|
|
"mean_token_accuracy": 0.13753967881202697,
|
|
"num_tokens": 2593385.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.774140882492065,
|
|
"epoch": 1.122503328894807,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004992652681011508,
|
|
"loss": 6.3959,
|
|
"mean_token_accuracy": 0.13451069965958595,
|
|
"num_tokens": 2603679.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.745535135269165,
|
|
"epoch": 1.1269418553040391,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000499237189683714,
|
|
"loss": 6.458,
|
|
"mean_token_accuracy": 0.1327129691839218,
|
|
"num_tokens": 2614029.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.658591461181641,
|
|
"epoch": 1.1313803817132713,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004992085856696259,
|
|
"loss": 6.4209,
|
|
"mean_token_accuracy": 0.1382272757589817,
|
|
"num_tokens": 2624063.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.722748279571533,
|
|
"epoch": 1.1358189081225034,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004991794561259325,
|
|
"loss": 6.3691,
|
|
"mean_token_accuracy": 0.14201630800962448,
|
|
"num_tokens": 2633792.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.6542223453521725,
|
|
"epoch": 1.1402574345317356,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004991498011209112,
|
|
"loss": 6.3232,
|
|
"mean_token_accuracy": 0.13645180314779282,
|
|
"num_tokens": 2643401.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.773741817474365,
|
|
"epoch": 1.1446959609409677,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004991196207240714,
|
|
"loss": 6.4539,
|
|
"mean_token_accuracy": 0.1282748505473137,
|
|
"num_tokens": 2653909.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.693137884140015,
|
|
"epoch": 1.1491344873501999,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004990889150061541,
|
|
"loss": 6.4317,
|
|
"mean_token_accuracy": 0.1349467732012272,
|
|
"num_tokens": 2664386.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.766120195388794,
|
|
"epoch": 1.1535730137594318,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004990576840391312,
|
|
"loss": 6.4865,
|
|
"mean_token_accuracy": 0.12709747180342673,
|
|
"num_tokens": 2675398.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.749423599243164,
|
|
"epoch": 1.158011540168664,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000499025927896206,
|
|
"loss": 6.389,
|
|
"mean_token_accuracy": 0.1319727636873722,
|
|
"num_tokens": 2684895.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.622655010223388,
|
|
"epoch": 1.162450066577896,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004989936466518127,
|
|
"loss": 6.367,
|
|
"mean_token_accuracy": 0.1397032156586647,
|
|
"num_tokens": 2694706.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.733213758468628,
|
|
"epoch": 1.1668885929871282,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004989608403816164,
|
|
"loss": 6.3576,
|
|
"mean_token_accuracy": 0.13502043187618257,
|
|
"num_tokens": 2706486.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.691570997238159,
|
|
"epoch": 1.1713271193963604,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004989275091625126,
|
|
"loss": 6.4414,
|
|
"mean_token_accuracy": 0.14210355877876282,
|
|
"num_tokens": 2716613.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.886738586425781,
|
|
"epoch": 1.1757656458055925,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004988936530726276,
|
|
"loss": 6.5302,
|
|
"mean_token_accuracy": 0.1268772892653942,
|
|
"num_tokens": 2727313.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.723033380508423,
|
|
"epoch": 1.1802041722148247,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004988592721913176,
|
|
"loss": 6.3995,
|
|
"mean_token_accuracy": 0.1361616112291813,
|
|
"num_tokens": 2737826.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.63746075630188,
|
|
"epoch": 1.1846426986240568,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0004988243665991692,
|
|
"loss": 6.3232,
|
|
"mean_token_accuracy": 0.13523943722248077,
|
|
"num_tokens": 2748962.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.7660548210144045,
|
|
"epoch": 1.189081225033289,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004987889363779985,
|
|
"loss": 6.3726,
|
|
"mean_token_accuracy": 0.13909099400043487,
|
|
"num_tokens": 2760116.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.642192268371582,
|
|
"epoch": 1.193519751442521,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004987529816108517,
|
|
"loss": 6.3868,
|
|
"mean_token_accuracy": 0.13063296377658845,
|
|
"num_tokens": 2771664.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.735338401794434,
|
|
"epoch": 1.1979582778517532,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004987165023820043,
|
|
"loss": 6.3339,
|
|
"mean_token_accuracy": 0.13763192743062974,
|
|
"num_tokens": 2781792.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.5532543659210205,
|
|
"epoch": 1.2023968042609854,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004986794987769611,
|
|
"loss": 6.3272,
|
|
"mean_token_accuracy": 0.1380511462688446,
|
|
"num_tokens": 2792203.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.720197010040283,
|
|
"epoch": 1.2068353306702175,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000498641970882456,
|
|
"loss": 6.4867,
|
|
"mean_token_accuracy": 0.13111176937818528,
|
|
"num_tokens": 2802077.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.638952589035034,
|
|
"epoch": 1.2112738570794497,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004986039187864518,
|
|
"loss": 6.3655,
|
|
"mean_token_accuracy": 0.14026057049632074,
|
|
"num_tokens": 2811993.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.676184177398682,
|
|
"epoch": 1.2157123834886818,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004985653425781401,
|
|
"loss": 6.402,
|
|
"mean_token_accuracy": 0.13089463859796524,
|
|
"num_tokens": 2821924.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.675514078140258,
|
|
"epoch": 1.220150909897914,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004985262423479408,
|
|
"loss": 6.3627,
|
|
"mean_token_accuracy": 0.1411583587527275,
|
|
"num_tokens": 2831521.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.686585426330566,
|
|
"epoch": 1.2245894363071461,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004984866181875021,
|
|
"loss": 6.3526,
|
|
"mean_token_accuracy": 0.13705047890543937,
|
|
"num_tokens": 2841864.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.604874324798584,
|
|
"epoch": 1.229027962716378,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004984464701897005,
|
|
"loss": 6.3588,
|
|
"mean_token_accuracy": 0.13925145491957663,
|
|
"num_tokens": 2852199.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.6554466724395756,
|
|
"epoch": 1.2334664891256102,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004984057984486402,
|
|
"loss": 6.3641,
|
|
"mean_token_accuracy": 0.1341533102095127,
|
|
"num_tokens": 2862167.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.686126089096069,
|
|
"epoch": 1.2379050155348423,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004983646030596527,
|
|
"loss": 6.4079,
|
|
"mean_token_accuracy": 0.13984917402267455,
|
|
"num_tokens": 2872604.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.672490835189819,
|
|
"epoch": 1.2423435419440745,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004983228841192975,
|
|
"loss": 6.3752,
|
|
"mean_token_accuracy": 0.13595345988869667,
|
|
"num_tokens": 2882401.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.631757831573486,
|
|
"epoch": 1.2467820683533066,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004982806417253607,
|
|
"loss": 6.3511,
|
|
"mean_token_accuracy": 0.13876891285181045,
|
|
"num_tokens": 2892004.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.63784236907959,
|
|
"epoch": 1.2512205947625388,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004982378759768557,
|
|
"loss": 6.3538,
|
|
"mean_token_accuracy": 0.13770614489912986,
|
|
"num_tokens": 2902369.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.656727027893067,
|
|
"epoch": 1.255659121171771,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004981945869740225,
|
|
"loss": 6.3934,
|
|
"mean_token_accuracy": 0.13234489262104035,
|
|
"num_tokens": 2912532.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.692376708984375,
|
|
"epoch": 1.260097647581003,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004981507748183276,
|
|
"loss": 6.3853,
|
|
"mean_token_accuracy": 0.1314159318804741,
|
|
"num_tokens": 2922621.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.710336446762085,
|
|
"epoch": 1.2645361739902352,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004981064396124635,
|
|
"loss": 6.3583,
|
|
"mean_token_accuracy": 0.13584646657109262,
|
|
"num_tokens": 2932510.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.6139873504638675,
|
|
"epoch": 1.2689747003994674,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004980615814603492,
|
|
"loss": 6.3846,
|
|
"mean_token_accuracy": 0.12972714975476266,
|
|
"num_tokens": 2942982.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.653766632080078,
|
|
"epoch": 1.2734132268086995,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004980162004671288,
|
|
"loss": 6.2817,
|
|
"mean_token_accuracy": 0.1397496670484543,
|
|
"num_tokens": 2953417.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.590647983551025,
|
|
"epoch": 1.2778517532179317,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004979702967391725,
|
|
"loss": 6.287,
|
|
"mean_token_accuracy": 0.14722541570663453,
|
|
"num_tokens": 2961771.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.587854480743408,
|
|
"epoch": 1.2822902796271638,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004979238703840755,
|
|
"loss": 6.3671,
|
|
"mean_token_accuracy": 0.13086313456296922,
|
|
"num_tokens": 2971980.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.69378170967102,
|
|
"epoch": 1.286728806036396,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004978769215106579,
|
|
"loss": 6.3792,
|
|
"mean_token_accuracy": 0.13028891906142234,
|
|
"num_tokens": 2982087.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.681559753417969,
|
|
"epoch": 1.291167332445628,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004978294502289646,
|
|
"loss": 6.4874,
|
|
"mean_token_accuracy": 0.12749797403812407,
|
|
"num_tokens": 2993424.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.710448789596557,
|
|
"epoch": 1.2956058588548602,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004977814566502651,
|
|
"loss": 6.3213,
|
|
"mean_token_accuracy": 0.1379827730357647,
|
|
"num_tokens": 3003081.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.677050495147705,
|
|
"epoch": 1.3000443852640924,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004977329408870532,
|
|
"loss": 6.4048,
|
|
"mean_token_accuracy": 0.1378847599029541,
|
|
"num_tokens": 3013829.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.576231288909912,
|
|
"epoch": 1.3044829116733245,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004976839030530464,
|
|
"loss": 6.3102,
|
|
"mean_token_accuracy": 0.1419086903333664,
|
|
"num_tokens": 3024511.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.593928623199463,
|
|
"epoch": 1.3089214380825567,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000497634343263186,
|
|
"loss": 6.4043,
|
|
"mean_token_accuracy": 0.12826052978634833,
|
|
"num_tokens": 3034503.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.648963260650635,
|
|
"epoch": 1.3133599644917888,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004975842616336369,
|
|
"loss": 6.3201,
|
|
"mean_token_accuracy": 0.141230496019125,
|
|
"num_tokens": 3044823.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.510059928894043,
|
|
"epoch": 1.317798490901021,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004975336582817869,
|
|
"loss": 6.33,
|
|
"mean_token_accuracy": 0.1362045116722584,
|
|
"num_tokens": 3056412.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.728832674026489,
|
|
"epoch": 1.3222370173102531,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004974825333262469,
|
|
"loss": 6.3945,
|
|
"mean_token_accuracy": 0.12927686050534248,
|
|
"num_tokens": 3067101.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.7084520816802975,
|
|
"epoch": 1.3266755437194853,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004974308868868501,
|
|
"loss": 6.3995,
|
|
"mean_token_accuracy": 0.12630158364772798,
|
|
"num_tokens": 3077638.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.551306915283203,
|
|
"epoch": 1.3311140701287172,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004973787190846524,
|
|
"loss": 6.3218,
|
|
"mean_token_accuracy": 0.14200132787227632,
|
|
"num_tokens": 3087970.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"epoch": 1.3311140701287172,
|
|
"eval_entropy": 6.368161134651891,
|
|
"eval_loss": 6.45974588394165,
|
|
"eval_mean_token_accuracy": 0.13263512823969623,
|
|
"eval_num_tokens": 3087970.0,
|
|
"eval_runtime": 2.0891,
|
|
"eval_samples_per_second": 1611.728,
|
|
"eval_steps_per_second": 201.526,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 6.630201387405395,
|
|
"epoch": 1.3355525965379493,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004973260300419316,
|
|
"loss": 6.3578,
|
|
"mean_token_accuracy": 0.13766367435455323,
|
|
"num_tokens": 3097895.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 6.590450048446655,
|
|
"epoch": 1.3399911229471815,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004972728198821871,
|
|
"loss": 6.3072,
|
|
"mean_token_accuracy": 0.1387806810438633,
|
|
"num_tokens": 3107042.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 6.625532007217407,
|
|
"epoch": 1.3444296493564136,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00049721908873014,
|
|
"loss": 6.315,
|
|
"mean_token_accuracy": 0.14337668791413308,
|
|
"num_tokens": 3117959.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 6.558200359344482,
|
|
"epoch": 1.3488681757656458,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004971648367117323,
|
|
"loss": 6.3543,
|
|
"mean_token_accuracy": 0.13464199230074883,
|
|
"num_tokens": 3128407.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 6.676009941101074,
|
|
"epoch": 1.353306702174878,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004971100639541271,
|
|
"loss": 6.3267,
|
|
"mean_token_accuracy": 0.14146978706121444,
|
|
"num_tokens": 3138265.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 6.57694182395935,
|
|
"epoch": 1.35774522858411,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000497054770585708,
|
|
"loss": 6.3353,
|
|
"mean_token_accuracy": 0.13715606033802033,
|
|
"num_tokens": 3148677.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 6.556269979476928,
|
|
"epoch": 1.3621837549933422,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004969989567360788,
|
|
"loss": 6.3503,
|
|
"mean_token_accuracy": 0.13542981594800949,
|
|
"num_tokens": 3158684.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 6.643629789352417,
|
|
"epoch": 1.3666222814025744,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004969426225360635,
|
|
"loss": 6.3543,
|
|
"mean_token_accuracy": 0.1378885067999363,
|
|
"num_tokens": 3169124.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 6.596967315673828,
|
|
"epoch": 1.3710608078118065,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004968857681177056,
|
|
"loss": 6.2667,
|
|
"mean_token_accuracy": 0.1364466980099678,
|
|
"num_tokens": 3179282.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 6.605694532394409,
|
|
"epoch": 1.3754993342210386,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004968283936142679,
|
|
"loss": 6.301,
|
|
"mean_token_accuracy": 0.13783995881676675,
|
|
"num_tokens": 3190144.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 6.484689426422119,
|
|
"epoch": 1.3799378606302708,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004967704991602322,
|
|
"loss": 6.3025,
|
|
"mean_token_accuracy": 0.14239512234926224,
|
|
"num_tokens": 3201277.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 6.584119367599487,
|
|
"epoch": 1.384376387039503,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004967120848912995,
|
|
"loss": 6.318,
|
|
"mean_token_accuracy": 0.13924861773848535,
|
|
"num_tokens": 3211691.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 6.51182188987732,
|
|
"epoch": 1.388814913448735,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004966531509443884,
|
|
"loss": 6.2972,
|
|
"mean_token_accuracy": 0.1423930622637272,
|
|
"num_tokens": 3220452.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 6.665063858032227,
|
|
"epoch": 1.3932534398579672,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004965936974576363,
|
|
"loss": 6.2572,
|
|
"mean_token_accuracy": 0.1434251219034195,
|
|
"num_tokens": 3230821.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 6.569358205795288,
|
|
"epoch": 1.3976919662671992,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004965337245703981,
|
|
"loss": 6.333,
|
|
"mean_token_accuracy": 0.13867368102073668,
|
|
"num_tokens": 3240621.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 6.564372253417969,
|
|
"epoch": 1.4021304926764313,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004964732324232462,
|
|
"loss": 6.302,
|
|
"mean_token_accuracy": 0.13816518783569337,
|
|
"num_tokens": 3251442.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 6.6255796432495115,
|
|
"epoch": 1.4065690190856635,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00049641222115797,
|
|
"loss": 6.3077,
|
|
"mean_token_accuracy": 0.13412498235702514,
|
|
"num_tokens": 3261164.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 6.539021825790405,
|
|
"epoch": 1.4110075454948956,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004963506909175758,
|
|
"loss": 6.3365,
|
|
"mean_token_accuracy": 0.13258393555879594,
|
|
"num_tokens": 3271442.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 6.633949136734008,
|
|
"epoch": 1.4154460719041277,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000496288641846286,
|
|
"loss": 6.3385,
|
|
"mean_token_accuracy": 0.13758450075984002,
|
|
"num_tokens": 3282159.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 6.485613584518433,
|
|
"epoch": 1.41988459831336,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004962260740895398,
|
|
"loss": 6.2791,
|
|
"mean_token_accuracy": 0.14571996703743934,
|
|
"num_tokens": 3291424.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 6.557862329483032,
|
|
"epoch": 1.424323124722592,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004961629877939913,
|
|
"loss": 6.3422,
|
|
"mean_token_accuracy": 0.1376090705394745,
|
|
"num_tokens": 3302098.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 6.554436111450196,
|
|
"epoch": 1.4287616511318242,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000496099383107511,
|
|
"loss": 6.3472,
|
|
"mean_token_accuracy": 0.13372454419732094,
|
|
"num_tokens": 3313292.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 6.600845575332642,
|
|
"epoch": 1.4332001775410563,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004960352601791835,
|
|
"loss": 6.2123,
|
|
"mean_token_accuracy": 0.14182500839233397,
|
|
"num_tokens": 3324272.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 6.54074878692627,
|
|
"epoch": 1.4376387039502885,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004959706191593087,
|
|
"loss": 6.3102,
|
|
"mean_token_accuracy": 0.1401732988655567,
|
|
"num_tokens": 3335299.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 6.616578388214111,
|
|
"epoch": 1.4420772303595206,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004959054601994007,
|
|
"loss": 6.2826,
|
|
"mean_token_accuracy": 0.1450162336230278,
|
|
"num_tokens": 3344652.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 6.52901258468628,
|
|
"epoch": 1.4465157567687528,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004958397834521878,
|
|
"loss": 6.3048,
|
|
"mean_token_accuracy": 0.139273801445961,
|
|
"num_tokens": 3355166.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 6.562482261657715,
|
|
"epoch": 1.450954283177985,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004957735890716115,
|
|
"loss": 6.2622,
|
|
"mean_token_accuracy": 0.1440594784915447,
|
|
"num_tokens": 3366330.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 6.549576234817505,
|
|
"epoch": 1.455392809587217,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004957068772128273,
|
|
"loss": 6.3315,
|
|
"mean_token_accuracy": 0.14034282714128493,
|
|
"num_tokens": 3375930.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 6.5108832836151125,
|
|
"epoch": 1.4598313359964492,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000495639648032203,
|
|
"loss": 6.1684,
|
|
"mean_token_accuracy": 0.16133927404880524,
|
|
"num_tokens": 3387310.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 6.47861647605896,
|
|
"epoch": 1.4642698624056814,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004955719016873192,
|
|
"loss": 6.2004,
|
|
"mean_token_accuracy": 0.1491427481174469,
|
|
"num_tokens": 3398022.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 6.540802431106568,
|
|
"epoch": 1.4687083888149135,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004955036383369688,
|
|
"loss": 6.216,
|
|
"mean_token_accuracy": 0.14632173478603364,
|
|
"num_tokens": 3407315.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 6.459036493301392,
|
|
"epoch": 1.4731469152241456,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004954348581411564,
|
|
"loss": 6.3148,
|
|
"mean_token_accuracy": 0.14125865548849106,
|
|
"num_tokens": 3418009.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 6.666120433807373,
|
|
"epoch": 1.4775854416333778,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000495365561261098,
|
|
"loss": 6.3262,
|
|
"mean_token_accuracy": 0.13915122225880622,
|
|
"num_tokens": 3428659.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 6.513209629058838,
|
|
"epoch": 1.48202396804261,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004952957478592209,
|
|
"loss": 6.2605,
|
|
"mean_token_accuracy": 0.14562821611762047,
|
|
"num_tokens": 3439102.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 6.495951080322266,
|
|
"epoch": 1.486462494451842,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004952254180991628,
|
|
"loss": 6.3143,
|
|
"mean_token_accuracy": 0.14086345955729485,
|
|
"num_tokens": 3450269.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 6.550183248519898,
|
|
"epoch": 1.4909010208610742,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004951545721457719,
|
|
"loss": 6.2323,
|
|
"mean_token_accuracy": 0.15024334490299224,
|
|
"num_tokens": 3459876.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 6.439149618148804,
|
|
"epoch": 1.4953395472703064,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004950832101651062,
|
|
"loss": 6.2879,
|
|
"mean_token_accuracy": 0.1395568035542965,
|
|
"num_tokens": 3469276.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 6.540930128097534,
|
|
"epoch": 1.4997780736795385,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004950113323244336,
|
|
"loss": 6.1873,
|
|
"mean_token_accuracy": 0.14649046510457991,
|
|
"num_tokens": 3479835.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 6.476472091674805,
|
|
"epoch": 1.5042166000887707,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004949389387922305,
|
|
"loss": 6.2472,
|
|
"mean_token_accuracy": 0.14312107488512993,
|
|
"num_tokens": 3491176.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 6.531066989898681,
|
|
"epoch": 1.5086551264980028,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004948660297381826,
|
|
"loss": 6.2262,
|
|
"mean_token_accuracy": 0.14610032215714455,
|
|
"num_tokens": 3501122.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 6.392983913421631,
|
|
"epoch": 1.5130936529072347,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004947926053331836,
|
|
"loss": 6.1558,
|
|
"mean_token_accuracy": 0.1542304791510105,
|
|
"num_tokens": 3510875.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 6.541310834884643,
|
|
"epoch": 1.517532179316467,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004947186657493354,
|
|
"loss": 6.2377,
|
|
"mean_token_accuracy": 0.1494641751050949,
|
|
"num_tokens": 3521798.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 6.438729286193848,
|
|
"epoch": 1.521970705725699,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004946442111599473,
|
|
"loss": 6.2043,
|
|
"mean_token_accuracy": 0.1451006591320038,
|
|
"num_tokens": 3531456.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 6.580926847457886,
|
|
"epoch": 1.5264092321349312,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004945692417395358,
|
|
"loss": 6.251,
|
|
"mean_token_accuracy": 0.14176156222820283,
|
|
"num_tokens": 3541315.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 6.448061084747314,
|
|
"epoch": 1.5308477585441633,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.000494493757663824,
|
|
"loss": 6.2238,
|
|
"mean_token_accuracy": 0.14351213127374648,
|
|
"num_tokens": 3552876.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 6.4509584426879885,
|
|
"epoch": 1.5352862849533955,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004944177591097415,
|
|
"loss": 6.2331,
|
|
"mean_token_accuracy": 0.14354335144162178,
|
|
"num_tokens": 3563967.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 6.457042932510376,
|
|
"epoch": 1.5397248113626276,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004943412462554236,
|
|
"loss": 6.2159,
|
|
"mean_token_accuracy": 0.1507117457687855,
|
|
"num_tokens": 3573510.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 6.546308755874634,
|
|
"epoch": 1.5441633377718598,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004942642192802114,
|
|
"loss": 6.3099,
|
|
"mean_token_accuracy": 0.13970684260129929,
|
|
"num_tokens": 3584863.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 6.455177211761475,
|
|
"epoch": 1.548601864181092,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004941866783646508,
|
|
"loss": 6.2268,
|
|
"mean_token_accuracy": 0.14618334472179412,
|
|
"num_tokens": 3594489.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 6.556396770477295,
|
|
"epoch": 1.553040390590324,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004941086236904923,
|
|
"loss": 6.2043,
|
|
"mean_token_accuracy": 0.14727460145950316,
|
|
"num_tokens": 3605159.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 6.481395721435547,
|
|
"epoch": 1.557478916999556,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004940300554406909,
|
|
"loss": 6.3012,
|
|
"mean_token_accuracy": 0.14313508197665215,
|
|
"num_tokens": 3615254.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 6.445766162872315,
|
|
"epoch": 1.5619174434087881,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000493950973799405,
|
|
"loss": 6.1703,
|
|
"mean_token_accuracy": 0.15018191039562226,
|
|
"num_tokens": 3625015.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 6.42508454322815,
|
|
"epoch": 1.5663559698180203,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004938713789519967,
|
|
"loss": 6.2497,
|
|
"mean_token_accuracy": 0.1505135826766491,
|
|
"num_tokens": 3636083.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 6.512313747406006,
|
|
"epoch": 1.5707944962272524,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004937912710850309,
|
|
"loss": 6.166,
|
|
"mean_token_accuracy": 0.1465531922876835,
|
|
"num_tokens": 3646363.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 6.370706844329834,
|
|
"epoch": 1.5752330226364846,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004937106503862749,
|
|
"loss": 6.1685,
|
|
"mean_token_accuracy": 0.1537150263786316,
|
|
"num_tokens": 3656868.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 6.539106035232544,
|
|
"epoch": 1.5796715490457167,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004936295170446981,
|
|
"loss": 6.3492,
|
|
"mean_token_accuracy": 0.1347063384950161,
|
|
"num_tokens": 3668105.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 6.607371664047241,
|
|
"epoch": 1.5841100754549489,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004935478712504715,
|
|
"loss": 6.2487,
|
|
"mean_token_accuracy": 0.1385604441165924,
|
|
"num_tokens": 3678841.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 6.4086525440216064,
|
|
"epoch": 1.588548601864181,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004934657131949674,
|
|
"loss": 6.2415,
|
|
"mean_token_accuracy": 0.14987936168909072,
|
|
"num_tokens": 3688603.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 6.529016494750977,
|
|
"epoch": 1.5929871282734132,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004933830430707585,
|
|
"loss": 6.2542,
|
|
"mean_token_accuracy": 0.13600879460573195,
|
|
"num_tokens": 3698312.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 6.4327093124389645,
|
|
"epoch": 1.5974256546826453,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000493299861071618,
|
|
"loss": 6.1204,
|
|
"mean_token_accuracy": 0.15229299366474153,
|
|
"num_tokens": 3708252.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 6.483391284942627,
|
|
"epoch": 1.6018641810918774,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004932161673925189,
|
|
"loss": 6.1691,
|
|
"mean_token_accuracy": 0.1462477594614029,
|
|
"num_tokens": 3718431.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 6.433411121368408,
|
|
"epoch": 1.6063027075011096,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004931319622296333,
|
|
"loss": 6.0884,
|
|
"mean_token_accuracy": 0.15784337967634202,
|
|
"num_tokens": 3729170.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 6.474455261230469,
|
|
"epoch": 1.6107412339103417,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004930472457803324,
|
|
"loss": 6.1861,
|
|
"mean_token_accuracy": 0.1482206791639328,
|
|
"num_tokens": 3738885.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 6.340183973312378,
|
|
"epoch": 1.6151797603195739,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004929620182431858,
|
|
"loss": 6.1744,
|
|
"mean_token_accuracy": 0.15203531160950662,
|
|
"num_tokens": 3748535.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 6.444498491287232,
|
|
"epoch": 1.619618286728806,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004928762798179612,
|
|
"loss": 6.237,
|
|
"mean_token_accuracy": 0.14974780827760698,
|
|
"num_tokens": 3758827.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 6.455253171920776,
|
|
"epoch": 1.6240568131380382,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004927900307056233,
|
|
"loss": 6.1615,
|
|
"mean_token_accuracy": 0.1510964810848236,
|
|
"num_tokens": 3768632.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 6.441723012924195,
|
|
"epoch": 1.6284953395472703,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004927032711083342,
|
|
"loss": 6.214,
|
|
"mean_token_accuracy": 0.14101684093475342,
|
|
"num_tokens": 3779324.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 6.381724882125854,
|
|
"epoch": 1.6329338659565025,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004926160012294526,
|
|
"loss": 6.1586,
|
|
"mean_token_accuracy": 0.15024627447128297,
|
|
"num_tokens": 3789744.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 6.370701789855957,
|
|
"epoch": 1.6373723923657346,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000492528221273533,
|
|
"loss": 6.1104,
|
|
"mean_token_accuracy": 0.1493580684065819,
|
|
"num_tokens": 3799831.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 6.542438220977783,
|
|
"epoch": 1.6418109187749668,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004924399314463258,
|
|
"loss": 6.2365,
|
|
"mean_token_accuracy": 0.14592818170785904,
|
|
"num_tokens": 3812102.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 6.4679584980010985,
|
|
"epoch": 1.646249445184199,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004923511319547761,
|
|
"loss": 6.2914,
|
|
"mean_token_accuracy": 0.13645976781845093,
|
|
"num_tokens": 3822750.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 6.439510631561279,
|
|
"epoch": 1.650687971593431,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000492261823007024,
|
|
"loss": 6.2064,
|
|
"mean_token_accuracy": 0.1489210918545723,
|
|
"num_tokens": 3832354.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 6.458003902435303,
|
|
"epoch": 1.6551264980026632,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004921720048124034,
|
|
"loss": 6.2224,
|
|
"mean_token_accuracy": 0.1468009516596794,
|
|
"num_tokens": 3842930.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 6.5039207458496096,
|
|
"epoch": 1.6595650244118954,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004920816775814422,
|
|
"loss": 6.2599,
|
|
"mean_token_accuracy": 0.14020814523100852,
|
|
"num_tokens": 3852952.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 6.454715347290039,
|
|
"epoch": 1.6640035508211275,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004919908415258612,
|
|
"loss": 6.2067,
|
|
"mean_token_accuracy": 0.14455220550298692,
|
|
"num_tokens": 3863070.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 6.4132616996765135,
|
|
"epoch": 1.6684420772303596,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000491899496858574,
|
|
"loss": 6.1501,
|
|
"mean_token_accuracy": 0.15156037211418152,
|
|
"num_tokens": 3872767.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 6.3353992938995365,
|
|
"epoch": 1.6728806036395918,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004918076437936859,
|
|
"loss": 6.1056,
|
|
"mean_token_accuracy": 0.15438561514019966,
|
|
"num_tokens": 3883183.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 6.433599185943604,
|
|
"epoch": 1.677319130048824,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004917152825464947,
|
|
"loss": 6.1996,
|
|
"mean_token_accuracy": 0.1471118450164795,
|
|
"num_tokens": 3893304.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 6.44781608581543,
|
|
"epoch": 1.681757656458056,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004916224133334885,
|
|
"loss": 6.135,
|
|
"mean_token_accuracy": 0.1504994697868824,
|
|
"num_tokens": 3903786.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 6.413273906707763,
|
|
"epoch": 1.686196182867288,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004915290363723465,
|
|
"loss": 6.1426,
|
|
"mean_token_accuracy": 0.14888948425650597,
|
|
"num_tokens": 3913703.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 6.3861230373382565,
|
|
"epoch": 1.6906347092765202,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004914351518819379,
|
|
"loss": 6.197,
|
|
"mean_token_accuracy": 0.14716671407222748,
|
|
"num_tokens": 3923653.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 6.3618451118469235,
|
|
"epoch": 1.6950732356857523,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004913407600823216,
|
|
"loss": 6.0837,
|
|
"mean_token_accuracy": 0.15539871603250505,
|
|
"num_tokens": 3933915.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 6.429026556015015,
|
|
"epoch": 1.6995117620949844,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004912458611947454,
|
|
"loss": 6.1613,
|
|
"mean_token_accuracy": 0.14607803374528885,
|
|
"num_tokens": 3943334.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 6.512318658828735,
|
|
"epoch": 1.7039502885042166,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000491150455441646,
|
|
"loss": 6.2164,
|
|
"mean_token_accuracy": 0.13555126562714576,
|
|
"num_tokens": 3952608.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 6.345305156707764,
|
|
"epoch": 1.7083888149134487,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004910545430466478,
|
|
"loss": 6.123,
|
|
"mean_token_accuracy": 0.1472136050462723,
|
|
"num_tokens": 3963398.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 6.38430609703064,
|
|
"epoch": 1.7128273413226809,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004909581242345628,
|
|
"loss": 6.1552,
|
|
"mean_token_accuracy": 0.14694450199604034,
|
|
"num_tokens": 3973878.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 6.430195760726929,
|
|
"epoch": 1.717265867731913,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004908611992313905,
|
|
"loss": 6.1648,
|
|
"mean_token_accuracy": 0.13664614632725716,
|
|
"num_tokens": 3983681.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 6.39329776763916,
|
|
"epoch": 1.7217043941411452,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004907637682643162,
|
|
"loss": 6.201,
|
|
"mean_token_accuracy": 0.14567800536751746,
|
|
"num_tokens": 3995026.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 6.430534839630127,
|
|
"epoch": 1.7261429205503773,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004906658315617112,
|
|
"loss": 6.2022,
|
|
"mean_token_accuracy": 0.14595392793416978,
|
|
"num_tokens": 4005089.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 6.359121513366699,
|
|
"epoch": 1.7305814469596092,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004905673893531331,
|
|
"loss": 6.1076,
|
|
"mean_token_accuracy": 0.15097568482160567,
|
|
"num_tokens": 4014795.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 6.326131105422974,
|
|
"epoch": 1.7350199733688414,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.0004904684418693231,
|
|
"loss": 6.1828,
|
|
"mean_token_accuracy": 0.14517654925584794,
|
|
"num_tokens": 4025563.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 6.499860239028931,
|
|
"epoch": 1.7394584997780735,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004903689893422077,
|
|
"loss": 6.116,
|
|
"mean_token_accuracy": 0.1486818477511406,
|
|
"num_tokens": 4035050.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 6.305110931396484,
|
|
"epoch": 1.7438970261873057,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004902690320048966,
|
|
"loss": 6.1345,
|
|
"mean_token_accuracy": 0.14928318560123444,
|
|
"num_tokens": 4046109.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 6.449776601791382,
|
|
"epoch": 1.7483355525965378,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000490168570091683,
|
|
"loss": 6.1848,
|
|
"mean_token_accuracy": 0.1390085846185684,
|
|
"num_tokens": 4056776.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 6.319344854354858,
|
|
"epoch": 1.75277407900577,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004900676038380429,
|
|
"loss": 6.0406,
|
|
"mean_token_accuracy": 0.16621824055910112,
|
|
"num_tokens": 4066326.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 6.454939699172973,
|
|
"epoch": 1.7572126054150021,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004899661334806342,
|
|
"loss": 6.1759,
|
|
"mean_token_accuracy": 0.15071096420288085,
|
|
"num_tokens": 4077478.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 6.262599611282349,
|
|
"epoch": 1.7616511318242343,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004898641592572965,
|
|
"loss": 6.1057,
|
|
"mean_token_accuracy": 0.15066718459129333,
|
|
"num_tokens": 4087661.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 6.442129898071289,
|
|
"epoch": 1.7660896582334664,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004897616814070505,
|
|
"loss": 6.1694,
|
|
"mean_token_accuracy": 0.15258410200476646,
|
|
"num_tokens": 4097450.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 6.382274055480957,
|
|
"epoch": 1.7705281846426986,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004896587001700972,
|
|
"loss": 6.2003,
|
|
"mean_token_accuracy": 0.14331723377108574,
|
|
"num_tokens": 4106850.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 6.4386186599731445,
|
|
"epoch": 1.7749667110519307,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004895552157878174,
|
|
"loss": 6.2273,
|
|
"mean_token_accuracy": 0.1446727268397808,
|
|
"num_tokens": 4117561.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 1.7749667110519307,
|
|
"eval_entropy": 6.304367476574315,
|
|
"eval_loss": 6.241267681121826,
|
|
"eval_mean_token_accuracy": 0.14483241421876214,
|
|
"eval_num_tokens": 4117561.0,
|
|
"eval_runtime": 2.0862,
|
|
"eval_samples_per_second": 1613.972,
|
|
"eval_steps_per_second": 201.806,
|
|
"step": 2000
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 11260,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 10,
|
|
"save_steps": 500,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 926375873863680.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|