506b135ec4
Model: fpadovani/rus-cyrl-10mb-after-ppt-Dp-10mb-ckpt500_seed10 Source: Original Platform
1046 lines
28 KiB
JSON
1046 lines
28 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.44385264092321347,
|
|
"eval_steps": 500,
|
|
"global_step": 500,
|
|
"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
|
|
}
|
|
],
|
|
"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": 230847095439360.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|