6046 lines
164 KiB
JSON
6046 lines
164 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.299895036737142,
|
|
"eval_steps": 3000,
|
|
"global_step": 3000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.691991329193115,
|
|
"epoch": 0.00049982506122857,
|
|
"grad_norm": 12.6875,
|
|
"learning_rate": 2e-06,
|
|
"loss": 10.7898,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 8821.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.691885375976563,
|
|
"epoch": 0.00099965012245714,
|
|
"grad_norm": 12.8125,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 10.7486,
|
|
"mean_token_accuracy": 0.0003800044767558575,
|
|
"num_tokens": 18288.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.690632152557374,
|
|
"epoch": 0.00149947518368571,
|
|
"grad_norm": 9.1875,
|
|
"learning_rate": 7e-06,
|
|
"loss": 10.5694,
|
|
"mean_token_accuracy": 0.028398845111951232,
|
|
"num_tokens": 28664.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.66879768371582,
|
|
"epoch": 0.00199930024491428,
|
|
"grad_norm": 9.75,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 10.2706,
|
|
"mean_token_accuracy": 0.042865953966975213,
|
|
"num_tokens": 37807.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.517071342468261,
|
|
"epoch": 0.00249912530614285,
|
|
"grad_norm": 3.78125,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 10.1036,
|
|
"mean_token_accuracy": 0.04603611640632153,
|
|
"num_tokens": 48086.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.393479061126708,
|
|
"epoch": 0.00299895036737142,
|
|
"grad_norm": 3.5,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 9.9637,
|
|
"mean_token_accuracy": 0.0450932290405035,
|
|
"num_tokens": 57427.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.494590854644775,
|
|
"epoch": 0.00349877542859999,
|
|
"grad_norm": 2.96875,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 9.8782,
|
|
"mean_token_accuracy": 0.05189286395907402,
|
|
"num_tokens": 66662.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.441628360748291,
|
|
"epoch": 0.00399860048982856,
|
|
"grad_norm": 2.625,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 9.9029,
|
|
"mean_token_accuracy": 0.0466872725635767,
|
|
"num_tokens": 77388.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.414823532104492,
|
|
"epoch": 0.00449842555105713,
|
|
"grad_norm": 2.484375,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 9.8107,
|
|
"mean_token_accuracy": 0.05048008672893047,
|
|
"num_tokens": 86093.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.454731464385986,
|
|
"epoch": 0.0049982506122857,
|
|
"grad_norm": 2.265625,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 9.8342,
|
|
"mean_token_accuracy": 0.05612768940627575,
|
|
"num_tokens": 96097.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.447371864318848,
|
|
"epoch": 0.00549807567351427,
|
|
"grad_norm": 2.171875,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.8003,
|
|
"mean_token_accuracy": 0.05726277977228165,
|
|
"num_tokens": 106328.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.290729427337647,
|
|
"epoch": 0.00599790073474284,
|
|
"grad_norm": 1.9140625,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.6517,
|
|
"mean_token_accuracy": 0.06179290115833282,
|
|
"num_tokens": 114822.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.229889583587646,
|
|
"epoch": 0.00649772579597141,
|
|
"grad_norm": 2.1875,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.6975,
|
|
"mean_token_accuracy": 0.05696648210287094,
|
|
"num_tokens": 124984.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.342562770843506,
|
|
"epoch": 0.00699755085719998,
|
|
"grad_norm": 1.7109375,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.5666,
|
|
"mean_token_accuracy": 0.06108379699289799,
|
|
"num_tokens": 134054.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.226979160308838,
|
|
"epoch": 0.00749737591842855,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.5292,
|
|
"mean_token_accuracy": 0.05906996764242649,
|
|
"num_tokens": 144561.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.289578342437744,
|
|
"epoch": 0.00799720097965712,
|
|
"grad_norm": 1.84375,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.5334,
|
|
"mean_token_accuracy": 0.057385103404521944,
|
|
"num_tokens": 153701.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.220948886871337,
|
|
"epoch": 0.00849702604088569,
|
|
"grad_norm": 1.7734375,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.3868,
|
|
"mean_token_accuracy": 0.06292971596121788,
|
|
"num_tokens": 164291.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.15461025238037,
|
|
"epoch": 0.00899685110211426,
|
|
"grad_norm": 1.7265625,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 9.3427,
|
|
"mean_token_accuracy": 0.0600983664393425,
|
|
"num_tokens": 174136.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.212667751312257,
|
|
"epoch": 0.00949667616334283,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 9.2902,
|
|
"mean_token_accuracy": 0.06364279538393021,
|
|
"num_tokens": 184368.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.123463916778565,
|
|
"epoch": 0.0099965012245714,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 9.2736,
|
|
"mean_token_accuracy": 0.055468232929706575,
|
|
"num_tokens": 194659.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.121500205993652,
|
|
"epoch": 0.01049632628579997,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 9.1603,
|
|
"mean_token_accuracy": 0.06068473868072033,
|
|
"num_tokens": 204875.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.076999092102051,
|
|
"epoch": 0.01099615134702854,
|
|
"grad_norm": 2.09375,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 9.1596,
|
|
"mean_token_accuracy": 0.05537644699215889,
|
|
"num_tokens": 217033.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 9.973392391204834,
|
|
"epoch": 0.01149597640825711,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 8.9827,
|
|
"mean_token_accuracy": 0.06727760210633278,
|
|
"num_tokens": 225946.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 9.899255275726318,
|
|
"epoch": 0.01199580146948568,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.9333,
|
|
"mean_token_accuracy": 0.064241661131382,
|
|
"num_tokens": 235413.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 9.898325634002685,
|
|
"epoch": 0.01249562653071425,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.9549,
|
|
"mean_token_accuracy": 0.05523654818534851,
|
|
"num_tokens": 247913.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 9.778405952453614,
|
|
"epoch": 0.01299545159194282,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 8.8377,
|
|
"mean_token_accuracy": 0.0660038247704506,
|
|
"num_tokens": 258486.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 9.669632911682129,
|
|
"epoch": 0.01349527665317139,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 8.7473,
|
|
"mean_token_accuracy": 0.062194935604929925,
|
|
"num_tokens": 268027.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 9.625946807861329,
|
|
"epoch": 0.01399510171439996,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 8.6678,
|
|
"mean_token_accuracy": 0.06665367037057876,
|
|
"num_tokens": 277470.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 9.440710926055909,
|
|
"epoch": 0.01449492677562853,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 8.6292,
|
|
"mean_token_accuracy": 0.0616190355271101,
|
|
"num_tokens": 286891.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 9.300586795806884,
|
|
"epoch": 0.0149947518368571,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 8.5505,
|
|
"mean_token_accuracy": 0.07438425682485103,
|
|
"num_tokens": 295636.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 9.237431240081786,
|
|
"epoch": 0.01549457689808567,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 8.5224,
|
|
"mean_token_accuracy": 0.06782945469021798,
|
|
"num_tokens": 306542.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 9.139766025543214,
|
|
"epoch": 0.01599440195931424,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 8.4669,
|
|
"mean_token_accuracy": 0.06699273698031902,
|
|
"num_tokens": 315956.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 9.095052909851074,
|
|
"epoch": 0.01649422702054281,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 8.4695,
|
|
"mean_token_accuracy": 0.0664449006319046,
|
|
"num_tokens": 325500.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 8.970187854766845,
|
|
"epoch": 0.01699405208177138,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 8.4878,
|
|
"mean_token_accuracy": 0.07641985937952996,
|
|
"num_tokens": 335194.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 8.9250994682312,
|
|
"epoch": 0.01749387714299995,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 8.4259,
|
|
"mean_token_accuracy": 0.07149134054780007,
|
|
"num_tokens": 344717.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.886880683898926,
|
|
"epoch": 0.01799370220422852,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 8.3504,
|
|
"mean_token_accuracy": 0.07034418731927872,
|
|
"num_tokens": 354476.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.828811931610108,
|
|
"epoch": 0.01849352726545709,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 8.4046,
|
|
"mean_token_accuracy": 0.06441363953053951,
|
|
"num_tokens": 363939.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 8.870282554626465,
|
|
"epoch": 0.01899335232668566,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 8.4495,
|
|
"mean_token_accuracy": 0.06985239945352077,
|
|
"num_tokens": 373187.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 8.721969509124756,
|
|
"epoch": 0.01949317738791423,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 8.3612,
|
|
"mean_token_accuracy": 0.07164071202278137,
|
|
"num_tokens": 382999.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 8.758165073394775,
|
|
"epoch": 0.0199930024491428,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 8.4243,
|
|
"mean_token_accuracy": 0.06688517518341541,
|
|
"num_tokens": 392639.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 8.722731494903565,
|
|
"epoch": 0.02049282751037137,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000102,
|
|
"loss": 8.4128,
|
|
"mean_token_accuracy": 0.06573708839714527,
|
|
"num_tokens": 403089.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 8.734565448760986,
|
|
"epoch": 0.02099265257159994,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 8.3621,
|
|
"mean_token_accuracy": 0.07012501321732997,
|
|
"num_tokens": 412321.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 8.65351037979126,
|
|
"epoch": 0.02149247763282851,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000107,
|
|
"loss": 8.3951,
|
|
"mean_token_accuracy": 0.0715017631649971,
|
|
"num_tokens": 421599.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 8.667902088165283,
|
|
"epoch": 0.02199230269405708,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 8.3828,
|
|
"mean_token_accuracy": 0.0675255998969078,
|
|
"num_tokens": 432012.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 8.715345764160157,
|
|
"epoch": 0.02249212775528565,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.000112,
|
|
"loss": 8.3796,
|
|
"mean_token_accuracy": 0.06614709198474884,
|
|
"num_tokens": 441989.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 8.6453782081604,
|
|
"epoch": 0.02299195281651422,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 8.3404,
|
|
"mean_token_accuracy": 0.07430519983172416,
|
|
"num_tokens": 451532.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 8.606919479370116,
|
|
"epoch": 0.02349177787774279,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 8.3501,
|
|
"mean_token_accuracy": 0.07073643431067467,
|
|
"num_tokens": 461815.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 8.688431644439698,
|
|
"epoch": 0.02399160293897136,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 8.4073,
|
|
"mean_token_accuracy": 0.06408721245825291,
|
|
"num_tokens": 470860.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 8.692715835571288,
|
|
"epoch": 0.02449142800019993,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000122,
|
|
"loss": 8.3683,
|
|
"mean_token_accuracy": 0.07005545422434807,
|
|
"num_tokens": 480649.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 8.633028984069824,
|
|
"epoch": 0.0249912530614285,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 8.3097,
|
|
"mean_token_accuracy": 0.07226504758000374,
|
|
"num_tokens": 489969.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 8.586344814300537,
|
|
"epoch": 0.02549107812265707,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000127,
|
|
"loss": 8.3083,
|
|
"mean_token_accuracy": 0.0720578346401453,
|
|
"num_tokens": 500000.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 8.595793437957763,
|
|
"epoch": 0.02599090318388564,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 8.2906,
|
|
"mean_token_accuracy": 0.07689355686306953,
|
|
"num_tokens": 510013.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 8.487995624542236,
|
|
"epoch": 0.026490728245114212,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000132,
|
|
"loss": 8.3216,
|
|
"mean_token_accuracy": 0.07494106702506542,
|
|
"num_tokens": 519698.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 8.628628921508788,
|
|
"epoch": 0.02699055330634278,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 8.3313,
|
|
"mean_token_accuracy": 0.07653104439377785,
|
|
"num_tokens": 531631.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 8.49343662261963,
|
|
"epoch": 0.02749037836757135,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 8.309,
|
|
"mean_token_accuracy": 0.07760142609477043,
|
|
"num_tokens": 542204.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 8.588697624206542,
|
|
"epoch": 0.02799020342879992,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 8.2178,
|
|
"mean_token_accuracy": 0.07421289905905723,
|
|
"num_tokens": 552269.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 8.481217861175537,
|
|
"epoch": 0.02849002849002849,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 8.2257,
|
|
"mean_token_accuracy": 0.07464087493717671,
|
|
"num_tokens": 561907.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 8.594965744018555,
|
|
"epoch": 0.02898985355125706,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 8.3486,
|
|
"mean_token_accuracy": 0.07167273983359337,
|
|
"num_tokens": 571283.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 8.453783988952637,
|
|
"epoch": 0.02948967861248563,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.000147,
|
|
"loss": 8.208,
|
|
"mean_token_accuracy": 0.08085252717137337,
|
|
"num_tokens": 579896.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 8.439711093902588,
|
|
"epoch": 0.0299895036737142,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 8.1769,
|
|
"mean_token_accuracy": 0.08574878796935081,
|
|
"num_tokens": 588856.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 8.485370445251466,
|
|
"epoch": 0.03048932873494277,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000152,
|
|
"loss": 8.2282,
|
|
"mean_token_accuracy": 0.08083872310817242,
|
|
"num_tokens": 598910.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 8.521125984191894,
|
|
"epoch": 0.03098915379617134,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 8.2775,
|
|
"mean_token_accuracy": 0.07208913937211037,
|
|
"num_tokens": 609248.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 8.504746055603027,
|
|
"epoch": 0.03148897885739991,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000157,
|
|
"loss": 8.2766,
|
|
"mean_token_accuracy": 0.07899871729314327,
|
|
"num_tokens": 619352.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 8.49971408843994,
|
|
"epoch": 0.03198880391862848,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 8.2799,
|
|
"mean_token_accuracy": 0.07424024008214473,
|
|
"num_tokens": 628300.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 8.551074409484864,
|
|
"epoch": 0.03248862897985705,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000162,
|
|
"loss": 8.2263,
|
|
"mean_token_accuracy": 0.0815961055457592,
|
|
"num_tokens": 637969.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 8.433257675170898,
|
|
"epoch": 0.03298845404108562,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 8.1501,
|
|
"mean_token_accuracy": 0.08082715310156345,
|
|
"num_tokens": 647518.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 8.569383907318116,
|
|
"epoch": 0.03348827910231419,
|
|
"grad_norm": 2.609375,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 8.3036,
|
|
"mean_token_accuracy": 0.07518826834857464,
|
|
"num_tokens": 658292.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 8.49070987701416,
|
|
"epoch": 0.03398810416354276,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 8.198,
|
|
"mean_token_accuracy": 0.08227207660675048,
|
|
"num_tokens": 667872.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 8.390650367736816,
|
|
"epoch": 0.03448792922477133,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 8.1071,
|
|
"mean_token_accuracy": 0.08415818810462952,
|
|
"num_tokens": 676976.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 8.475732135772706,
|
|
"epoch": 0.0349877542859999,
|
|
"grad_norm": 2.0625,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 8.125,
|
|
"mean_token_accuracy": 0.08281808346509933,
|
|
"num_tokens": 686149.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 8.361923885345458,
|
|
"epoch": 0.03548757934722847,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000177,
|
|
"loss": 8.0348,
|
|
"mean_token_accuracy": 0.09114877879619598,
|
|
"num_tokens": 696439.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 8.345643234252929,
|
|
"epoch": 0.03598740440845704,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 8.0586,
|
|
"mean_token_accuracy": 0.08257722929120064,
|
|
"num_tokens": 705902.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 8.347632789611817,
|
|
"epoch": 0.03648722946968561,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000182,
|
|
"loss": 8.2259,
|
|
"mean_token_accuracy": 0.08338720053434372,
|
|
"num_tokens": 715973.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 8.380148220062257,
|
|
"epoch": 0.03698705453091418,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 8.1158,
|
|
"mean_token_accuracy": 0.08314864560961724,
|
|
"num_tokens": 725463.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 8.452237033843994,
|
|
"epoch": 0.03748687959214275,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000187,
|
|
"loss": 8.1331,
|
|
"mean_token_accuracy": 0.07936708480119706,
|
|
"num_tokens": 734941.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 8.413053894042969,
|
|
"epoch": 0.03798670465337132,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 8.1623,
|
|
"mean_token_accuracy": 0.07542651332914829,
|
|
"num_tokens": 744302.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 8.298501968383789,
|
|
"epoch": 0.03848652971459989,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000192,
|
|
"loss": 8.0864,
|
|
"mean_token_accuracy": 0.0863127164542675,
|
|
"num_tokens": 754341.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 8.491098117828368,
|
|
"epoch": 0.03898635477582846,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 8.2233,
|
|
"mean_token_accuracy": 0.07063181065022946,
|
|
"num_tokens": 764087.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 8.280704689025878,
|
|
"epoch": 0.03948617983705703,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 8.1443,
|
|
"mean_token_accuracy": 0.0792117640376091,
|
|
"num_tokens": 773728.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 8.518323612213134,
|
|
"epoch": 0.0399860048982856,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 8.1695,
|
|
"mean_token_accuracy": 0.07747316658496857,
|
|
"num_tokens": 784839.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 8.327428245544434,
|
|
"epoch": 0.04048582995951417,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000202,
|
|
"loss": 8.1449,
|
|
"mean_token_accuracy": 0.08543298915028572,
|
|
"num_tokens": 794514.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 8.39577693939209,
|
|
"epoch": 0.04098565502074274,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 8.12,
|
|
"mean_token_accuracy": 0.08610004633665085,
|
|
"num_tokens": 805254.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 8.327196311950683,
|
|
"epoch": 0.04148548008197131,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000207,
|
|
"loss": 8.0291,
|
|
"mean_token_accuracy": 0.08783004581928253,
|
|
"num_tokens": 814135.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 8.247317600250245,
|
|
"epoch": 0.04198530514319988,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 8.0769,
|
|
"mean_token_accuracy": 0.0855021134018898,
|
|
"num_tokens": 823531.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 8.29983892440796,
|
|
"epoch": 0.04248513020442845,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000212,
|
|
"loss": 7.9907,
|
|
"mean_token_accuracy": 0.08977121785283089,
|
|
"num_tokens": 832833.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 8.302264976501466,
|
|
"epoch": 0.04298495526565702,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 7.9899,
|
|
"mean_token_accuracy": 0.08786641731858254,
|
|
"num_tokens": 841541.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 8.21257724761963,
|
|
"epoch": 0.04348478032688559,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 8.0154,
|
|
"mean_token_accuracy": 0.08381219580769539,
|
|
"num_tokens": 850689.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 8.272824764251709,
|
|
"epoch": 0.04398460538811416,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 7.9668,
|
|
"mean_token_accuracy": 0.09021262973546981,
|
|
"num_tokens": 860573.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 8.297366428375245,
|
|
"epoch": 0.04448443044934273,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000222,
|
|
"loss": 8.0741,
|
|
"mean_token_accuracy": 0.08352044560015201,
|
|
"num_tokens": 869784.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 8.307674121856689,
|
|
"epoch": 0.0449842555105713,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 8.0295,
|
|
"mean_token_accuracy": 0.08401608467102051,
|
|
"num_tokens": 878716.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 8.190884590148926,
|
|
"epoch": 0.04548408057179987,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 8.1069,
|
|
"mean_token_accuracy": 0.08195061162114144,
|
|
"num_tokens": 888334.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 8.293374824523926,
|
|
"epoch": 0.04598390563302844,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 8.0342,
|
|
"mean_token_accuracy": 0.08340942040085793,
|
|
"num_tokens": 897701.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 8.285488986968994,
|
|
"epoch": 0.04648373069425701,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 8.0071,
|
|
"mean_token_accuracy": 0.08369808346033096,
|
|
"num_tokens": 907278.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 8.194073104858399,
|
|
"epoch": 0.04698355575548558,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 8.0156,
|
|
"mean_token_accuracy": 0.08289322704076767,
|
|
"num_tokens": 916194.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 8.201291465759278,
|
|
"epoch": 0.04748338081671415,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000237,
|
|
"loss": 8.0324,
|
|
"mean_token_accuracy": 0.08584462963044644,
|
|
"num_tokens": 926743.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 8.315499877929687,
|
|
"epoch": 0.04798320587794272,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 8.0711,
|
|
"mean_token_accuracy": 0.07998838126659394,
|
|
"num_tokens": 936059.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 8.14477767944336,
|
|
"epoch": 0.04848303093917129,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000242,
|
|
"loss": 7.9935,
|
|
"mean_token_accuracy": 0.08685875385999679,
|
|
"num_tokens": 945990.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 8.329904556274414,
|
|
"epoch": 0.04898285600039986,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 8.0686,
|
|
"mean_token_accuracy": 0.08279181569814682,
|
|
"num_tokens": 956626.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 8.200769996643066,
|
|
"epoch": 0.04948268106162843,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000247,
|
|
"loss": 8.0158,
|
|
"mean_token_accuracy": 0.08317816480994225,
|
|
"num_tokens": 966200.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 8.185665225982666,
|
|
"epoch": 0.049982506122857,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 7.9803,
|
|
"mean_token_accuracy": 0.08663359805941581,
|
|
"num_tokens": 975651.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 8.208883476257324,
|
|
"epoch": 0.05048233118408557,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000252,
|
|
"loss": 7.9563,
|
|
"mean_token_accuracy": 0.08982686921954156,
|
|
"num_tokens": 985264.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 8.222821950912476,
|
|
"epoch": 0.05098215624531414,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 7.9719,
|
|
"mean_token_accuracy": 0.08723178803920746,
|
|
"num_tokens": 996723.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 8.118862915039063,
|
|
"epoch": 0.05148198130654271,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000257,
|
|
"loss": 7.9813,
|
|
"mean_token_accuracy": 0.08684109896421432,
|
|
"num_tokens": 1006098.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 8.217711162567138,
|
|
"epoch": 0.05198180636777128,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 8.0171,
|
|
"mean_token_accuracy": 0.08192172124981881,
|
|
"num_tokens": 1016595.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 8.109809970855713,
|
|
"epoch": 0.05248163142899985,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000262,
|
|
"loss": 7.8715,
|
|
"mean_token_accuracy": 0.09216257557272911,
|
|
"num_tokens": 1026281.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 8.174291276931763,
|
|
"epoch": 0.052981456490228424,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 7.9929,
|
|
"mean_token_accuracy": 0.0979285828769207,
|
|
"num_tokens": 1036697.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 8.011242246627807,
|
|
"epoch": 0.05348128155145699,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 7.9404,
|
|
"mean_token_accuracy": 0.08916681483387948,
|
|
"num_tokens": 1047433.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 8.192585468292236,
|
|
"epoch": 0.05398110661268556,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 7.9844,
|
|
"mean_token_accuracy": 0.08336958363652229,
|
|
"num_tokens": 1056774.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 8.207756519317627,
|
|
"epoch": 0.05448093167391413,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 7.8969,
|
|
"mean_token_accuracy": 0.09069060757756234,
|
|
"num_tokens": 1068056.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 8.064545297622681,
|
|
"epoch": 0.0549807567351427,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 7.8949,
|
|
"mean_token_accuracy": 0.09357433989644051,
|
|
"num_tokens": 1077698.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 8.047876501083374,
|
|
"epoch": 0.05548058179637127,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000277,
|
|
"loss": 7.8442,
|
|
"mean_token_accuracy": 0.09086285978555679,
|
|
"num_tokens": 1086485.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 8.150215339660644,
|
|
"epoch": 0.05598040685759984,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 7.9309,
|
|
"mean_token_accuracy": 0.08365092799067497,
|
|
"num_tokens": 1097329.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 8.007211875915527,
|
|
"epoch": 0.05648023191882841,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 7.8926,
|
|
"mean_token_accuracy": 0.08841023817658425,
|
|
"num_tokens": 1107990.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 8.063920402526856,
|
|
"epoch": 0.05698005698005698,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 7.8199,
|
|
"mean_token_accuracy": 0.09871476218104362,
|
|
"num_tokens": 1117798.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 7.987555932998657,
|
|
"epoch": 0.05747988204128555,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000287,
|
|
"loss": 7.9727,
|
|
"mean_token_accuracy": 0.08568176589906215,
|
|
"num_tokens": 1126976.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 8.147423028945923,
|
|
"epoch": 0.05797970710251412,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 7.8779,
|
|
"mean_token_accuracy": 0.09310147389769555,
|
|
"num_tokens": 1136410.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 8.087569284439088,
|
|
"epoch": 0.05847953216374269,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000292,
|
|
"loss": 7.909,
|
|
"mean_token_accuracy": 0.09139843881130219,
|
|
"num_tokens": 1147162.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 8.01126503944397,
|
|
"epoch": 0.05897935722497126,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 7.8668,
|
|
"mean_token_accuracy": 0.09075593948364258,
|
|
"num_tokens": 1157578.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 8.09571590423584,
|
|
"epoch": 0.05947918228619983,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000297,
|
|
"loss": 7.7723,
|
|
"mean_token_accuracy": 0.09766411408782005,
|
|
"num_tokens": 1166914.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 8.103814792633056,
|
|
"epoch": 0.0599790073474284,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 7.9517,
|
|
"mean_token_accuracy": 0.07924456372857094,
|
|
"num_tokens": 1176740.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 7.950633239746094,
|
|
"epoch": 0.06047883240865697,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000302,
|
|
"loss": 7.8391,
|
|
"mean_token_accuracy": 0.08747319504618645,
|
|
"num_tokens": 1185497.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 8.162421655654907,
|
|
"epoch": 0.06097865746988554,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 7.9355,
|
|
"mean_token_accuracy": 0.08849193602800369,
|
|
"num_tokens": 1197390.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 7.9715574264526365,
|
|
"epoch": 0.06147848253111411,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000307,
|
|
"loss": 7.8416,
|
|
"mean_token_accuracy": 0.09039048105478287,
|
|
"num_tokens": 1208293.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 8.039429759979248,
|
|
"epoch": 0.06197830759234268,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 7.8144,
|
|
"mean_token_accuracy": 0.09343133047223091,
|
|
"num_tokens": 1217860.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 7.971710109710694,
|
|
"epoch": 0.06247813265357125,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000312,
|
|
"loss": 7.8269,
|
|
"mean_token_accuracy": 0.09038410261273384,
|
|
"num_tokens": 1227259.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 7.8878796100616455,
|
|
"epoch": 0.06297795771479982,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 7.731,
|
|
"mean_token_accuracy": 0.09711387455463409,
|
|
"num_tokens": 1236775.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 8.014493560791015,
|
|
"epoch": 0.06347778277602839,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000317,
|
|
"loss": 7.8452,
|
|
"mean_token_accuracy": 0.09392437562346459,
|
|
"num_tokens": 1246307.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 8.040218830108643,
|
|
"epoch": 0.06397760783725696,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 7.8016,
|
|
"mean_token_accuracy": 0.09335492551326752,
|
|
"num_tokens": 1255197.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 7.904662942886352,
|
|
"epoch": 0.06447743289848552,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000322,
|
|
"loss": 7.8099,
|
|
"mean_token_accuracy": 0.09469617754220963,
|
|
"num_tokens": 1265056.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 7.957584476470947,
|
|
"epoch": 0.0649772579597141,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 7.7651,
|
|
"mean_token_accuracy": 0.09336479231715203,
|
|
"num_tokens": 1274718.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 8.030276775360107,
|
|
"epoch": 0.06547708302094267,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 7.8535,
|
|
"mean_token_accuracy": 0.09047859162092209,
|
|
"num_tokens": 1284364.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 7.932621622085572,
|
|
"epoch": 0.06597690808217124,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 7.8481,
|
|
"mean_token_accuracy": 0.08901502788066865,
|
|
"num_tokens": 1294186.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 7.92815752029419,
|
|
"epoch": 0.06647673314339982,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 7.749,
|
|
"mean_token_accuracy": 0.09501210823655129,
|
|
"num_tokens": 1303393.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 7.907820129394532,
|
|
"epoch": 0.06697655820462838,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 7.759,
|
|
"mean_token_accuracy": 0.08786283545196057,
|
|
"num_tokens": 1312819.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 7.962007904052735,
|
|
"epoch": 0.06747638326585695,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000337,
|
|
"loss": 7.8036,
|
|
"mean_token_accuracy": 0.08635997176170349,
|
|
"num_tokens": 1323001.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 7.896900320053101,
|
|
"epoch": 0.06797620832708552,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 7.7621,
|
|
"mean_token_accuracy": 0.0913906916975975,
|
|
"num_tokens": 1333260.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 7.922715091705323,
|
|
"epoch": 0.0684760333883141,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000342,
|
|
"loss": 7.752,
|
|
"mean_token_accuracy": 0.09890134558081627,
|
|
"num_tokens": 1342696.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 7.876401472091675,
|
|
"epoch": 0.06897585844954265,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 7.7177,
|
|
"mean_token_accuracy": 0.09715954437851906,
|
|
"num_tokens": 1353104.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 7.830476951599121,
|
|
"epoch": 0.06947568351077123,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000347,
|
|
"loss": 7.7889,
|
|
"mean_token_accuracy": 0.09046038314700126,
|
|
"num_tokens": 1362871.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 7.993965768814087,
|
|
"epoch": 0.0699755085719998,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 7.779,
|
|
"mean_token_accuracy": 0.09142390564084053,
|
|
"num_tokens": 1373622.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 7.7379943370819095,
|
|
"epoch": 0.07047533363322837,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000352,
|
|
"loss": 7.6555,
|
|
"mean_token_accuracy": 0.09681056961417198,
|
|
"num_tokens": 1382707.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 7.919323539733886,
|
|
"epoch": 0.07097515869445695,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 7.7344,
|
|
"mean_token_accuracy": 0.0933229923248291,
|
|
"num_tokens": 1392048.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 7.728861141204834,
|
|
"epoch": 0.0714749837556855,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000357,
|
|
"loss": 7.6032,
|
|
"mean_token_accuracy": 0.09870319068431854,
|
|
"num_tokens": 1401387.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 7.828720331192017,
|
|
"epoch": 0.07197480881691408,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 7.7248,
|
|
"mean_token_accuracy": 0.09169024452567101,
|
|
"num_tokens": 1411404.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 7.813608837127686,
|
|
"epoch": 0.07247463387814265,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000362,
|
|
"loss": 7.7136,
|
|
"mean_token_accuracy": 0.0950439676642418,
|
|
"num_tokens": 1420292.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 7.8692755699157715,
|
|
"epoch": 0.07297445893937123,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 7.6719,
|
|
"mean_token_accuracy": 0.0976124607026577,
|
|
"num_tokens": 1429990.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 7.831142568588257,
|
|
"epoch": 0.07347428400059978,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000367,
|
|
"loss": 7.7069,
|
|
"mean_token_accuracy": 0.09147948920726776,
|
|
"num_tokens": 1440148.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 7.764329719543457,
|
|
"epoch": 0.07397410906182836,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 7.6201,
|
|
"mean_token_accuracy": 0.09635840207338334,
|
|
"num_tokens": 1450015.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 7.804030704498291,
|
|
"epoch": 0.07447393412305693,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000372,
|
|
"loss": 7.6751,
|
|
"mean_token_accuracy": 0.0914132259786129,
|
|
"num_tokens": 1459582.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 7.778511476516724,
|
|
"epoch": 0.0749737591842855,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 7.7152,
|
|
"mean_token_accuracy": 0.09422787129878998,
|
|
"num_tokens": 1468151.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 7.706881713867188,
|
|
"epoch": 0.07547358424551406,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000377,
|
|
"loss": 7.6081,
|
|
"mean_token_accuracy": 0.10060534551739693,
|
|
"num_tokens": 1479099.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 7.84734468460083,
|
|
"epoch": 0.07597340930674264,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 7.7093,
|
|
"mean_token_accuracy": 0.0981033094227314,
|
|
"num_tokens": 1488060.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 7.740744495391846,
|
|
"epoch": 0.07647323436797121,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000382,
|
|
"loss": 7.7024,
|
|
"mean_token_accuracy": 0.08865071311593056,
|
|
"num_tokens": 1498230.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 7.745259141921997,
|
|
"epoch": 0.07697305942919978,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 7.6757,
|
|
"mean_token_accuracy": 0.09528728574514389,
|
|
"num_tokens": 1507962.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 7.917555141448974,
|
|
"epoch": 0.07747288449042836,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 7.7555,
|
|
"mean_token_accuracy": 0.09370075911283493,
|
|
"num_tokens": 1518607.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 7.787389945983887,
|
|
"epoch": 0.07797270955165692,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 7.6177,
|
|
"mean_token_accuracy": 0.09257503524422646,
|
|
"num_tokens": 1527166.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 7.814159679412842,
|
|
"epoch": 0.07847253461288549,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 7.7204,
|
|
"mean_token_accuracy": 0.09208195880055428,
|
|
"num_tokens": 1536683.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 7.740466594696045,
|
|
"epoch": 0.07897235967411406,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 7.6038,
|
|
"mean_token_accuracy": 0.09592815190553665,
|
|
"num_tokens": 1546243.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 7.751312637329102,
|
|
"epoch": 0.07947218473534264,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 7.6224,
|
|
"mean_token_accuracy": 0.09321612268686294,
|
|
"num_tokens": 1555916.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 7.638057899475098,
|
|
"epoch": 0.0799720097965712,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 7.5463,
|
|
"mean_token_accuracy": 0.1019765131175518,
|
|
"num_tokens": 1566617.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 7.6898962497711185,
|
|
"epoch": 0.08047183485779977,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000402,
|
|
"loss": 7.5999,
|
|
"mean_token_accuracy": 0.09289236888289451,
|
|
"num_tokens": 1575413.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 7.713872575759888,
|
|
"epoch": 0.08097165991902834,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 7.613,
|
|
"mean_token_accuracy": 0.09581428095698356,
|
|
"num_tokens": 1586863.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 7.698768568038941,
|
|
"epoch": 0.08147148498025691,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 7.6881,
|
|
"mean_token_accuracy": 0.09687362313270569,
|
|
"num_tokens": 1597685.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 7.732133245468139,
|
|
"epoch": 0.08197131004148547,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 7.5717,
|
|
"mean_token_accuracy": 0.09641181379556656,
|
|
"num_tokens": 1607103.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 7.7222553253173825,
|
|
"epoch": 0.08247113510271405,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000412,
|
|
"loss": 7.615,
|
|
"mean_token_accuracy": 0.09867498949170113,
|
|
"num_tokens": 1617919.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 7.688009023666382,
|
|
"epoch": 0.08297096016394262,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 7.6484,
|
|
"mean_token_accuracy": 0.0951889581978321,
|
|
"num_tokens": 1627293.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 7.735793828964233,
|
|
"epoch": 0.08347078522517119,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000417,
|
|
"loss": 7.6721,
|
|
"mean_token_accuracy": 0.09414145052433014,
|
|
"num_tokens": 1638440.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 7.663995361328125,
|
|
"epoch": 0.08397061028639977,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 7.5891,
|
|
"mean_token_accuracy": 0.09871478825807571,
|
|
"num_tokens": 1648148.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 7.626646089553833,
|
|
"epoch": 0.08447043534762833,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000422,
|
|
"loss": 7.5982,
|
|
"mean_token_accuracy": 0.10008721724152565,
|
|
"num_tokens": 1658334.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 7.790194845199585,
|
|
"epoch": 0.0849702604088569,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 7.6443,
|
|
"mean_token_accuracy": 0.09095977023243904,
|
|
"num_tokens": 1667963.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 7.602978897094727,
|
|
"epoch": 0.08547008547008547,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000427,
|
|
"loss": 7.5584,
|
|
"mean_token_accuracy": 0.09558436721563339,
|
|
"num_tokens": 1677270.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 7.686446809768677,
|
|
"epoch": 0.08596991053131404,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 7.5341,
|
|
"mean_token_accuracy": 0.1028837390244007,
|
|
"num_tokens": 1686770.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 7.618714475631714,
|
|
"epoch": 0.0864697355925426,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000432,
|
|
"loss": 7.6122,
|
|
"mean_token_accuracy": 0.093927763402462,
|
|
"num_tokens": 1697507.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 7.563453435897827,
|
|
"epoch": 0.08696956065377118,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 7.5464,
|
|
"mean_token_accuracy": 0.09653227850794792,
|
|
"num_tokens": 1708253.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 7.721293497085571,
|
|
"epoch": 0.08746938571499975,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000437,
|
|
"loss": 7.6172,
|
|
"mean_token_accuracy": 0.09264569208025933,
|
|
"num_tokens": 1717827.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 7.440291261672973,
|
|
"epoch": 0.08796921077622832,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 7.4133,
|
|
"mean_token_accuracy": 0.10668823942542076,
|
|
"num_tokens": 1726678.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 7.65731291770935,
|
|
"epoch": 0.0884690358374569,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.000442,
|
|
"loss": 7.5619,
|
|
"mean_token_accuracy": 0.09504320323467255,
|
|
"num_tokens": 1737518.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 7.595208358764649,
|
|
"epoch": 0.08896886089868546,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 7.4525,
|
|
"mean_token_accuracy": 0.10350560992956162,
|
|
"num_tokens": 1746338.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 7.5602888584136965,
|
|
"epoch": 0.08946868595991403,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000447,
|
|
"loss": 7.5298,
|
|
"mean_token_accuracy": 0.10117225348949432,
|
|
"num_tokens": 1756538.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 7.619009017944336,
|
|
"epoch": 0.0899685110211426,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 7.5422,
|
|
"mean_token_accuracy": 0.09560575634241104,
|
|
"num_tokens": 1766722.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 7.486926317214966,
|
|
"epoch": 0.09046833608237118,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 7.4575,
|
|
"mean_token_accuracy": 0.1022283561527729,
|
|
"num_tokens": 1777039.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 7.649388551712036,
|
|
"epoch": 0.09096816114359974,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 7.5762,
|
|
"mean_token_accuracy": 0.09183706119656562,
|
|
"num_tokens": 1787621.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 7.558070182800293,
|
|
"epoch": 0.09146798620482831,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 7.5862,
|
|
"mean_token_accuracy": 0.09815951213240623,
|
|
"num_tokens": 1798345.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 7.605602025985718,
|
|
"epoch": 0.09196781126605688,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 7.5492,
|
|
"mean_token_accuracy": 0.10024880468845368,
|
|
"num_tokens": 1807994.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 7.5432651996612545,
|
|
"epoch": 0.09246763632728545,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000462,
|
|
"loss": 7.4606,
|
|
"mean_token_accuracy": 0.09140576496720314,
|
|
"num_tokens": 1817965.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 7.537735462188721,
|
|
"epoch": 0.09296746138851401,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 7.5073,
|
|
"mean_token_accuracy": 0.09909802079200744,
|
|
"num_tokens": 1827094.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 7.568693351745606,
|
|
"epoch": 0.09346728644974259,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000467,
|
|
"loss": 7.4965,
|
|
"mean_token_accuracy": 0.09650534391403198,
|
|
"num_tokens": 1837754.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 7.49898247718811,
|
|
"epoch": 0.09396711151097116,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 7.4844,
|
|
"mean_token_accuracy": 0.09962845072150231,
|
|
"num_tokens": 1847500.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 7.550319385528565,
|
|
"epoch": 0.09446693657219973,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000472,
|
|
"loss": 7.5407,
|
|
"mean_token_accuracy": 0.09482819139957428,
|
|
"num_tokens": 1857663.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 7.619623422622681,
|
|
"epoch": 0.0949667616334283,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 7.5932,
|
|
"mean_token_accuracy": 0.09658025428652764,
|
|
"num_tokens": 1867752.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 7.511770153045655,
|
|
"epoch": 0.09546658669465687,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000477,
|
|
"loss": 7.5312,
|
|
"mean_token_accuracy": 0.09528448730707169,
|
|
"num_tokens": 1877479.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 7.57475905418396,
|
|
"epoch": 0.09596641175588544,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 7.5499,
|
|
"mean_token_accuracy": 0.09801071211695671,
|
|
"num_tokens": 1887944.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 7.543986701965332,
|
|
"epoch": 0.09646623681711401,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000482,
|
|
"loss": 7.462,
|
|
"mean_token_accuracy": 0.10098568946123124,
|
|
"num_tokens": 1896991.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 7.511151647567749,
|
|
"epoch": 0.09696606187834259,
|
|
"grad_norm": 1.796875,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 7.4906,
|
|
"mean_token_accuracy": 0.10490255281329156,
|
|
"num_tokens": 1907216.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 7.548860692977906,
|
|
"epoch": 0.09746588693957114,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000487,
|
|
"loss": 7.4929,
|
|
"mean_token_accuracy": 0.09408583045005799,
|
|
"num_tokens": 1917206.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 7.441623258590698,
|
|
"epoch": 0.09796571200079972,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 7.5821,
|
|
"mean_token_accuracy": 0.09114983677864075,
|
|
"num_tokens": 1929669.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 7.6039916515350345,
|
|
"epoch": 0.09846553706202829,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000492,
|
|
"loss": 7.4331,
|
|
"mean_token_accuracy": 0.10135036930441857,
|
|
"num_tokens": 1939881.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 7.386138916015625,
|
|
"epoch": 0.09896536212325686,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 7.4611,
|
|
"mean_token_accuracy": 0.09856282249093055,
|
|
"num_tokens": 1949423.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 7.534242963790893,
|
|
"epoch": 0.09946518718448542,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000497,
|
|
"loss": 7.5083,
|
|
"mean_token_accuracy": 0.09706381112337112,
|
|
"num_tokens": 1959214.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 7.59117021560669,
|
|
"epoch": 0.099965012245714,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 7.4778,
|
|
"mean_token_accuracy": 0.09641126468777657,
|
|
"num_tokens": 1968963.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 7.452836179733277,
|
|
"epoch": 0.10046483730694257,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999999981884985,
|
|
"loss": 7.4616,
|
|
"mean_token_accuracy": 0.10010318234562873,
|
|
"num_tokens": 1979551.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 7.487920141220092,
|
|
"epoch": 0.10096466236817114,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004999999908292739,
|
|
"loss": 7.4158,
|
|
"mean_token_accuracy": 0.10269866734743119,
|
|
"num_tokens": 1988432.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 7.458759784698486,
|
|
"epoch": 0.10146448742939972,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004999999778091072,
|
|
"loss": 7.463,
|
|
"mean_token_accuracy": 0.10001362189650535,
|
|
"num_tokens": 1997509.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 7.585840272903442,
|
|
"epoch": 0.10196431249062828,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004999999591279991,
|
|
"loss": 7.6465,
|
|
"mean_token_accuracy": 0.0948883354663849,
|
|
"num_tokens": 2007406.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 7.562847900390625,
|
|
"epoch": 0.10246413755185685,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00049999993478595,
|
|
"loss": 7.4447,
|
|
"mean_token_accuracy": 0.09490034580230713,
|
|
"num_tokens": 2016760.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 7.384311056137085,
|
|
"epoch": 0.10296396261308542,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004999999047829604,
|
|
"loss": 7.472,
|
|
"mean_token_accuracy": 0.09740648344159127,
|
|
"num_tokens": 2027776.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 7.500061655044556,
|
|
"epoch": 0.103463787674314,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000499999869119031,
|
|
"loss": 7.4209,
|
|
"mean_token_accuracy": 0.10009032934904098,
|
|
"num_tokens": 2036955.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 7.439191389083862,
|
|
"epoch": 0.10396361273554255,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004999998277941629,
|
|
"loss": 7.3367,
|
|
"mean_token_accuracy": 0.1029628187417984,
|
|
"num_tokens": 2046118.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 7.363403367996216,
|
|
"epoch": 0.10446343779677113,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004999997808083569,
|
|
"loss": 7.4052,
|
|
"mean_token_accuracy": 0.0971752054989338,
|
|
"num_tokens": 2056266.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 7.50301456451416,
|
|
"epoch": 0.1049632628579997,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004999997281616144,
|
|
"loss": 7.4578,
|
|
"mean_token_accuracy": 0.10237492248415947,
|
|
"num_tokens": 2067416.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 7.432632303237915,
|
|
"epoch": 0.10546308791922827,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004999996698539366,
|
|
"loss": 7.436,
|
|
"mean_token_accuracy": 0.09863532111048698,
|
|
"num_tokens": 2076926.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 7.411193323135376,
|
|
"epoch": 0.10596291298045685,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000499999605885325,
|
|
"loss": 7.4407,
|
|
"mean_token_accuracy": 0.10239719152450562,
|
|
"num_tokens": 2088017.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 7.435057353973389,
|
|
"epoch": 0.1064627380416854,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004999995362557813,
|
|
"loss": 7.4726,
|
|
"mean_token_accuracy": 0.10063011795282364,
|
|
"num_tokens": 2097249.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 7.433179426193237,
|
|
"epoch": 0.10696256310291398,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000499999460965307,
|
|
"loss": 7.4226,
|
|
"mean_token_accuracy": 0.09669319242238998,
|
|
"num_tokens": 2105778.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 7.43740439414978,
|
|
"epoch": 0.10746238816414255,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999993800139043,
|
|
"loss": 7.4346,
|
|
"mean_token_accuracy": 0.10024228915572167,
|
|
"num_tokens": 2114710.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 7.437659692764282,
|
|
"epoch": 0.10796221322537113,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999992934015752,
|
|
"loss": 7.3675,
|
|
"mean_token_accuracy": 0.1006879135966301,
|
|
"num_tokens": 2124002.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 7.470679759979248,
|
|
"epoch": 0.10846203828659969,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004999992011283215,
|
|
"loss": 7.4596,
|
|
"mean_token_accuracy": 0.10169444680213928,
|
|
"num_tokens": 2134592.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 7.366870260238647,
|
|
"epoch": 0.10896186334782826,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999991031941459,
|
|
"loss": 7.3004,
|
|
"mean_token_accuracy": 0.10656143799424171,
|
|
"num_tokens": 2143641.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 7.437468814849853,
|
|
"epoch": 0.10946168840905683,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004999989995990509,
|
|
"loss": 7.4242,
|
|
"mean_token_accuracy": 0.09810039848089218,
|
|
"num_tokens": 2154738.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 7.3005725860595705,
|
|
"epoch": 0.1099615134702854,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999988903430389,
|
|
"loss": 7.3807,
|
|
"mean_token_accuracy": 0.10800745934247971,
|
|
"num_tokens": 2164365.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 7.38071026802063,
|
|
"epoch": 0.11046133853151396,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004999987754261126,
|
|
"loss": 7.256,
|
|
"mean_token_accuracy": 0.106747804582119,
|
|
"num_tokens": 2173185.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 7.309447431564331,
|
|
"epoch": 0.11096116359274254,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004999986548482751,
|
|
"loss": 7.3739,
|
|
"mean_token_accuracy": 0.10424032732844353,
|
|
"num_tokens": 2184717.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 7.390005302429199,
|
|
"epoch": 0.11146098865397111,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999985286095293,
|
|
"loss": 7.3417,
|
|
"mean_token_accuracy": 0.1053157016634941,
|
|
"num_tokens": 2193681.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 7.360709953308105,
|
|
"epoch": 0.11196081371519968,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004999983967098785,
|
|
"loss": 7.3575,
|
|
"mean_token_accuracy": 0.10127927213907242,
|
|
"num_tokens": 2203366.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 7.370172548294067,
|
|
"epoch": 0.11246063877642826,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004999982591493258,
|
|
"loss": 7.367,
|
|
"mean_token_accuracy": 0.10329278036952019,
|
|
"num_tokens": 2213982.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 7.42004804611206,
|
|
"epoch": 0.11296046383765682,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004999981159278749,
|
|
"loss": 7.4133,
|
|
"mean_token_accuracy": 0.0983288586139679,
|
|
"num_tokens": 2222897.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 7.345064115524292,
|
|
"epoch": 0.11346028889888539,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004999979670455292,
|
|
"loss": 7.3589,
|
|
"mean_token_accuracy": 0.10666252076625823,
|
|
"num_tokens": 2232472.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 7.34918098449707,
|
|
"epoch": 0.11396011396011396,
|
|
"grad_norm": 1.8515625,
|
|
"learning_rate": 0.0004999978125022926,
|
|
"loss": 7.3859,
|
|
"mean_token_accuracy": 0.1046237476170063,
|
|
"num_tokens": 2242666.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 7.447892522811889,
|
|
"epoch": 0.11445993902134254,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000499997652298169,
|
|
"loss": 7.4574,
|
|
"mean_token_accuracy": 0.09490783661603927,
|
|
"num_tokens": 2254586.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 7.438595724105835,
|
|
"epoch": 0.1149597640825711,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999974864331624,
|
|
"loss": 7.3063,
|
|
"mean_token_accuracy": 0.10331938341259957,
|
|
"num_tokens": 2264619.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 7.417053699493408,
|
|
"epoch": 0.11545958914379967,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999973149072768,
|
|
"loss": 7.3842,
|
|
"mean_token_accuracy": 0.09593874737620353,
|
|
"num_tokens": 2274114.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 7.170158815383911,
|
|
"epoch": 0.11595941420502824,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004999971377205167,
|
|
"loss": 7.2906,
|
|
"mean_token_accuracy": 0.10702734366059304,
|
|
"num_tokens": 2283887.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 7.404069900512695,
|
|
"epoch": 0.11645923926625681,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004999969548728864,
|
|
"loss": 7.3116,
|
|
"mean_token_accuracy": 0.10856308713555336,
|
|
"num_tokens": 2293090.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 7.295200777053833,
|
|
"epoch": 0.11695906432748537,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999967663643908,
|
|
"loss": 7.4181,
|
|
"mean_token_accuracy": 0.10627190247178078,
|
|
"num_tokens": 2302828.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 7.4557257175445555,
|
|
"epoch": 0.11745888938871395,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999965721950345,
|
|
"loss": 7.2855,
|
|
"mean_token_accuracy": 0.10752936452627182,
|
|
"num_tokens": 2311415.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 7.324864387512207,
|
|
"epoch": 0.11795871444994252,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004999963723648222,
|
|
"loss": 7.3818,
|
|
"mean_token_accuracy": 0.10108949542045594,
|
|
"num_tokens": 2321512.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 7.343777084350586,
|
|
"epoch": 0.1184585395111711,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000499996166873759,
|
|
"loss": 7.2702,
|
|
"mean_token_accuracy": 0.10485923290252686,
|
|
"num_tokens": 2331398.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 7.232042646408081,
|
|
"epoch": 0.11895836457239967,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004999959557218502,
|
|
"loss": 7.2743,
|
|
"mean_token_accuracy": 0.10458537936210632,
|
|
"num_tokens": 2341751.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 7.423449993133545,
|
|
"epoch": 0.11945818963362823,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999957389091012,
|
|
"loss": 7.2923,
|
|
"mean_token_accuracy": 0.10130665153264999,
|
|
"num_tokens": 2352538.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 7.314671564102173,
|
|
"epoch": 0.1199580146948568,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004999955164355171,
|
|
"loss": 7.3717,
|
|
"mean_token_accuracy": 0.09852588474750519,
|
|
"num_tokens": 2362433.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 7.369090938568116,
|
|
"epoch": 0.12045783975608537,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004999952883011038,
|
|
"loss": 7.3447,
|
|
"mean_token_accuracy": 0.09998081587255,
|
|
"num_tokens": 2372431.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 7.257458639144898,
|
|
"epoch": 0.12095766481731395,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000499995054505867,
|
|
"loss": 7.3622,
|
|
"mean_token_accuracy": 0.1072867564857006,
|
|
"num_tokens": 2382483.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 7.43514986038208,
|
|
"epoch": 0.1214574898785425,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999948150498124,
|
|
"loss": 7.3292,
|
|
"mean_token_accuracy": 0.09574694111943245,
|
|
"num_tokens": 2392835.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 7.278194236755371,
|
|
"epoch": 0.12195731493977108,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004999945699329462,
|
|
"loss": 7.3013,
|
|
"mean_token_accuracy": 0.09891690164804459,
|
|
"num_tokens": 2402305.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 7.221072483062744,
|
|
"epoch": 0.12245714000099965,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999943191552745,
|
|
"loss": 7.2329,
|
|
"mean_token_accuracy": 0.10779139772057533,
|
|
"num_tokens": 2412093.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 7.276859903335572,
|
|
"epoch": 0.12295696506222822,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999940627168037,
|
|
"loss": 7.2007,
|
|
"mean_token_accuracy": 0.11122392266988754,
|
|
"num_tokens": 2421349.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 7.224908828735352,
|
|
"epoch": 0.12345679012345678,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00049999380061754,
|
|
"loss": 7.1752,
|
|
"mean_token_accuracy": 0.11676926836371422,
|
|
"num_tokens": 2430672.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 7.250015306472778,
|
|
"epoch": 0.12395661518468536,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004999935328574903,
|
|
"loss": 7.276,
|
|
"mean_token_accuracy": 0.10409534648060799,
|
|
"num_tokens": 2439355.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 7.27557520866394,
|
|
"epoch": 0.12445644024591393,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999932594366613,
|
|
"loss": 7.2616,
|
|
"mean_token_accuracy": 0.11316191554069518,
|
|
"num_tokens": 2448860.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 7.245042085647583,
|
|
"epoch": 0.1249562653071425,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999929803550597,
|
|
"loss": 7.3048,
|
|
"mean_token_accuracy": 0.103059022128582,
|
|
"num_tokens": 2458276.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 7.231337165832519,
|
|
"epoch": 0.12545609036837108,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999926956126927,
|
|
"loss": 7.221,
|
|
"mean_token_accuracy": 0.1075199268758297,
|
|
"num_tokens": 2467791.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 7.315781545639038,
|
|
"epoch": 0.12595591542959964,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999924052095672,
|
|
"loss": 7.3538,
|
|
"mean_token_accuracy": 0.10306788422167301,
|
|
"num_tokens": 2477549.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 7.283037090301514,
|
|
"epoch": 0.12645574049082822,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004999921091456907,
|
|
"loss": 7.2505,
|
|
"mean_token_accuracy": 0.10314335748553276,
|
|
"num_tokens": 2487549.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 7.3570441722869875,
|
|
"epoch": 0.12695556555205678,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999918074210707,
|
|
"loss": 7.2521,
|
|
"mean_token_accuracy": 0.10893722623586655,
|
|
"num_tokens": 2496799.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 7.3477424621582035,
|
|
"epoch": 0.12745539061328534,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999915000357147,
|
|
"loss": 7.359,
|
|
"mean_token_accuracy": 0.10306784734129906,
|
|
"num_tokens": 2505795.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 7.267300128936768,
|
|
"epoch": 0.12795521567451393,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004999911869896305,
|
|
"loss": 7.1675,
|
|
"mean_token_accuracy": 0.10215755999088287,
|
|
"num_tokens": 2515289.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 7.293365049362182,
|
|
"epoch": 0.1284550407357425,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004999908682828258,
|
|
"loss": 7.3458,
|
|
"mean_token_accuracy": 0.09611370787024498,
|
|
"num_tokens": 2524563.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 7.298129272460938,
|
|
"epoch": 0.12895486579697105,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004999905439153089,
|
|
"loss": 7.2354,
|
|
"mean_token_accuracy": 0.10675213858485222,
|
|
"num_tokens": 2534057.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 7.170830202102661,
|
|
"epoch": 0.12945469085819963,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004999902138870878,
|
|
"loss": 7.2159,
|
|
"mean_token_accuracy": 0.11319980695843697,
|
|
"num_tokens": 2544646.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 7.339636278152466,
|
|
"epoch": 0.1299545159194282,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004999898781981707,
|
|
"loss": 7.2845,
|
|
"mean_token_accuracy": 0.10319616049528121,
|
|
"num_tokens": 2553843.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 7.230478286743164,
|
|
"epoch": 0.13045434098065678,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999895368485662,
|
|
"loss": 7.3418,
|
|
"mean_token_accuracy": 0.09818191155791282,
|
|
"num_tokens": 2564061.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 7.2596722602844235,
|
|
"epoch": 0.13095416604188534,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000499989189838283,
|
|
"loss": 7.2158,
|
|
"mean_token_accuracy": 0.1089217223227024,
|
|
"num_tokens": 2574758.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 7.273542308807373,
|
|
"epoch": 0.1314539911031139,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004999888371673295,
|
|
"loss": 7.2483,
|
|
"mean_token_accuracy": 0.10606862902641297,
|
|
"num_tokens": 2585213.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 7.301839447021484,
|
|
"epoch": 0.13195381616434249,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004999884788357147,
|
|
"loss": 7.3233,
|
|
"mean_token_accuracy": 0.10537952035665513,
|
|
"num_tokens": 2595511.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 7.240723466873169,
|
|
"epoch": 0.13245364122557104,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999881148434478,
|
|
"loss": 7.2557,
|
|
"mean_token_accuracy": 0.10606002658605576,
|
|
"num_tokens": 2605930.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 7.290654516220092,
|
|
"epoch": 0.13295346628679963,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004999877451905378,
|
|
"loss": 7.1118,
|
|
"mean_token_accuracy": 0.11601563394069672,
|
|
"num_tokens": 2614224.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 7.235601758956909,
|
|
"epoch": 0.1334532913480282,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999873698769941,
|
|
"loss": 7.3573,
|
|
"mean_token_accuracy": 0.09951390027999878,
|
|
"num_tokens": 2624879.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 7.25201997756958,
|
|
"epoch": 0.13395311640925675,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000499986988902826,
|
|
"loss": 7.2007,
|
|
"mean_token_accuracy": 0.1086399532854557,
|
|
"num_tokens": 2633836.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 7.1708330631256105,
|
|
"epoch": 0.13445294147048534,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004999866022680431,
|
|
"loss": 7.2625,
|
|
"mean_token_accuracy": 0.10410139337182045,
|
|
"num_tokens": 2643754.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 7.410620546340942,
|
|
"epoch": 0.1349527665317139,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999862099726552,
|
|
"loss": 7.3104,
|
|
"mean_token_accuracy": 0.10062430128455162,
|
|
"num_tokens": 2652708.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 7.168396949768066,
|
|
"epoch": 0.13545259159294248,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004999858120166721,
|
|
"loss": 7.2749,
|
|
"mean_token_accuracy": 0.10011175870895386,
|
|
"num_tokens": 2662353.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 7.309360504150391,
|
|
"epoch": 0.13595241665417104,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999854084001039,
|
|
"loss": 7.1765,
|
|
"mean_token_accuracy": 0.10962674990296364,
|
|
"num_tokens": 2672376.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 7.2507658958435055,
|
|
"epoch": 0.1364522417153996,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004999849991229608,
|
|
"loss": 7.2769,
|
|
"mean_token_accuracy": 0.09988101571798325,
|
|
"num_tokens": 2682115.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 7.25983362197876,
|
|
"epoch": 0.1369520667766282,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000499984584185253,
|
|
"loss": 7.2561,
|
|
"mean_token_accuracy": 0.10595569089055061,
|
|
"num_tokens": 2692429.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 7.227681732177734,
|
|
"epoch": 0.13745189183785675,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004999841635869909,
|
|
"loss": 7.2038,
|
|
"mean_token_accuracy": 0.10670246928930283,
|
|
"num_tokens": 2701373.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 7.357850885391235,
|
|
"epoch": 0.1379517168990853,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004999837373281852,
|
|
"loss": 7.3842,
|
|
"mean_token_accuracy": 0.09883339330554008,
|
|
"num_tokens": 2710717.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 7.200265598297119,
|
|
"epoch": 0.1384515419603139,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004999833054088465,
|
|
"loss": 7.238,
|
|
"mean_token_accuracy": 0.10604067370295525,
|
|
"num_tokens": 2720936.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 7.334603643417358,
|
|
"epoch": 0.13895136702154245,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999828678289858,
|
|
"loss": 7.189,
|
|
"mean_token_accuracy": 0.11226848065853119,
|
|
"num_tokens": 2730504.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 7.173786735534668,
|
|
"epoch": 0.13945119208277104,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000499982424588614,
|
|
"loss": 7.1891,
|
|
"mean_token_accuracy": 0.10550703704357148,
|
|
"num_tokens": 2739069.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 7.247704267501831,
|
|
"epoch": 0.1399510171439996,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999819756877422,
|
|
"loss": 7.2625,
|
|
"mean_token_accuracy": 0.10297654867172241,
|
|
"num_tokens": 2748941.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 7.265053176879883,
|
|
"epoch": 0.14045084220522816,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999815211263819,
|
|
"loss": 7.2332,
|
|
"mean_token_accuracy": 0.10695202872157097,
|
|
"num_tokens": 2758915.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 7.296236133575439,
|
|
"epoch": 0.14095066726645675,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999810609045444,
|
|
"loss": 7.2574,
|
|
"mean_token_accuracy": 0.10682384073734283,
|
|
"num_tokens": 2768972.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 7.190992832183838,
|
|
"epoch": 0.1414504923276853,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999805950222412,
|
|
"loss": 7.1027,
|
|
"mean_token_accuracy": 0.10736067742109298,
|
|
"num_tokens": 2777989.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 7.189531517028809,
|
|
"epoch": 0.1419503173889139,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999801234794843,
|
|
"loss": 7.2003,
|
|
"mean_token_accuracy": 0.11025712713599205,
|
|
"num_tokens": 2788227.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 7.243151473999023,
|
|
"epoch": 0.14245014245014245,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999796462762853,
|
|
"loss": 7.2024,
|
|
"mean_token_accuracy": 0.11139077916741372,
|
|
"num_tokens": 2797729.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 7.215033388137817,
|
|
"epoch": 0.142949967511371,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999791634126562,
|
|
"loss": 7.1988,
|
|
"mean_token_accuracy": 0.10713230073451996,
|
|
"num_tokens": 2807215.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 7.180344343185425,
|
|
"epoch": 0.1434497925725996,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004999786748886094,
|
|
"loss": 7.1224,
|
|
"mean_token_accuracy": 0.1083896778523922,
|
|
"num_tokens": 2816945.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 7.13452730178833,
|
|
"epoch": 0.14394961763382816,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999781807041569,
|
|
"loss": 7.2,
|
|
"mean_token_accuracy": 0.10707568377256393,
|
|
"num_tokens": 2827072.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 7.242374563217163,
|
|
"epoch": 0.14444944269505672,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999776808593113,
|
|
"loss": 7.1277,
|
|
"mean_token_accuracy": 0.10868938043713569,
|
|
"num_tokens": 2837928.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 7.085830926895142,
|
|
"epoch": 0.1449492677562853,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999771753540852,
|
|
"loss": 7.1543,
|
|
"mean_token_accuracy": 0.11084548160433769,
|
|
"num_tokens": 2847861.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 7.249849414825439,
|
|
"epoch": 0.14544909281751386,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999766641884912,
|
|
"loss": 7.1643,
|
|
"mean_token_accuracy": 0.1070607289671898,
|
|
"num_tokens": 2857885.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 7.199644470214844,
|
|
"epoch": 0.14594891787874245,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004999761473625422,
|
|
"loss": 7.1822,
|
|
"mean_token_accuracy": 0.10687580332159996,
|
|
"num_tokens": 2867004.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 7.189074420928955,
|
|
"epoch": 0.146448742939971,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999756248762513,
|
|
"loss": 7.0531,
|
|
"mean_token_accuracy": 0.1135893739759922,
|
|
"num_tokens": 2876285.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 7.109452152252198,
|
|
"epoch": 0.14694856800119957,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004999750967296315,
|
|
"loss": 7.2097,
|
|
"mean_token_accuracy": 0.10470318049192429,
|
|
"num_tokens": 2886967.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 7.199790811538696,
|
|
"epoch": 0.14744839306242816,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004999745629226962,
|
|
"loss": 7.1824,
|
|
"mean_token_accuracy": 0.10584060102701187,
|
|
"num_tokens": 2897137.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 7.102796745300293,
|
|
"epoch": 0.14794821812365672,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004999740234554589,
|
|
"loss": 7.1333,
|
|
"mean_token_accuracy": 0.114309261739254,
|
|
"num_tokens": 2907658.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 7.184971666336059,
|
|
"epoch": 0.1484480431848853,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999734783279327,
|
|
"loss": 7.1394,
|
|
"mean_token_accuracy": 0.1090355396270752,
|
|
"num_tokens": 2917801.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 7.183087253570557,
|
|
"epoch": 0.14894786824611386,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000499972927540132,
|
|
"loss": 7.2457,
|
|
"mean_token_accuracy": 0.10699436143040657,
|
|
"num_tokens": 2927188.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 7.182110166549682,
|
|
"epoch": 0.14944769330734242,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004999723710920703,
|
|
"loss": 7.1084,
|
|
"mean_token_accuracy": 0.1068830780684948,
|
|
"num_tokens": 2936781.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 7.15032000541687,
|
|
"epoch": 0.149947518368571,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004999718089837616,
|
|
"loss": 7.0845,
|
|
"mean_token_accuracy": 0.1083772599697113,
|
|
"num_tokens": 2946426.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 7.071782922744751,
|
|
"epoch": 0.15044734342979957,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00049997124121522,
|
|
"loss": 7.1271,
|
|
"mean_token_accuracy": 0.10690554976463318,
|
|
"num_tokens": 2956246.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 7.193260526657104,
|
|
"epoch": 0.15094716849102813,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999706677864599,
|
|
"loss": 7.1187,
|
|
"mean_token_accuracy": 0.10903280302882194,
|
|
"num_tokens": 2965930.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 7.124616432189941,
|
|
"epoch": 0.15144699355225671,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004999700886974958,
|
|
"loss": 7.0448,
|
|
"mean_token_accuracy": 0.1127635382115841,
|
|
"num_tokens": 2975975.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 7.083345556259156,
|
|
"epoch": 0.15194681861348527,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000499969503948342,
|
|
"loss": 7.0777,
|
|
"mean_token_accuracy": 0.11200629472732544,
|
|
"num_tokens": 2985343.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 7.118203258514404,
|
|
"epoch": 0.15244664367471386,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004999689135390134,
|
|
"loss": 7.0889,
|
|
"mean_token_accuracy": 0.10904469192028046,
|
|
"num_tokens": 2994420.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 7.146843528747558,
|
|
"epoch": 0.15294646873594242,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999683174695249,
|
|
"loss": 7.2178,
|
|
"mean_token_accuracy": 0.10726030841469765,
|
|
"num_tokens": 3004447.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 7.2564233303070065,
|
|
"epoch": 0.15344629379717098,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004999677157398914,
|
|
"loss": 7.0998,
|
|
"mean_token_accuracy": 0.10736021399497986,
|
|
"num_tokens": 3014596.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 7.050740432739258,
|
|
"epoch": 0.15394611885839957,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004999671083501281,
|
|
"loss": 7.0311,
|
|
"mean_token_accuracy": 0.11356010809540748,
|
|
"num_tokens": 3024528.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 7.148781967163086,
|
|
"epoch": 0.15444594391962813,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004999664953002502,
|
|
"loss": 7.1504,
|
|
"mean_token_accuracy": 0.10772878602147103,
|
|
"num_tokens": 3035233.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 7.117992353439331,
|
|
"epoch": 0.1549457689808567,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000499965876590273,
|
|
"loss": 7.148,
|
|
"mean_token_accuracy": 0.11055759713053703,
|
|
"num_tokens": 3045890.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 7.103748226165772,
|
|
"epoch": 0.15544559404208527,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004999652522202125,
|
|
"loss": 7.0532,
|
|
"mean_token_accuracy": 0.11696677058935165,
|
|
"num_tokens": 3056640.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 7.117296743392944,
|
|
"epoch": 0.15594541910331383,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004999646221900839,
|
|
"loss": 7.0671,
|
|
"mean_token_accuracy": 0.11334137469530106,
|
|
"num_tokens": 3066938.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 7.140928316116333,
|
|
"epoch": 0.15644524416454242,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004999639864999034,
|
|
"loss": 7.1843,
|
|
"mean_token_accuracy": 0.10920159071683884,
|
|
"num_tokens": 3076791.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 7.207313203811646,
|
|
"epoch": 0.15694506922577098,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000499963345149687,
|
|
"loss": 7.0487,
|
|
"mean_token_accuracy": 0.11203598901629448,
|
|
"num_tokens": 3087532.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 7.021874189376831,
|
|
"epoch": 0.15744489428699954,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004999626981394506,
|
|
"loss": 7.1736,
|
|
"mean_token_accuracy": 0.11182766035199165,
|
|
"num_tokens": 3097465.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 7.218239116668701,
|
|
"epoch": 0.15794471934822812,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999620454692106,
|
|
"loss": 7.0643,
|
|
"mean_token_accuracy": 0.11003529652953148,
|
|
"num_tokens": 3107162.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 7.055271244049072,
|
|
"epoch": 0.15844454440945668,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999613871389836,
|
|
"loss": 7.0638,
|
|
"mean_token_accuracy": 0.10946442261338234,
|
|
"num_tokens": 3117154.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 7.16115403175354,
|
|
"epoch": 0.15894436947068527,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004999607231487858,
|
|
"loss": 7.0747,
|
|
"mean_token_accuracy": 0.10850517675280572,
|
|
"num_tokens": 3127635.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 7.163178443908691,
|
|
"epoch": 0.15944419453191383,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000499960053498634,
|
|
"loss": 7.1198,
|
|
"mean_token_accuracy": 0.10815898403525352,
|
|
"num_tokens": 3137806.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 7.110098171234131,
|
|
"epoch": 0.1599440195931424,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004999593781885454,
|
|
"loss": 7.1597,
|
|
"mean_token_accuracy": 0.10361172184348107,
|
|
"num_tokens": 3147704.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 7.177382183074951,
|
|
"epoch": 0.16044384465437098,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004999586972185366,
|
|
"loss": 7.1209,
|
|
"mean_token_accuracy": 0.106394574791193,
|
|
"num_tokens": 3158486.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 7.113349390029907,
|
|
"epoch": 0.16094366971559954,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004999580105886248,
|
|
"loss": 7.0974,
|
|
"mean_token_accuracy": 0.1099392257630825,
|
|
"num_tokens": 3168518.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 7.062984132766724,
|
|
"epoch": 0.16144349477682812,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999573182988275,
|
|
"loss": 6.9995,
|
|
"mean_token_accuracy": 0.1146053321659565,
|
|
"num_tokens": 3177502.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 7.128028869628906,
|
|
"epoch": 0.16194331983805668,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999566203491619,
|
|
"loss": 7.0611,
|
|
"mean_token_accuracy": 0.10871210768818855,
|
|
"num_tokens": 3187408.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 7.144448757171631,
|
|
"epoch": 0.16244314489928524,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004999559167396455,
|
|
"loss": 7.1239,
|
|
"mean_token_accuracy": 0.1121080070734024,
|
|
"num_tokens": 3196812.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 7.011726713180542,
|
|
"epoch": 0.16294296996051383,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004999552074702963,
|
|
"loss": 6.9848,
|
|
"mean_token_accuracy": 0.11447064578533173,
|
|
"num_tokens": 3206159.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 7.103731536865235,
|
|
"epoch": 0.1634427950217424,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004999544925411318,
|
|
"loss": 6.9641,
|
|
"mean_token_accuracy": 0.11926594749093056,
|
|
"num_tokens": 3217329.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 7.0044941902160645,
|
|
"epoch": 0.16394262008297095,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004999537719521703,
|
|
"loss": 7.0956,
|
|
"mean_token_accuracy": 0.11171141639351845,
|
|
"num_tokens": 3227558.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 7.228980493545532,
|
|
"epoch": 0.16444244514419953,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999530457034297,
|
|
"loss": 7.1913,
|
|
"mean_token_accuracy": 0.10457537770271301,
|
|
"num_tokens": 3237719.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 7.0681822299957275,
|
|
"epoch": 0.1649422702054281,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999523137949283,
|
|
"loss": 7.1056,
|
|
"mean_token_accuracy": 0.10843203067779542,
|
|
"num_tokens": 3246483.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 7.033597707748413,
|
|
"epoch": 0.16544209526665668,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004999515762266846,
|
|
"loss": 6.9882,
|
|
"mean_token_accuracy": 0.11709940508008003,
|
|
"num_tokens": 3256955.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 7.148850917816162,
|
|
"epoch": 0.16594192032788524,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999508329987173,
|
|
"loss": 7.0666,
|
|
"mean_token_accuracy": 0.10870269536972046,
|
|
"num_tokens": 3266799.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 7.035959720611572,
|
|
"epoch": 0.1664417453891138,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999500841110447,
|
|
"loss": 7.0619,
|
|
"mean_token_accuracy": 0.1083013042807579,
|
|
"num_tokens": 3276595.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 7.106384897232056,
|
|
"epoch": 0.16694157045034239,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000499949329563686,
|
|
"loss": 7.0058,
|
|
"mean_token_accuracy": 0.11427201181650162,
|
|
"num_tokens": 3286047.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 7.13128981590271,
|
|
"epoch": 0.16744139551157095,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00049994856935666,
|
|
"loss": 7.1015,
|
|
"mean_token_accuracy": 0.10778715461492538,
|
|
"num_tokens": 3295364.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 6.995987176895142,
|
|
"epoch": 0.16794122057279953,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000499947803489986,
|
|
"loss": 6.9963,
|
|
"mean_token_accuracy": 0.11492641419172286,
|
|
"num_tokens": 3305319.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 7.075168037414551,
|
|
"epoch": 0.1684410456340281,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004999470319636829,
|
|
"loss": 7.02,
|
|
"mean_token_accuracy": 0.11598107293248176,
|
|
"num_tokens": 3315349.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 7.090064430236817,
|
|
"epoch": 0.16894087069525665,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004999462547777705,
|
|
"loss": 7.0895,
|
|
"mean_token_accuracy": 0.10790005549788476,
|
|
"num_tokens": 3325603.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 7.018190336227417,
|
|
"epoch": 0.16944069575648524,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999454719322682,
|
|
"loss": 7.0041,
|
|
"mean_token_accuracy": 0.11242652088403701,
|
|
"num_tokens": 3334693.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 7.13174524307251,
|
|
"epoch": 0.1699405208177138,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004999446834271958,
|
|
"loss": 7.0239,
|
|
"mean_token_accuracy": 0.10740584284067153,
|
|
"num_tokens": 3345259.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 6.988541936874389,
|
|
"epoch": 0.17044034587894236,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999438892625728,
|
|
"loss": 7.0064,
|
|
"mean_token_accuracy": 0.10854029953479767,
|
|
"num_tokens": 3354072.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 7.207061100006103,
|
|
"epoch": 0.17094017094017094,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999430894384198,
|
|
"loss": 6.9929,
|
|
"mean_token_accuracy": 0.11092336103320122,
|
|
"num_tokens": 3364881.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 6.9689531326293945,
|
|
"epoch": 0.1714399960013995,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999422839547564,
|
|
"loss": 6.9744,
|
|
"mean_token_accuracy": 0.11487517282366752,
|
|
"num_tokens": 3374030.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 7.004118824005127,
|
|
"epoch": 0.1719398210626281,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004999414728116029,
|
|
"loss": 6.9781,
|
|
"mean_token_accuracy": 0.11029164642095565,
|
|
"num_tokens": 3384009.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 7.1930708408355715,
|
|
"epoch": 0.17243964612385665,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999406560089798,
|
|
"loss": 7.1166,
|
|
"mean_token_accuracy": 0.10778723657131195,
|
|
"num_tokens": 3393660.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 7.073626899719239,
|
|
"epoch": 0.1729394711850852,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004999398335469078,
|
|
"loss": 7.1265,
|
|
"mean_token_accuracy": 0.11251236200332641,
|
|
"num_tokens": 3404364.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 7.160039472579956,
|
|
"epoch": 0.1734392962463138,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999390054254074,
|
|
"loss": 7.0297,
|
|
"mean_token_accuracy": 0.1090213879942894,
|
|
"num_tokens": 3412990.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 6.935382604598999,
|
|
"epoch": 0.17393912130754235,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004999381716444996,
|
|
"loss": 6.9892,
|
|
"mean_token_accuracy": 0.11384793221950532,
|
|
"num_tokens": 3422339.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 7.049477767944336,
|
|
"epoch": 0.17443894636877094,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004999373322042052,
|
|
"loss": 6.9789,
|
|
"mean_token_accuracy": 0.11349833980202675,
|
|
"num_tokens": 3433093.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 7.139243173599243,
|
|
"epoch": 0.1749387714299995,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004999364871045455,
|
|
"loss": 7.0647,
|
|
"mean_token_accuracy": 0.10994341224431992,
|
|
"num_tokens": 3443464.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 7.052545499801636,
|
|
"epoch": 0.17543859649122806,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004999356363455415,
|
|
"loss": 6.9928,
|
|
"mean_token_accuracy": 0.11283324584364891,
|
|
"num_tokens": 3453380.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 7.093253469467163,
|
|
"epoch": 0.17593842155245665,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000499934779927215,
|
|
"loss": 7.0465,
|
|
"mean_token_accuracy": 0.11127711459994316,
|
|
"num_tokens": 3463583.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 6.952481842041015,
|
|
"epoch": 0.1764382466136852,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004999339178495872,
|
|
"loss": 6.947,
|
|
"mean_token_accuracy": 0.120367631316185,
|
|
"num_tokens": 3473640.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 7.034084367752075,
|
|
"epoch": 0.1769380716749138,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999330501126798,
|
|
"loss": 7.0118,
|
|
"mean_token_accuracy": 0.11708684265613556,
|
|
"num_tokens": 3483827.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 7.028639554977417,
|
|
"epoch": 0.17743789673614235,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004999321767165148,
|
|
"loss": 6.9672,
|
|
"mean_token_accuracy": 0.12046092599630356,
|
|
"num_tokens": 3493763.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 7.066556453704834,
|
|
"epoch": 0.1779377217973709,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999312976611142,
|
|
"loss": 7.0814,
|
|
"mean_token_accuracy": 0.11231729686260224,
|
|
"num_tokens": 3503119.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 7.147392320632934,
|
|
"epoch": 0.1784375468585995,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004999304129465,
|
|
"loss": 7.0178,
|
|
"mean_token_accuracy": 0.1075978696346283,
|
|
"num_tokens": 3513235.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 7.093100214004517,
|
|
"epoch": 0.17893737191982806,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004999295225726945,
|
|
"loss": 7.1301,
|
|
"mean_token_accuracy": 0.10513050705194474,
|
|
"num_tokens": 3523633.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 7.043914461135865,
|
|
"epoch": 0.17943719698105662,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00049992862653972,
|
|
"loss": 6.9709,
|
|
"mean_token_accuracy": 0.11173391118645667,
|
|
"num_tokens": 3532888.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 7.064423990249634,
|
|
"epoch": 0.1799370220422852,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004999277248475992,
|
|
"loss": 6.9886,
|
|
"mean_token_accuracy": 0.11316701844334602,
|
|
"num_tokens": 3541942.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 7.019074440002441,
|
|
"epoch": 0.18043684710351376,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004999268174963547,
|
|
"loss": 6.9535,
|
|
"mean_token_accuracy": 0.11337752267718315,
|
|
"num_tokens": 3551544.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 6.995860528945923,
|
|
"epoch": 0.18093667216474235,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999259044860093,
|
|
"loss": 6.9649,
|
|
"mean_token_accuracy": 0.11671746969223022,
|
|
"num_tokens": 3561024.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 7.052825117111206,
|
|
"epoch": 0.1814364972259709,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999249858165861,
|
|
"loss": 6.9448,
|
|
"mean_token_accuracy": 0.11900644525885581,
|
|
"num_tokens": 3570962.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 6.871934175491333,
|
|
"epoch": 0.18193632228719947,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004999240614881082,
|
|
"loss": 6.9231,
|
|
"mean_token_accuracy": 0.11965297684073448,
|
|
"num_tokens": 3580669.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 7.107207250595093,
|
|
"epoch": 0.18243614734842806,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004999231315005987,
|
|
"loss": 6.9517,
|
|
"mean_token_accuracy": 0.1128486692905426,
|
|
"num_tokens": 3590627.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 6.98112735748291,
|
|
"epoch": 0.18293597240965662,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004999221958540811,
|
|
"loss": 7.0279,
|
|
"mean_token_accuracy": 0.10484765991568565,
|
|
"num_tokens": 3599872.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 7.036418867111206,
|
|
"epoch": 0.1834357974708852,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004999212545485789,
|
|
"loss": 7.0079,
|
|
"mean_token_accuracy": 0.11393587961792946,
|
|
"num_tokens": 3609809.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 7.130139350891113,
|
|
"epoch": 0.18393562253211376,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004999203075841159,
|
|
"loss": 7.0305,
|
|
"mean_token_accuracy": 0.11348742470145226,
|
|
"num_tokens": 3619385.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 6.894102430343628,
|
|
"epoch": 0.18443544759334232,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004999193549607157,
|
|
"loss": 6.8742,
|
|
"mean_token_accuracy": 0.12475180551409722,
|
|
"num_tokens": 3628867.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 7.0008539199829105,
|
|
"epoch": 0.1849352726545709,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999183966784026,
|
|
"loss": 6.8653,
|
|
"mean_token_accuracy": 0.12164018377661705,
|
|
"num_tokens": 3637528.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 6.99360876083374,
|
|
"epoch": 0.18543509771579947,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004999174327372004,
|
|
"loss": 7.0004,
|
|
"mean_token_accuracy": 0.11094492375850677,
|
|
"num_tokens": 3648011.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 7.009715986251831,
|
|
"epoch": 0.18593492277702803,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004999164631371335,
|
|
"loss": 6.9968,
|
|
"mean_token_accuracy": 0.11623972430825233,
|
|
"num_tokens": 3657648.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 7.061586046218872,
|
|
"epoch": 0.18643474783825661,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004999154878782262,
|
|
"loss": 6.9918,
|
|
"mean_token_accuracy": 0.11201639547944069,
|
|
"num_tokens": 3666965.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 7.05910816192627,
|
|
"epoch": 0.18693457289948517,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000499914506960503,
|
|
"loss": 7.0192,
|
|
"mean_token_accuracy": 0.10654903575778008,
|
|
"num_tokens": 3676609.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 6.931095027923584,
|
|
"epoch": 0.18743439796071376,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999135203839889,
|
|
"loss": 6.9009,
|
|
"mean_token_accuracy": 0.11747965291142463,
|
|
"num_tokens": 3685884.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 7.015621614456177,
|
|
"epoch": 0.18793422302194232,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999125281487084,
|
|
"loss": 7.0263,
|
|
"mean_token_accuracy": 0.11326177418231964,
|
|
"num_tokens": 3695436.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 7.053316402435303,
|
|
"epoch": 0.18843404808317088,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999115302546866,
|
|
"loss": 6.8708,
|
|
"mean_token_accuracy": 0.1211979329586029,
|
|
"num_tokens": 3704806.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 6.911389636993408,
|
|
"epoch": 0.18893387314439947,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0004999105267019486,
|
|
"loss": 6.9271,
|
|
"mean_token_accuracy": 0.1165336862206459,
|
|
"num_tokens": 3713927.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 7.068922758102417,
|
|
"epoch": 0.18943369820562803,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004999095174905195,
|
|
"loss": 7.0344,
|
|
"mean_token_accuracy": 0.1089947208762169,
|
|
"num_tokens": 3723685.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 6.937805414199829,
|
|
"epoch": 0.1899335232668566,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999085026204249,
|
|
"loss": 6.9937,
|
|
"mean_token_accuracy": 0.112867072224617,
|
|
"num_tokens": 3733918.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 7.059433317184448,
|
|
"epoch": 0.19043334832808517,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004999074820916903,
|
|
"loss": 6.9371,
|
|
"mean_token_accuracy": 0.11250495687127113,
|
|
"num_tokens": 3744466.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 6.984634971618652,
|
|
"epoch": 0.19093317338931373,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999064559043412,
|
|
"loss": 6.8978,
|
|
"mean_token_accuracy": 0.11224280893802643,
|
|
"num_tokens": 3754082.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 6.9544031620025635,
|
|
"epoch": 0.19143299845054232,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004999054240584037,
|
|
"loss": 6.9246,
|
|
"mean_token_accuracy": 0.11677428260445595,
|
|
"num_tokens": 3763111.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 7.080189561843872,
|
|
"epoch": 0.19193282351177088,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004999043865539035,
|
|
"loss": 6.9652,
|
|
"mean_token_accuracy": 0.11014738902449608,
|
|
"num_tokens": 3772555.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 7.0365804672241214,
|
|
"epoch": 0.19243264857299944,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999033433908667,
|
|
"loss": 7.0159,
|
|
"mean_token_accuracy": 0.11137505248188972,
|
|
"num_tokens": 3783073.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 6.855137062072754,
|
|
"epoch": 0.19293247363422802,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999022945693198,
|
|
"loss": 6.9465,
|
|
"mean_token_accuracy": 0.11666482612490654,
|
|
"num_tokens": 3792277.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 7.001844072341919,
|
|
"epoch": 0.19343229869545658,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000499901240089289,
|
|
"loss": 6.9055,
|
|
"mean_token_accuracy": 0.11711104214191437,
|
|
"num_tokens": 3801098.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 6.975520658493042,
|
|
"epoch": 0.19393212375668517,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999001799508009,
|
|
"loss": 6.9512,
|
|
"mean_token_accuracy": 0.11243590638041497,
|
|
"num_tokens": 3810614.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 6.998220157623291,
|
|
"epoch": 0.19443194881791373,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004998991141538821,
|
|
"loss": 6.9268,
|
|
"mean_token_accuracy": 0.11685932651162148,
|
|
"num_tokens": 3818989.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 6.912782478332519,
|
|
"epoch": 0.1949317738791423,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004998980426985596,
|
|
"loss": 6.9684,
|
|
"mean_token_accuracy": 0.11471363082528115,
|
|
"num_tokens": 3829319.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 7.026929330825806,
|
|
"epoch": 0.19543159894037088,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00049989696558486,
|
|
"loss": 6.9881,
|
|
"mean_token_accuracy": 0.11341556683182716,
|
|
"num_tokens": 3839545.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 6.994954204559326,
|
|
"epoch": 0.19593142400159944,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004998958828128108,
|
|
"loss": 6.9334,
|
|
"mean_token_accuracy": 0.11814818978309631,
|
|
"num_tokens": 3849029.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 6.998789691925049,
|
|
"epoch": 0.19643124906282802,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004998947943824389,
|
|
"loss": 6.9873,
|
|
"mean_token_accuracy": 0.1150098517537117,
|
|
"num_tokens": 3859463.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 7.099795150756836,
|
|
"epoch": 0.19693107412405658,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004998937002937719,
|
|
"loss": 7.0047,
|
|
"mean_token_accuracy": 0.1137129820883274,
|
|
"num_tokens": 3869695.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 6.943069839477539,
|
|
"epoch": 0.19743089918528514,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004998926005468373,
|
|
"loss": 6.9697,
|
|
"mean_token_accuracy": 0.11416480243206024,
|
|
"num_tokens": 3879608.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 7.00183253288269,
|
|
"epoch": 0.19793072424651373,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004998914951416627,
|
|
"loss": 6.8927,
|
|
"mean_token_accuracy": 0.11571493595838547,
|
|
"num_tokens": 3888709.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 6.841829586029053,
|
|
"epoch": 0.1984305493077423,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000499890384078276,
|
|
"loss": 6.8369,
|
|
"mean_token_accuracy": 0.11708793714642525,
|
|
"num_tokens": 3898392.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 6.987730503082275,
|
|
"epoch": 0.19893037436897085,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004998892673567051,
|
|
"loss": 7.0549,
|
|
"mean_token_accuracy": 0.10358659774065018,
|
|
"num_tokens": 3909508.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 7.001489162445068,
|
|
"epoch": 0.19943019943019943,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004998881449769781,
|
|
"loss": 6.8967,
|
|
"mean_token_accuracy": 0.11786960512399673,
|
|
"num_tokens": 3918155.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 7.019080638885498,
|
|
"epoch": 0.199930024491428,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004998870169391232,
|
|
"loss": 6.9337,
|
|
"mean_token_accuracy": 0.11424290388822556,
|
|
"num_tokens": 3928478.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 6.898446416854858,
|
|
"epoch": 0.20042984955265658,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000499885883243169,
|
|
"loss": 6.9702,
|
|
"mean_token_accuracy": 0.12253575176000595,
|
|
"num_tokens": 3938126.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 6.876339244842529,
|
|
"epoch": 0.20092967461388514,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004998847438891437,
|
|
"loss": 6.816,
|
|
"mean_token_accuracy": 0.11575676202774048,
|
|
"num_tokens": 3948176.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 7.003563070297242,
|
|
"epoch": 0.2014294996751137,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004998835988770761,
|
|
"loss": 6.9322,
|
|
"mean_token_accuracy": 0.11722335889935494,
|
|
"num_tokens": 3958983.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 6.995720767974854,
|
|
"epoch": 0.20192932473634229,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000499882448206995,
|
|
"loss": 6.8842,
|
|
"mean_token_accuracy": 0.11819658502936363,
|
|
"num_tokens": 3968766.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 6.937156057357788,
|
|
"epoch": 0.20242914979757085,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004998812918789293,
|
|
"loss": 6.9056,
|
|
"mean_token_accuracy": 0.11888515204191208,
|
|
"num_tokens": 3979487.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 6.9709443092346195,
|
|
"epoch": 0.20292897485879943,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004998801298929084,
|
|
"loss": 6.8821,
|
|
"mean_token_accuracy": 0.11899071708321571,
|
|
"num_tokens": 3988450.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 6.9239765167236325,
|
|
"epoch": 0.203428799920028,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004998789622489611,
|
|
"loss": 6.8673,
|
|
"mean_token_accuracy": 0.11786953210830689,
|
|
"num_tokens": 3997492.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 6.913021612167358,
|
|
"epoch": 0.20392862498125655,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004998777889471171,
|
|
"loss": 6.8757,
|
|
"mean_token_accuracy": 0.12316029518842697,
|
|
"num_tokens": 4007298.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 6.9504697799682615,
|
|
"epoch": 0.20442845004248514,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004998766099874056,
|
|
"loss": 6.9186,
|
|
"mean_token_accuracy": 0.11500022858381272,
|
|
"num_tokens": 4017727.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 6.997060394287109,
|
|
"epoch": 0.2049282751037137,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004998754253698566,
|
|
"loss": 6.9233,
|
|
"mean_token_accuracy": 0.1129599705338478,
|
|
"num_tokens": 4027618.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 6.9302326202392575,
|
|
"epoch": 0.20542810016494226,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004998742350944998,
|
|
"loss": 6.9392,
|
|
"mean_token_accuracy": 0.1133469320833683,
|
|
"num_tokens": 4038831.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 6.96685037612915,
|
|
"epoch": 0.20592792522617084,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004998730391613651,
|
|
"loss": 6.9081,
|
|
"mean_token_accuracy": 0.11602651923894883,
|
|
"num_tokens": 4048736.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 6.967009258270264,
|
|
"epoch": 0.2064277502873994,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004998718375704825,
|
|
"loss": 6.8127,
|
|
"mean_token_accuracy": 0.12039854824542999,
|
|
"num_tokens": 4057633.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 6.901197910308838,
|
|
"epoch": 0.206927575348628,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004998706303218825,
|
|
"loss": 6.9488,
|
|
"mean_token_accuracy": 0.12238204479217529,
|
|
"num_tokens": 4067883.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 7.028725862503052,
|
|
"epoch": 0.20742740040985655,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004998694174155952,
|
|
"loss": 6.8041,
|
|
"mean_token_accuracy": 0.12112294137477875,
|
|
"num_tokens": 4078147.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 6.877385187149048,
|
|
"epoch": 0.2079272254710851,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004998681988516512,
|
|
"loss": 6.7772,
|
|
"mean_token_accuracy": 0.12411027252674103,
|
|
"num_tokens": 4087130.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 6.799959278106689,
|
|
"epoch": 0.2084270505323137,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004998669746300811,
|
|
"loss": 6.913,
|
|
"mean_token_accuracy": 0.12494281083345413,
|
|
"num_tokens": 4096840.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 7.074980068206787,
|
|
"epoch": 0.20892687559354226,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004998657447509159,
|
|
"loss": 6.9152,
|
|
"mean_token_accuracy": 0.11520147994160652,
|
|
"num_tokens": 4105842.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 6.862983131408692,
|
|
"epoch": 0.20942670065477084,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004998645092141864,
|
|
"loss": 6.8277,
|
|
"mean_token_accuracy": 0.11900869160890579,
|
|
"num_tokens": 4114865.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 6.963934278488159,
|
|
"epoch": 0.2099265257159994,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004998632680199238,
|
|
"loss": 6.9615,
|
|
"mean_token_accuracy": 0.11277624815702439,
|
|
"num_tokens": 4124954.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 6.93606743812561,
|
|
"epoch": 0.21042635077722796,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004998620211681591,
|
|
"loss": 6.8275,
|
|
"mean_token_accuracy": 0.12393845915794373,
|
|
"num_tokens": 4134091.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 6.960611724853516,
|
|
"epoch": 0.21092617583845655,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004998607686589239,
|
|
"loss": 6.9149,
|
|
"mean_token_accuracy": 0.11502190008759498,
|
|
"num_tokens": 4143566.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 6.919644498825074,
|
|
"epoch": 0.2114260008996851,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004998595104922496,
|
|
"loss": 6.8746,
|
|
"mean_token_accuracy": 0.11764942035079003,
|
|
"num_tokens": 4153053.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 6.968874740600586,
|
|
"epoch": 0.2119258259609137,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004998582466681678,
|
|
"loss": 6.855,
|
|
"mean_token_accuracy": 0.12172348126769066,
|
|
"num_tokens": 4162925.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 6.78645544052124,
|
|
"epoch": 0.21242565102214225,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004998569771867105,
|
|
"loss": 6.7861,
|
|
"mean_token_accuracy": 0.12430796623229981,
|
|
"num_tokens": 4172077.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 6.884827756881714,
|
|
"epoch": 0.2129254760833708,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004998557020479095,
|
|
"loss": 6.8712,
|
|
"mean_token_accuracy": 0.11964266151189804,
|
|
"num_tokens": 4182134.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 7.031274938583374,
|
|
"epoch": 0.2134253011445994,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004998544212517968,
|
|
"loss": 6.8856,
|
|
"mean_token_accuracy": 0.11359192878007889,
|
|
"num_tokens": 4191963.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 6.858419895172119,
|
|
"epoch": 0.21392512620582796,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004998531347984047,
|
|
"loss": 6.8442,
|
|
"mean_token_accuracy": 0.11866675019264221,
|
|
"num_tokens": 4202180.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 6.953751468658448,
|
|
"epoch": 0.21442495126705652,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004998518426877656,
|
|
"loss": 6.8626,
|
|
"mean_token_accuracy": 0.11622112393379211,
|
|
"num_tokens": 4211874.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 6.913810157775879,
|
|
"epoch": 0.2149247763282851,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004998505449199122,
|
|
"loss": 6.8299,
|
|
"mean_token_accuracy": 0.11711556166410446,
|
|
"num_tokens": 4221610.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 6.752934122085572,
|
|
"epoch": 0.21542460138951366,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004998492414948768,
|
|
"loss": 6.7874,
|
|
"mean_token_accuracy": 0.12737127915024757,
|
|
"num_tokens": 4230610.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 6.917922449111939,
|
|
"epoch": 0.21592442645074225,
|
|
"grad_norm": 0.81640625,
|
|
"learning_rate": 0.0004998479324126924,
|
|
"loss": 6.9086,
|
|
"mean_token_accuracy": 0.11124092862010002,
|
|
"num_tokens": 4241389.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 6.996273183822632,
|
|
"epoch": 0.2164242515119708,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004998466176733918,
|
|
"loss": 6.8422,
|
|
"mean_token_accuracy": 0.117838254570961,
|
|
"num_tokens": 4250617.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 6.813162040710449,
|
|
"epoch": 0.21692407657319937,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004998452972770082,
|
|
"loss": 6.8345,
|
|
"mean_token_accuracy": 0.12334902957081795,
|
|
"num_tokens": 4260000.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 6.872658777236938,
|
|
"epoch": 0.21742390163442796,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004998439712235747,
|
|
"loss": 6.7642,
|
|
"mean_token_accuracy": 0.12139535620808602,
|
|
"num_tokens": 4269751.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 6.920974588394165,
|
|
"epoch": 0.21792372669565652,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004998426395131248,
|
|
"loss": 6.8374,
|
|
"mean_token_accuracy": 0.12188360318541527,
|
|
"num_tokens": 4279210.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 6.908169555664062,
|
|
"epoch": 0.2184235517568851,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004998413021456919,
|
|
"loss": 6.7937,
|
|
"mean_token_accuracy": 0.12465235516428948,
|
|
"num_tokens": 4288573.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 6.862871980667114,
|
|
"epoch": 0.21892337681811366,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004998399591213098,
|
|
"loss": 6.8086,
|
|
"mean_token_accuracy": 0.1203602097928524,
|
|
"num_tokens": 4298438.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 6.8648326873779295,
|
|
"epoch": 0.21942320187934222,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004998386104400122,
|
|
"loss": 6.7329,
|
|
"mean_token_accuracy": 0.1275396764278412,
|
|
"num_tokens": 4307174.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 6.967341136932373,
|
|
"epoch": 0.2199230269405708,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004998372561018329,
|
|
"loss": 6.9337,
|
|
"mean_token_accuracy": 0.11480155438184739,
|
|
"num_tokens": 4317220.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 6.963476037979126,
|
|
"epoch": 0.22042285200179937,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004998358961068062,
|
|
"loss": 6.8363,
|
|
"mean_token_accuracy": 0.12407503426074981,
|
|
"num_tokens": 4326496.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 6.850760078430175,
|
|
"epoch": 0.22092267706302793,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004998345304549661,
|
|
"loss": 6.8271,
|
|
"mean_token_accuracy": 0.12148844972252845,
|
|
"num_tokens": 4335705.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 6.9391758918762205,
|
|
"epoch": 0.22142250212425652,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004998331591463471,
|
|
"loss": 6.8674,
|
|
"mean_token_accuracy": 0.12010736986994744,
|
|
"num_tokens": 4344711.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 6.898680257797241,
|
|
"epoch": 0.22192232718548507,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004998317821809837,
|
|
"loss": 6.882,
|
|
"mean_token_accuracy": 0.11965162381529808,
|
|
"num_tokens": 4355448.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 6.817663192749023,
|
|
"epoch": 0.22242215224671366,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004998303995589104,
|
|
"loss": 6.7461,
|
|
"mean_token_accuracy": 0.12404740154743195,
|
|
"num_tokens": 4365104.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 6.794039249420166,
|
|
"epoch": 0.22292197730794222,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004998290112801623,
|
|
"loss": 6.7307,
|
|
"mean_token_accuracy": 0.12114329189062119,
|
|
"num_tokens": 4375219.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 6.937287187576294,
|
|
"epoch": 0.22342180236917078,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000499827617344774,
|
|
"loss": 6.8694,
|
|
"mean_token_accuracy": 0.11553702056407929,
|
|
"num_tokens": 4385294.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 6.932056283950805,
|
|
"epoch": 0.22392162743039937,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004998262177527807,
|
|
"loss": 6.8277,
|
|
"mean_token_accuracy": 0.1127886414527893,
|
|
"num_tokens": 4394305.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 6.870424127578735,
|
|
"epoch": 0.22442145249162793,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004998248125042177,
|
|
"loss": 6.6962,
|
|
"mean_token_accuracy": 0.12111802026629448,
|
|
"num_tokens": 4403732.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 6.840139436721802,
|
|
"epoch": 0.2249212775528565,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004998234015991202,
|
|
"loss": 6.9208,
|
|
"mean_token_accuracy": 0.11533247455954551,
|
|
"num_tokens": 4413730.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 7.017762231826782,
|
|
"epoch": 0.22542110261408507,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004998219850375238,
|
|
"loss": 6.9207,
|
|
"mean_token_accuracy": 0.11020428538322449,
|
|
"num_tokens": 4423051.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 6.969862747192383,
|
|
"epoch": 0.22592092767531363,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004998205628194641,
|
|
"loss": 6.9154,
|
|
"mean_token_accuracy": 0.11847143098711968,
|
|
"num_tokens": 4433363.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 6.95698299407959,
|
|
"epoch": 0.22642075273654222,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004998191349449768,
|
|
"loss": 6.8742,
|
|
"mean_token_accuracy": 0.11326471045613289,
|
|
"num_tokens": 4442984.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 6.874368715286255,
|
|
"epoch": 0.22692057779777078,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004998177014140981,
|
|
"loss": 6.8188,
|
|
"mean_token_accuracy": 0.115499959141016,
|
|
"num_tokens": 4452364.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 6.949122142791748,
|
|
"epoch": 0.22742040285899934,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004998162622268637,
|
|
"loss": 6.8461,
|
|
"mean_token_accuracy": 0.11230619102716446,
|
|
"num_tokens": 4461904.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 6.843392372131348,
|
|
"epoch": 0.22792022792022792,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004998148173833101,
|
|
"loss": 6.7891,
|
|
"mean_token_accuracy": 0.11898191049695014,
|
|
"num_tokens": 4470633.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 6.851195049285889,
|
|
"epoch": 0.22842005298145648,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004998133668834735,
|
|
"loss": 6.7984,
|
|
"mean_token_accuracy": 0.12105753719806671,
|
|
"num_tokens": 4480673.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 6.894360876083374,
|
|
"epoch": 0.22891987804268507,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0004998119107273905,
|
|
"loss": 6.7919,
|
|
"mean_token_accuracy": 0.11570360809564591,
|
|
"num_tokens": 4489923.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 6.800668478012085,
|
|
"epoch": 0.22941970310391363,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004998104489150975,
|
|
"loss": 6.7628,
|
|
"mean_token_accuracy": 0.11922774538397789,
|
|
"num_tokens": 4501007.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 6.874569845199585,
|
|
"epoch": 0.2299195281651422,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004998089814466318,
|
|
"loss": 6.7858,
|
|
"mean_token_accuracy": 0.11911295130848884,
|
|
"num_tokens": 4510460.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 6.9255266189575195,
|
|
"epoch": 0.23041935322637078,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004998075083220295,
|
|
"loss": 6.8655,
|
|
"mean_token_accuracy": 0.11979968324303628,
|
|
"num_tokens": 4521026.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 6.814203453063965,
|
|
"epoch": 0.23091917828759934,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004998060295413286,
|
|
"loss": 6.786,
|
|
"mean_token_accuracy": 0.12223186492919921,
|
|
"num_tokens": 4530818.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 6.782166814804077,
|
|
"epoch": 0.23141900334882792,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004998045451045656,
|
|
"loss": 6.7239,
|
|
"mean_token_accuracy": 0.1288319930434227,
|
|
"num_tokens": 4539512.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 6.962126302719116,
|
|
"epoch": 0.23191882841005648,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004998030550117782,
|
|
"loss": 6.9126,
|
|
"mean_token_accuracy": 0.10850706845521926,
|
|
"num_tokens": 4549030.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 6.811104106903076,
|
|
"epoch": 0.23241865347128504,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004998015592630037,
|
|
"loss": 6.7625,
|
|
"mean_token_accuracy": 0.12445050030946732,
|
|
"num_tokens": 4559347.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 6.906196594238281,
|
|
"epoch": 0.23291847853251363,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0004998000578582798,
|
|
"loss": 6.7899,
|
|
"mean_token_accuracy": 0.11643932834267616,
|
|
"num_tokens": 4568721.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 6.858196830749511,
|
|
"epoch": 0.2334183035937422,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004997985507976443,
|
|
"loss": 6.8111,
|
|
"mean_token_accuracy": 0.12294815853238106,
|
|
"num_tokens": 4578265.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 6.801019620895386,
|
|
"epoch": 0.23391812865497075,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004997970380811351,
|
|
"loss": 6.787,
|
|
"mean_token_accuracy": 0.12455012276768684,
|
|
"num_tokens": 4588694.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 6.872055768966675,
|
|
"epoch": 0.23441795371619933,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004997955197087904,
|
|
"loss": 6.7748,
|
|
"mean_token_accuracy": 0.1263163171708584,
|
|
"num_tokens": 4597754.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 6.826019239425659,
|
|
"epoch": 0.2349177787774279,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004997939956806481,
|
|
"loss": 6.7868,
|
|
"mean_token_accuracy": 0.12200037539005279,
|
|
"num_tokens": 4606874.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 6.829957675933838,
|
|
"epoch": 0.23541760383865648,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004997924659967467,
|
|
"loss": 6.6847,
|
|
"mean_token_accuracy": 0.12210214510560036,
|
|
"num_tokens": 4615588.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 6.771548700332642,
|
|
"epoch": 0.23591742889988504,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004997909306571248,
|
|
"loss": 6.8354,
|
|
"mean_token_accuracy": 0.12239153161644936,
|
|
"num_tokens": 4625919.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 6.874808311462402,
|
|
"epoch": 0.2364172539611136,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004997893896618209,
|
|
"loss": 6.7683,
|
|
"mean_token_accuracy": 0.11741094887256623,
|
|
"num_tokens": 4635539.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 6.882049226760865,
|
|
"epoch": 0.2369170790223422,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004997878430108738,
|
|
"loss": 6.7736,
|
|
"mean_token_accuracy": 0.1185309499502182,
|
|
"num_tokens": 4645833.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 6.7860688209533695,
|
|
"epoch": 0.23741690408357075,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0004997862907043224,
|
|
"loss": 6.8908,
|
|
"mean_token_accuracy": 0.12114740163087845,
|
|
"num_tokens": 4655166.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 7.0036725997924805,
|
|
"epoch": 0.23791672914479933,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004997847327422059,
|
|
"loss": 6.7559,
|
|
"mean_token_accuracy": 0.12411316707730294,
|
|
"num_tokens": 4664884.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 6.804525756835938,
|
|
"epoch": 0.2384165542060279,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004997831691245632,
|
|
"loss": 6.8501,
|
|
"mean_token_accuracy": 0.11812512278556823,
|
|
"num_tokens": 4675172.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 6.920410919189453,
|
|
"epoch": 0.23891637926725645,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004997815998514337,
|
|
"loss": 6.7558,
|
|
"mean_token_accuracy": 0.12874482199549675,
|
|
"num_tokens": 4684588.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 6.8410602569580075,
|
|
"epoch": 0.23941620432848504,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004997800249228572,
|
|
"loss": 6.8633,
|
|
"mean_token_accuracy": 0.12056856080889702,
|
|
"num_tokens": 4694434.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 6.833298397064209,
|
|
"epoch": 0.2399160293897136,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004997784443388732,
|
|
"loss": 6.8069,
|
|
"mean_token_accuracy": 0.12126857340335846,
|
|
"num_tokens": 4705080.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 6.929813718795776,
|
|
"epoch": 0.24041585445094216,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0004997768580995212,
|
|
"loss": 6.8913,
|
|
"mean_token_accuracy": 0.11827507093548775,
|
|
"num_tokens": 4714466.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 6.816767978668213,
|
|
"epoch": 0.24091567951217074,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004997752662048413,
|
|
"loss": 6.7902,
|
|
"mean_token_accuracy": 0.12225700989365577,
|
|
"num_tokens": 4724047.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 6.883993434906006,
|
|
"epoch": 0.2414155045733993,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004997736686548736,
|
|
"loss": 6.8239,
|
|
"mean_token_accuracy": 0.11739473715424538,
|
|
"num_tokens": 4733153.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 6.902164316177368,
|
|
"epoch": 0.2419153296346279,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004997720654496582,
|
|
"loss": 6.6874,
|
|
"mean_token_accuracy": 0.1274111844599247,
|
|
"num_tokens": 4743239.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 6.779417943954468,
|
|
"epoch": 0.24241515469585645,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004997704565892355,
|
|
"loss": 6.7309,
|
|
"mean_token_accuracy": 0.11772564947605133,
|
|
"num_tokens": 4753128.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 6.967568254470825,
|
|
"epoch": 0.242914979757085,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004997688420736459,
|
|
"loss": 6.9923,
|
|
"mean_token_accuracy": 0.1105314128100872,
|
|
"num_tokens": 4764056.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 6.921211242675781,
|
|
"epoch": 0.2434148048183136,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004997672219029301,
|
|
"loss": 6.803,
|
|
"mean_token_accuracy": 0.11966260820627213,
|
|
"num_tokens": 4773354.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 6.768605518341064,
|
|
"epoch": 0.24391462987954216,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004997655960771288,
|
|
"loss": 6.7419,
|
|
"mean_token_accuracy": 0.12112644538283349,
|
|
"num_tokens": 4783367.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 6.760331010818481,
|
|
"epoch": 0.24441445494077074,
|
|
"grad_norm": 1.8359375,
|
|
"learning_rate": 0.000499763964596283,
|
|
"loss": 6.6365,
|
|
"mean_token_accuracy": 0.13002105727791785,
|
|
"num_tokens": 4793028.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 6.887047004699707,
|
|
"epoch": 0.2449142800019993,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004997623274604336,
|
|
"loss": 6.8154,
|
|
"mean_token_accuracy": 0.11559386104345322,
|
|
"num_tokens": 4804032.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 6.851502513885498,
|
|
"epoch": 0.24541410506322786,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000499760684669622,
|
|
"loss": 6.8398,
|
|
"mean_token_accuracy": 0.12432057932019233,
|
|
"num_tokens": 4813294.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 6.841673564910889,
|
|
"epoch": 0.24591393012445645,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004997590362238894,
|
|
"loss": 6.7551,
|
|
"mean_token_accuracy": 0.12083819210529327,
|
|
"num_tokens": 4823461.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 6.800539541244507,
|
|
"epoch": 0.246413755185685,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004997573821232772,
|
|
"loss": 6.7896,
|
|
"mean_token_accuracy": 0.12408755049109459,
|
|
"num_tokens": 4832464.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 6.831569385528565,
|
|
"epoch": 0.24691358024691357,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004997557223678272,
|
|
"loss": 6.7568,
|
|
"mean_token_accuracy": 0.12164122313261032,
|
|
"num_tokens": 4842697.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 6.906039190292359,
|
|
"epoch": 0.24741340530814215,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004997540569575809,
|
|
"loss": 6.8111,
|
|
"mean_token_accuracy": 0.12153081223368645,
|
|
"num_tokens": 4853129.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 6.879016637802124,
|
|
"epoch": 0.2479132303693707,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004997523858925805,
|
|
"loss": 6.8479,
|
|
"mean_token_accuracy": 0.11290789991617203,
|
|
"num_tokens": 4864048.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 6.748233461380005,
|
|
"epoch": 0.2484130554305993,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004997507091728678,
|
|
"loss": 6.7881,
|
|
"mean_token_accuracy": 0.12026874572038651,
|
|
"num_tokens": 4874368.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 6.947058010101318,
|
|
"epoch": 0.24891288049182786,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004997490267984853,
|
|
"loss": 6.7615,
|
|
"mean_token_accuracy": 0.12401909828186035,
|
|
"num_tokens": 4884813.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 6.790718698501587,
|
|
"epoch": 0.24941270555305642,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004997473387694749,
|
|
"loss": 6.8472,
|
|
"mean_token_accuracy": 0.12006180807948112,
|
|
"num_tokens": 4895594.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 6.864915561676026,
|
|
"epoch": 0.249912530614285,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004997456450858793,
|
|
"loss": 6.7273,
|
|
"mean_token_accuracy": 0.12555136010050774,
|
|
"num_tokens": 4904922.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 6.8340576171875,
|
|
"epoch": 0.25041235567551356,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004997439457477412,
|
|
"loss": 6.7292,
|
|
"mean_token_accuracy": 0.11903331875801086,
|
|
"num_tokens": 4915009.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 6.849345684051514,
|
|
"epoch": 0.25091218073674215,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004997422407551032,
|
|
"loss": 6.8718,
|
|
"mean_token_accuracy": 0.1152284488081932,
|
|
"num_tokens": 4924916.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 6.824352502822876,
|
|
"epoch": 0.2514120057979707,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004997405301080082,
|
|
"loss": 6.7613,
|
|
"mean_token_accuracy": 0.11944366693496704,
|
|
"num_tokens": 4935127.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 6.818592929840088,
|
|
"epoch": 0.25191183085919927,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004997388138064994,
|
|
"loss": 6.7721,
|
|
"mean_token_accuracy": 0.12516257539391518,
|
|
"num_tokens": 4944749.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 6.791562271118164,
|
|
"epoch": 0.25241165592042786,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004997370918506198,
|
|
"loss": 6.6298,
|
|
"mean_token_accuracy": 0.1356276325881481,
|
|
"num_tokens": 4953868.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 6.78893814086914,
|
|
"epoch": 0.25291148098165644,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004997353642404129,
|
|
"loss": 6.7301,
|
|
"mean_token_accuracy": 0.12474445253610611,
|
|
"num_tokens": 4963282.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 6.679150104522705,
|
|
"epoch": 0.253411306042885,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004997336309759219,
|
|
"loss": 6.593,
|
|
"mean_token_accuracy": 0.13212000578641891,
|
|
"num_tokens": 4973099.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 6.789412307739258,
|
|
"epoch": 0.25391113110411356,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004997318920571907,
|
|
"loss": 6.7625,
|
|
"mean_token_accuracy": 0.12050073370337486,
|
|
"num_tokens": 4983771.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 6.7664862155914305,
|
|
"epoch": 0.25441095616534215,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000499730147484263,
|
|
"loss": 6.6511,
|
|
"mean_token_accuracy": 0.12306877970695496,
|
|
"num_tokens": 4993648.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 6.757249116897583,
|
|
"epoch": 0.2549107812265707,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004997283972571824,
|
|
"loss": 6.7678,
|
|
"mean_token_accuracy": 0.12360350862145424,
|
|
"num_tokens": 5002821.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 6.85500750541687,
|
|
"epoch": 0.25541060628779927,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004997266413759934,
|
|
"loss": 6.7784,
|
|
"mean_token_accuracy": 0.11761773452162742,
|
|
"num_tokens": 5012655.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 6.707156181335449,
|
|
"epoch": 0.25591043134902786,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004997248798407397,
|
|
"loss": 6.7499,
|
|
"mean_token_accuracy": 0.1209179200232029,
|
|
"num_tokens": 5023649.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 6.851459789276123,
|
|
"epoch": 0.2564102564102564,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004997231126514659,
|
|
"loss": 6.7017,
|
|
"mean_token_accuracy": 0.12884948179125785,
|
|
"num_tokens": 5031945.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 6.8375133037567135,
|
|
"epoch": 0.256910081471485,
|
|
"grad_norm": 1.796875,
|
|
"learning_rate": 0.0004997213398082165,
|
|
"loss": 6.7484,
|
|
"mean_token_accuracy": 0.12101335003972054,
|
|
"num_tokens": 5042036.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 6.778736925125122,
|
|
"epoch": 0.25740990653271356,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000499719561311036,
|
|
"loss": 6.7588,
|
|
"mean_token_accuracy": 0.12266019508242607,
|
|
"num_tokens": 5050986.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 6.868758630752564,
|
|
"epoch": 0.2579097315939421,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004997177771599691,
|
|
"loss": 6.7456,
|
|
"mean_token_accuracy": 0.12436506897211075,
|
|
"num_tokens": 5061408.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 6.7382057189941404,
|
|
"epoch": 0.2584095566551707,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004997159873550607,
|
|
"loss": 6.6532,
|
|
"mean_token_accuracy": 0.12418296411633492,
|
|
"num_tokens": 5071646.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 6.6940131187438965,
|
|
"epoch": 0.25890938171639927,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004997141918963561,
|
|
"loss": 6.7626,
|
|
"mean_token_accuracy": 0.12861955985426904,
|
|
"num_tokens": 5083185.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 6.887934303283691,
|
|
"epoch": 0.25940920677762785,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004997123907839001,
|
|
"loss": 6.7215,
|
|
"mean_token_accuracy": 0.12268028780817986,
|
|
"num_tokens": 5092993.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 6.8641517639160154,
|
|
"epoch": 0.2599090318388564,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004997105840177381,
|
|
"loss": 6.7823,
|
|
"mean_token_accuracy": 0.11887935623526573,
|
|
"num_tokens": 5102709.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 6.770284795761109,
|
|
"epoch": 0.260408856900085,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004997087715979157,
|
|
"loss": 6.7178,
|
|
"mean_token_accuracy": 0.1205891989171505,
|
|
"num_tokens": 5112858.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 6.862639713287353,
|
|
"epoch": 0.26090868196131356,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004997069535244785,
|
|
"loss": 6.6414,
|
|
"mean_token_accuracy": 0.12935668230056763,
|
|
"num_tokens": 5123573.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 6.701737117767334,
|
|
"epoch": 0.2614085070225421,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004997051297974722,
|
|
"loss": 6.8159,
|
|
"mean_token_accuracy": 0.11847961246967316,
|
|
"num_tokens": 5132908.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 6.825256443023681,
|
|
"epoch": 0.2619083320837707,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004997033004169425,
|
|
"loss": 6.6951,
|
|
"mean_token_accuracy": 0.128434656560421,
|
|
"num_tokens": 5142331.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 6.889538669586182,
|
|
"epoch": 0.26240815714499927,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004997014653829357,
|
|
"loss": 6.7775,
|
|
"mean_token_accuracy": 0.11645150184631348,
|
|
"num_tokens": 5152517.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 6.763758563995362,
|
|
"epoch": 0.2629079822062278,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004996996246954977,
|
|
"loss": 6.7249,
|
|
"mean_token_accuracy": 0.1246179036796093,
|
|
"num_tokens": 5162418.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 6.751154136657715,
|
|
"epoch": 0.2634078072674564,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004996977783546752,
|
|
"loss": 6.748,
|
|
"mean_token_accuracy": 0.11966144517064095,
|
|
"num_tokens": 5171948.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 6.8422363758087155,
|
|
"epoch": 0.26390763232868497,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004996959263605143,
|
|
"loss": 6.6607,
|
|
"mean_token_accuracy": 0.126600893586874,
|
|
"num_tokens": 5180963.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 6.765673446655273,
|
|
"epoch": 0.2644074573899135,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004996940687130617,
|
|
"loss": 6.7731,
|
|
"mean_token_accuracy": 0.12072838991880416,
|
|
"num_tokens": 5192305.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 6.7567198276519775,
|
|
"epoch": 0.2649072824511421,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004996922054123641,
|
|
"loss": 6.6534,
|
|
"mean_token_accuracy": 0.12300853133201599,
|
|
"num_tokens": 5201503.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 6.784299898147583,
|
|
"epoch": 0.2654071075123707,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004996903364584685,
|
|
"loss": 6.6341,
|
|
"mean_token_accuracy": 0.1213750496506691,
|
|
"num_tokens": 5210206.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 6.587752819061279,
|
|
"epoch": 0.26590693257359926,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004996884618514219,
|
|
"loss": 6.6488,
|
|
"mean_token_accuracy": 0.12771403416991234,
|
|
"num_tokens": 5220256.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 6.779465198516846,
|
|
"epoch": 0.2664067576348278,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004996865815912713,
|
|
"loss": 6.68,
|
|
"mean_token_accuracy": 0.1303165651857853,
|
|
"num_tokens": 5229128.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 6.789926338195801,
|
|
"epoch": 0.2669065826960564,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004996846956780642,
|
|
"loss": 6.6227,
|
|
"mean_token_accuracy": 0.124240493029356,
|
|
"num_tokens": 5239468.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 6.7334638118743895,
|
|
"epoch": 0.26740640775728497,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000499682804111848,
|
|
"loss": 6.7304,
|
|
"mean_token_accuracy": 0.12104033678770065,
|
|
"num_tokens": 5247957.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 6.8382179737091064,
|
|
"epoch": 0.2679062328185135,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004996809068926702,
|
|
"loss": 6.7457,
|
|
"mean_token_accuracy": 0.11961550340056419,
|
|
"num_tokens": 5257960.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 6.6922729969024655,
|
|
"epoch": 0.2684060578797421,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004996790040205786,
|
|
"loss": 6.601,
|
|
"mean_token_accuracy": 0.12626811265945434,
|
|
"num_tokens": 5266696.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 6.728348159790039,
|
|
"epoch": 0.2689058829409707,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004996770954956211,
|
|
"loss": 6.5716,
|
|
"mean_token_accuracy": 0.13157410100102424,
|
|
"num_tokens": 5275619.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 6.75075330734253,
|
|
"epoch": 0.2694057080021992,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004996751813178456,
|
|
"loss": 6.7428,
|
|
"mean_token_accuracy": 0.12377356737852097,
|
|
"num_tokens": 5284884.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 6.719533443450928,
|
|
"epoch": 0.2699055330634278,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004996732614873005,
|
|
"loss": 6.634,
|
|
"mean_token_accuracy": 0.13089523389935492,
|
|
"num_tokens": 5293909.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 6.679500532150269,
|
|
"epoch": 0.2704053581246564,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004996713360040339,
|
|
"loss": 6.7119,
|
|
"mean_token_accuracy": 0.1190476343035698,
|
|
"num_tokens": 5303142.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 6.799370670318604,
|
|
"epoch": 0.27090518318588497,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004996694048680942,
|
|
"loss": 6.6442,
|
|
"mean_token_accuracy": 0.12190694138407707,
|
|
"num_tokens": 5313055.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 6.73793511390686,
|
|
"epoch": 0.2714050082471135,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004996674680795302,
|
|
"loss": 6.7527,
|
|
"mean_token_accuracy": 0.11705323457717895,
|
|
"num_tokens": 5323773.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 6.625799989700317,
|
|
"epoch": 0.2719048333083421,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004996655256383905,
|
|
"loss": 6.5626,
|
|
"mean_token_accuracy": 0.13518762066960335,
|
|
"num_tokens": 5334613.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 6.747070741653443,
|
|
"epoch": 0.2724046583695707,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000499663577544724,
|
|
"loss": 6.6161,
|
|
"mean_token_accuracy": 0.13200468420982361,
|
|
"num_tokens": 5344151.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 6.723088645935059,
|
|
"epoch": 0.2729044834307992,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004996616237985796,
|
|
"loss": 6.5994,
|
|
"mean_token_accuracy": 0.1294735737144947,
|
|
"num_tokens": 5352425.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 6.737382936477661,
|
|
"epoch": 0.2734043084920278,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004996596644000066,
|
|
"loss": 6.8021,
|
|
"mean_token_accuracy": 0.11943930014967918,
|
|
"num_tokens": 5361811.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 6.840099048614502,
|
|
"epoch": 0.2739041335532564,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004996576993490543,
|
|
"loss": 6.7398,
|
|
"mean_token_accuracy": 0.11546388566493988,
|
|
"num_tokens": 5371463.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 6.748827409744263,
|
|
"epoch": 0.2744039586144849,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004996557286457721,
|
|
"loss": 6.5958,
|
|
"mean_token_accuracy": 0.13211297914385794,
|
|
"num_tokens": 5381464.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 6.657773780822754,
|
|
"epoch": 0.2749037836757135,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004996537522902096,
|
|
"loss": 6.69,
|
|
"mean_token_accuracy": 0.12324602678418159,
|
|
"num_tokens": 5391012.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 6.880642366409302,
|
|
"epoch": 0.2754036087369421,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004996517702824163,
|
|
"loss": 6.7774,
|
|
"mean_token_accuracy": 0.12375206127762794,
|
|
"num_tokens": 5401198.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 6.836090850830078,
|
|
"epoch": 0.2759034337981706,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004996497826224425,
|
|
"loss": 6.7114,
|
|
"mean_token_accuracy": 0.12381514981389045,
|
|
"num_tokens": 5410592.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 6.669005727767944,
|
|
"epoch": 0.2764032588593992,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004996477893103379,
|
|
"loss": 6.6853,
|
|
"mean_token_accuracy": 0.12954680174589156,
|
|
"num_tokens": 5420878.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 6.798803281784058,
|
|
"epoch": 0.2769030839206278,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004996457903461527,
|
|
"loss": 6.7194,
|
|
"mean_token_accuracy": 0.12248257845640183,
|
|
"num_tokens": 5431199.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 6.745610475540161,
|
|
"epoch": 0.2774029089818564,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004996437857299373,
|
|
"loss": 6.6419,
|
|
"mean_token_accuracy": 0.1277368612587452,
|
|
"num_tokens": 5441661.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 6.669736242294311,
|
|
"epoch": 0.2779027340430849,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000499641775461742,
|
|
"loss": 6.6342,
|
|
"mean_token_accuracy": 0.1298973187804222,
|
|
"num_tokens": 5451863.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 6.779951810836792,
|
|
"epoch": 0.2784025591043135,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004996397595416173,
|
|
"loss": 6.6731,
|
|
"mean_token_accuracy": 0.1280485101044178,
|
|
"num_tokens": 5461600.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 6.832998275756836,
|
|
"epoch": 0.2789023841655421,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004996377379696143,
|
|
"loss": 6.7265,
|
|
"mean_token_accuracy": 0.11907677501440048,
|
|
"num_tokens": 5470097.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 6.808766031265259,
|
|
"epoch": 0.2794022092267706,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004996357107457834,
|
|
"loss": 6.844,
|
|
"mean_token_accuracy": 0.11644840613007545,
|
|
"num_tokens": 5480906.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 6.882464647293091,
|
|
"epoch": 0.2799020342879992,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000499633677870176,
|
|
"loss": 6.7426,
|
|
"mean_token_accuracy": 0.12104857787489891,
|
|
"num_tokens": 5489732.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 6.612109041213989,
|
|
"epoch": 0.2804018593492278,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004996316393428429,
|
|
"loss": 6.5461,
|
|
"mean_token_accuracy": 0.1322778083384037,
|
|
"num_tokens": 5498422.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 6.659309577941895,
|
|
"epoch": 0.2809016844104563,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004996295951638357,
|
|
"loss": 6.6973,
|
|
"mean_token_accuracy": 0.12405995801091194,
|
|
"num_tokens": 5507958.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 6.862936639785767,
|
|
"epoch": 0.2814015094716849,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004996275453332057,
|
|
"loss": 6.7103,
|
|
"mean_token_accuracy": 0.12772561535239219,
|
|
"num_tokens": 5516752.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 6.719772815704346,
|
|
"epoch": 0.2819013345329135,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004996254898510043,
|
|
"loss": 6.7309,
|
|
"mean_token_accuracy": 0.12634454071521758,
|
|
"num_tokens": 5527211.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 6.701810216903686,
|
|
"epoch": 0.282401159594142,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004996234287172835,
|
|
"loss": 6.6265,
|
|
"mean_token_accuracy": 0.130839204788208,
|
|
"num_tokens": 5536925.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 6.777781772613525,
|
|
"epoch": 0.2829009846553706,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.000499621361932095,
|
|
"loss": 6.7919,
|
|
"mean_token_accuracy": 0.11580215096473694,
|
|
"num_tokens": 5546800.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 6.7316731929779055,
|
|
"epoch": 0.2834008097165992,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004996192894954909,
|
|
"loss": 6.6298,
|
|
"mean_token_accuracy": 0.12401928007602692,
|
|
"num_tokens": 5557266.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 6.772225618362427,
|
|
"epoch": 0.2839006347778278,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004996172114075232,
|
|
"loss": 6.6175,
|
|
"mean_token_accuracy": 0.12422640845179558,
|
|
"num_tokens": 5566169.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 6.7197812557220455,
|
|
"epoch": 0.2844004598390563,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004996151276682443,
|
|
"loss": 6.702,
|
|
"mean_token_accuracy": 0.1240242712199688,
|
|
"num_tokens": 5575566.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 6.651896619796753,
|
|
"epoch": 0.2849002849002849,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004996130382777065,
|
|
"loss": 6.5968,
|
|
"mean_token_accuracy": 0.12551093101501465,
|
|
"num_tokens": 5584804.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 6.729376792907715,
|
|
"epoch": 0.2854001099615135,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004996109432359625,
|
|
"loss": 6.6672,
|
|
"mean_token_accuracy": 0.12371092364192009,
|
|
"num_tokens": 5594361.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 6.803797245025635,
|
|
"epoch": 0.285899935022742,
|
|
"grad_norm": 2.09375,
|
|
"learning_rate": 0.0004996088425430651,
|
|
"loss": 6.6738,
|
|
"mean_token_accuracy": 0.11989209428429604,
|
|
"num_tokens": 5605682.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 6.728023386001587,
|
|
"epoch": 0.2863997600839706,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004996067361990669,
|
|
"loss": 6.7392,
|
|
"mean_token_accuracy": 0.12079543843865395,
|
|
"num_tokens": 5617251.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 6.742342758178711,
|
|
"epoch": 0.2868995851451992,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004996046242040209,
|
|
"loss": 6.6256,
|
|
"mean_token_accuracy": 0.12656520083546638,
|
|
"num_tokens": 5627140.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 6.648843479156494,
|
|
"epoch": 0.28739941020642773,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004996025065579805,
|
|
"loss": 6.5455,
|
|
"mean_token_accuracy": 0.13074300885200502,
|
|
"num_tokens": 5635486.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 6.760076904296875,
|
|
"epoch": 0.2878992352676563,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004996003832609988,
|
|
"loss": 6.725,
|
|
"mean_token_accuracy": 0.11879278272390366,
|
|
"num_tokens": 5645621.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 6.839390230178833,
|
|
"epoch": 0.2883990603288849,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004995982543131292,
|
|
"loss": 6.7974,
|
|
"mean_token_accuracy": 0.11533181294798851,
|
|
"num_tokens": 5655626.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 6.745677804946899,
|
|
"epoch": 0.28889888539011344,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004995961197144254,
|
|
"loss": 6.7284,
|
|
"mean_token_accuracy": 0.12294788435101509,
|
|
"num_tokens": 5664547.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 6.777517175674438,
|
|
"epoch": 0.289398710451342,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000499593979464941,
|
|
"loss": 6.6933,
|
|
"mean_token_accuracy": 0.12205062508583069,
|
|
"num_tokens": 5674861.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 6.781017160415649,
|
|
"epoch": 0.2898985355125706,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004995918335647298,
|
|
"loss": 6.6468,
|
|
"mean_token_accuracy": 0.11987538114190102,
|
|
"num_tokens": 5684143.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 6.746604013442993,
|
|
"epoch": 0.2903983605737992,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004995896820138459,
|
|
"loss": 6.6189,
|
|
"mean_token_accuracy": 0.12228641510009766,
|
|
"num_tokens": 5693837.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 6.7990189552307125,
|
|
"epoch": 0.29089818563502773,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004995875248123434,
|
|
"loss": 6.7048,
|
|
"mean_token_accuracy": 0.1190277747809887,
|
|
"num_tokens": 5704827.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 6.760340070724487,
|
|
"epoch": 0.2913980106962563,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004995853619602766,
|
|
"loss": 6.7344,
|
|
"mean_token_accuracy": 0.12114278227090836,
|
|
"num_tokens": 5714419.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 6.722689962387085,
|
|
"epoch": 0.2918978357574849,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004995831934576998,
|
|
"loss": 6.655,
|
|
"mean_token_accuracy": 0.12388033345341683,
|
|
"num_tokens": 5724983.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 6.699277067184449,
|
|
"epoch": 0.29239766081871343,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004995810193046677,
|
|
"loss": 6.5855,
|
|
"mean_token_accuracy": 0.1329462192952633,
|
|
"num_tokens": 5734471.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 6.746148777008057,
|
|
"epoch": 0.292897485879942,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000499578839501235,
|
|
"loss": 6.7124,
|
|
"mean_token_accuracy": 0.12724108770489692,
|
|
"num_tokens": 5744801.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 6.716687440872192,
|
|
"epoch": 0.2933973109411706,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004995766540474564,
|
|
"loss": 6.6777,
|
|
"mean_token_accuracy": 0.12074058502912521,
|
|
"num_tokens": 5754267.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 6.670263338088989,
|
|
"epoch": 0.29389713600239914,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.000499574462943387,
|
|
"loss": 6.5904,
|
|
"mean_token_accuracy": 0.1308640405535698,
|
|
"num_tokens": 5764466.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 6.760790157318115,
|
|
"epoch": 0.2943969610636277,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000499572266189082,
|
|
"loss": 6.7179,
|
|
"mean_token_accuracy": 0.12045061811804772,
|
|
"num_tokens": 5775158.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 6.731432962417602,
|
|
"epoch": 0.2948967861248563,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004995700637845964,
|
|
"loss": 6.5224,
|
|
"mean_token_accuracy": 0.1331545315682888,
|
|
"num_tokens": 5783320.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 6.582903528213501,
|
|
"epoch": 0.29539661118608485,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004995678557299859,
|
|
"loss": 6.6728,
|
|
"mean_token_accuracy": 0.1265764907002449,
|
|
"num_tokens": 5792980.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 6.848560953140259,
|
|
"epoch": 0.29589643624731343,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004995656420253058,
|
|
"loss": 6.7094,
|
|
"mean_token_accuracy": 0.12386684119701385,
|
|
"num_tokens": 5802338.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 6.654112339019775,
|
|
"epoch": 0.296396261308542,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000499563422670612,
|
|
"loss": 6.6071,
|
|
"mean_token_accuracy": 0.12872216925024987,
|
|
"num_tokens": 5811388.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 6.689492273330688,
|
|
"epoch": 0.2968960863697706,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004995611976659604,
|
|
"loss": 6.6413,
|
|
"mean_token_accuracy": 0.1257068432867527,
|
|
"num_tokens": 5821108.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 6.742918539047241,
|
|
"epoch": 0.29739591143099914,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004995589670114067,
|
|
"loss": 6.6979,
|
|
"mean_token_accuracy": 0.11968054994940758,
|
|
"num_tokens": 5831389.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 6.61173996925354,
|
|
"epoch": 0.2978957364922277,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004995567307070072,
|
|
"loss": 6.5494,
|
|
"mean_token_accuracy": 0.13193634301424026,
|
|
"num_tokens": 5840392.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 6.747926092147827,
|
|
"epoch": 0.2983955615534563,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004995544887528182,
|
|
"loss": 6.568,
|
|
"mean_token_accuracy": 0.13146094381809234,
|
|
"num_tokens": 5850032.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 6.696498203277588,
|
|
"epoch": 0.29889538661468484,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004995522411488959,
|
|
"loss": 6.6238,
|
|
"mean_token_accuracy": 0.13100717812776566,
|
|
"num_tokens": 5859204.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 6.6779335021972654,
|
|
"epoch": 0.29939521167591343,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004995499878952971,
|
|
"loss": 6.6449,
|
|
"mean_token_accuracy": 0.1274212934076786,
|
|
"num_tokens": 5868582.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 6.717268085479736,
|
|
"epoch": 0.299895036737142,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004995477289920784,
|
|
"loss": 6.5901,
|
|
"mean_token_accuracy": 0.12910295873880387,
|
|
"num_tokens": 5879601.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 0.299895036737142,
|
|
"eval_entropy": 6.621060275687474,
|
|
"eval_loss": 6.689310073852539,
|
|
"eval_mean_token_accuracy": 0.12919487868870266,
|
|
"eval_num_tokens": 5879601.0,
|
|
"eval_runtime": 24.0707,
|
|
"eval_samples_per_second": 1289.742,
|
|
"eval_steps_per_second": 161.233,
|
|
"step": 3000
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 100030,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 10,
|
|
"save_steps": 3000,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 9383935887360000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|