fpadovani/jpn-jpan-100mb-after-ppt-Dp-10mb-ckpt500_seed3407
1
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity
Files
fa35c70621eedba6acd0d3899289858267167880
jpn-jpan-100mb-after-ppt-Dp…/checkpoint-1500/trainer_state.json

3068 lines
82 KiB
JSON
Raw Normal View History

初始化项目，由ModelHub XC社区提供模型 Model: fpadovani/jpn-jpan-100mb-after-ppt-Dp-10mb-ckpt500_seed3407 Source: Original Platform
2026-07-30 03:39:19 +08:00
{
"best_global_step": null,
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 1.7543859649122808,
"eval_steps": 500,
"global_step": 1500,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"entropy": 7.7522684097290036,
"epoch": 0.005847953216374269,
"grad_norm": 1.3359375,
"learning_rate": 2e-06,
"loss": 7.7708,
"mean_token_accuracy": 0.11476039662957191,
"num_tokens": 13999.0,
"step": 5
},
{
"entropy": 7.7173127174377445,
"epoch": 0.011695906432748537,
"grad_norm": 1.3203125,
"learning_rate": 4.5e-06,
"loss": 7.712,
"mean_token_accuracy": 0.1273074522614479,
"num_tokens": 25888.0,
"step": 10
},
{
"entropy": 7.66694827079773,
"epoch": 0.017543859649122806,
"grad_norm": 1.1953125,
"learning_rate": 7e-06,
"loss": 7.7338,
"mean_token_accuracy": 0.1251409351825714,
"num_tokens": 37699.0,
"step": 15
},
{
"entropy": 7.703865337371826,
"epoch": 0.023391812865497075,
"grad_norm": 1.0546875,
"learning_rate": 9.5e-06,
"loss": 7.7576,
"mean_token_accuracy": 0.12076537683606148,
"num_tokens": 49648.0,
"step": 20
},
{
"entropy": 7.764259910583496,
"epoch": 0.029239766081871343,
"grad_norm": 1.0234375,
"learning_rate": 1.2e-05,
"loss": 7.6864,
"mean_token_accuracy": 0.12975538596510888,
"num_tokens": 62255.0,
"step": 25
},
{
"entropy": 7.769811201095581,
"epoch": 0.03508771929824561,
"grad_norm": 0.99609375,
"learning_rate": 1.4500000000000002e-05,
"loss": 7.7319,
"mean_token_accuracy": 0.11793469190597534,
"num_tokens": 76003.0,
"step": 30
},
{
"entropy": 7.772569704055786,
"epoch": 0.04093567251461988,
"grad_norm": 0.9609375,
"learning_rate": 1.7000000000000003e-05,
"loss": 7.7096,
"mean_token_accuracy": 0.12931248769164086,
"num_tokens": 87585.0,
"step": 35
},
{
"entropy": 7.823340511322021,
"epoch": 0.04678362573099415,
"grad_norm": 1.1953125,
"learning_rate": 1.95e-05,
"loss": 7.678,
"mean_token_accuracy": 0.13150709196925164,
"num_tokens": 100404.0,
"step": 40
},
{
"entropy": 7.852498149871826,
"epoch": 0.05263157894736842,
"grad_norm": 0.98046875,
"learning_rate": 2.2e-05,
"loss": 7.6639,
"mean_token_accuracy": 0.12596714943647386,
"num_tokens": 113130.0,
"step": 45
},
{
"entropy": 7.778350687026977,
"epoch": 0.05847953216374269,
"grad_norm": 1.3828125,
"learning_rate": 2.4500000000000003e-05,
"loss": 7.6141,
"mean_token_accuracy": 0.1402079403400421,
"num_tokens": 125430.0,
"step": 50
},
{
"entropy": 7.769484996795654,
"epoch": 0.06432748538011696,
"grad_norm": 1.234375,
"learning_rate": 2.7e-05,
"loss": 7.6257,
"mean_token_accuracy": 0.1291840709745884,
"num_tokens": 137550.0,
"step": 55
},
{
"entropy": 7.694907188415527,
"epoch": 0.07017543859649122,
"grad_norm": 1.125,
"learning_rate": 2.95e-05,
"loss": 7.616,
"mean_token_accuracy": 0.13245319500565528,
"num_tokens": 149530.0,
"step": 60
},
{
"entropy": 7.757121801376343,
"epoch": 0.07602339181286549,
"grad_norm": 1.03125,
"learning_rate": 3.2e-05,
"loss": 7.6541,
"mean_token_accuracy": 0.12211295440793038,
"num_tokens": 161671.0,
"step": 65
},
{
"entropy": 7.8260785102844235,
"epoch": 0.08187134502923976,
"grad_norm": 1.03125,
"learning_rate": 3.4500000000000005e-05,
"loss": 7.6742,
"mean_token_accuracy": 0.12763192057609557,
"num_tokens": 174160.0,
"step": 70
},
{
"entropy": 7.788219547271728,
"epoch": 0.08771929824561403,
"grad_norm": 1.0625,
"learning_rate": 3.7e-05,
"loss": 7.6979,
"mean_token_accuracy": 0.12294255271553993,
"num_tokens": 186969.0,
"step": 75
},
{
"entropy": 7.8217706203460695,
"epoch": 0.0935672514619883,
"grad_norm": 0.9375,
"learning_rate": 3.95e-05,
"loss": 7.7355,
"mean_token_accuracy": 0.12626317963004113,
"num_tokens": 199470.0,
"step": 80
},
{
"entropy": 7.854635381698609,
"epoch": 0.09941520467836257,
"grad_norm": 1.0,
"learning_rate": 4.2000000000000004e-05,
"loss": 7.6322,
"mean_token_accuracy": 0.12727101594209672,
"num_tokens": 211306.0,
"step": 85
},
{
"entropy": 7.728736543655396,
"epoch": 0.10526315789473684,
"grad_norm": 1.0703125,
"learning_rate": 4.45e-05,
"loss": 7.5895,
"mean_token_accuracy": 0.12458127737045288,
"num_tokens": 223323.0,
"step": 90
},
{
"entropy": 7.74114408493042,
"epoch": 0.1111111111111111,
"grad_norm": 1.1484375,
"learning_rate": 4.7000000000000004e-05,
"loss": 7.6427,
"mean_token_accuracy": 0.1298776350915432,
"num_tokens": 235896.0,
"step": 95
},
{
"entropy": 7.772605514526367,
"epoch": 0.11695906432748537,
"grad_norm": 1.015625,
"learning_rate": 4.9500000000000004e-05,
"loss": 7.6593,
"mean_token_accuracy": 0.1269608959555626,
"num_tokens": 247879.0,
"step": 100
},
{
"entropy": 7.805002164840698,
"epoch": 0.12280701754385964,
"grad_norm": 0.96484375,
"learning_rate": 5.2e-05,
"loss": 7.5639,
"mean_token_accuracy": 0.13140838518738746,
"num_tokens": 259714.0,
"step": 105
},
{
"entropy": 7.809066724777222,
"epoch": 0.1286549707602339,
"grad_norm": 0.98046875,
"learning_rate": 5.45e-05,
"loss": 7.6391,
"mean_token_accuracy": 0.13385654166340827,
"num_tokens": 272429.0,
"step": 110
},
{
"entropy": 7.754804039001465,
"epoch": 0.13450292397660818,
"grad_norm": 1.0234375,
"learning_rate": 5.7e-05,
"loss": 7.6567,
"mean_token_accuracy": 0.1242964319884777,
"num_tokens": 285924.0,
"step": 115
},
{
"entropy": 7.7574787616729735,
"epoch": 0.14035087719298245,
"grad_norm": 0.8828125,
"learning_rate": 5.9499999999999996e-05,
"loss": 7.6416,
"mean_token_accuracy": 0.13203905522823334,
"num_tokens": 299453.0,
"step": 120
},
{
"entropy": 7.7849445819854735,
"epoch": 0.14619883040935672,
"grad_norm": 0.94921875,
"learning_rate": 6.2e-05,
"loss": 7.6541,
"mean_token_accuracy": 0.12795712649822236,
"num_tokens": 312295.0,
"step": 125
},
{
"entropy": 7.830564260482788,
"epoch": 0.15204678362573099,
"grad_norm": 1.03125,
"learning_rate": 6.450000000000001e-05,
"loss": 7.5997,
"mean_token_accuracy": 0.12964659035205842,
"num_tokens": 324730.0,
"step": 130
},
{
"entropy": 7.782875347137451,
"epoch": 0.15789473684210525,
"grad_norm": 0.9921875,
"learning_rate": 6.7e-05,
"loss": 7.5392,
"mean_token_accuracy": 0.1364399917423725,
"num_tokens": 336481.0,
"step": 135
},
{
"entropy": 7.790884017944336,
"epoch": 0.16374269005847952,
"grad_norm": 1.0234375,
"learning_rate": 6.950000000000001e-05,
"loss": 7.6003,
"mean_token_accuracy": 0.13024170324206352,
"num_tokens": 348987.0,
"step": 140
},
{
"entropy": 7.8166107654571535,
"epoch": 0.1695906432748538,
"grad_norm": 1.0078125,
"learning_rate": 7.2e-05,
"loss": 7.5865,
"mean_token_accuracy": 0.13023216873407364,
"num_tokens": 360390.0,
"step": 145
},
{
"entropy": 7.732658720016479,
"epoch": 0.17543859649122806,
"grad_norm": 1.0390625,
"learning_rate": 7.45e-05,
"loss": 7.5953,
"mean_token_accuracy": 0.1306314319372177,
"num_tokens": 372456.0,
"step": 150
},
{
"entropy": 7.831650114059448,
"epoch": 0.18128654970760233,
"grad_norm": 1.0703125,
"learning_rate": 7.7e-05,
"loss": 7.6024,
"mean_token_accuracy": 0.1245631642639637,
"num_tokens": 384185.0,
"step": 155
},
{
"entropy": 7.759004211425781,
"epoch": 0.1871345029239766,
"grad_norm": 1.09375,
"learning_rate": 7.950000000000001e-05,
"loss": 7.6135,
"mean_token_accuracy": 0.12875869423151015,
"num_tokens": 395399.0,
"step": 160
},
{
"entropy": 7.860350131988525,
"epoch": 0.19298245614035087,
"grad_norm": 1.0859375,
"learning_rate": 8.2e-05,
"loss": 7.611,
"mean_token_accuracy": 0.13090025410056114,
"num_tokens": 408676.0,
"step": 165
},
{
"entropy": 7.830869817733765,
"epoch": 0.19883040935672514,
"grad_norm": 1.0546875,
"learning_rate": 8.450000000000001e-05,
"loss": 7.6944,
"mean_token_accuracy": 0.1266174718737602,
"num_tokens": 421861.0,
"step": 170
},
{
"entropy": 7.768104171752929,
"epoch": 0.2046783625730994,
"grad_norm": 1.0625,
"learning_rate": 8.7e-05,
"loss": 7.623,
"mean_token_accuracy": 0.12830190733075142,
"num_tokens": 434852.0,
"step": 175
},
{
"entropy": 7.895460224151611,
"epoch": 0.21052631578947367,
"grad_norm": 1.0546875,
"learning_rate": 8.95e-05,
"loss": 7.6433,
"mean_token_accuracy": 0.12308355942368507,
"num_tokens": 447313.0,
"step": 180
},
{
"entropy": 7.782856607437134,
"epoch": 0.21637426900584794,
"grad_norm": 1.0390625,
"learning_rate": 9.2e-05,
"loss": 7.603,
"mean_token_accuracy": 0.13218898475170135,
"num_tokens": 460271.0,
"step": 185
},
{
"entropy": 7.745411825180054,
"epoch": 0.2222222222222222,
"grad_norm": 0.98046875,
"learning_rate": 9.45e-05,
"loss": 7.5913,
"mean_token_accuracy": 0.1339310348033905,
"num_tokens": 473006.0,
"step": 190
},
{
"entropy": 7.795257663726806,
"epoch": 0.22807017543859648,
"grad_norm": 1.09375,
"learning_rate": 9.7e-05,
"loss": 7.5571,
"mean_token_accuracy": 0.12507525384426116,
"num_tokens": 486100.0,
"step": 195
},
{
"entropy": 7.703856325149536,
"epoch": 0.23391812865497075,
"grad_norm": 1.2265625,
"learning_rate": 9.95e-05,
"loss": 7.4979,
"mean_token_accuracy": 0.1399431958794594,
"num_tokens": 499460.0,
"step": 200
},
{
"entropy": 7.747776317596435,
"epoch": 0.23976608187134502,
"grad_norm": 1.109375,
"learning_rate": 0.000102,
"loss": 7.5644,
"mean_token_accuracy": 0.13211438804864883,
"num_tokens": 511795.0,
"step": 205
},
{
"entropy": 7.779720592498779,
"epoch": 0.24561403508771928,
"grad_norm": 1.2265625,
"learning_rate": 0.00010449999999999999,
"loss": 7.6583,
"mean_token_accuracy": 0.13016858398914338,
"num_tokens": 524897.0,
"step": 210
},
{
"entropy": 7.713862800598145,
"epoch": 0.25146198830409355,
"grad_norm": 1.03125,
"learning_rate": 0.000107,
"loss": 7.5462,
"mean_token_accuracy": 0.12787842825055123,
"num_tokens": 538325.0,
"step": 215
},
{
"entropy": 7.847299242019654,
"epoch": 0.2573099415204678,
"grad_norm": 1.109375,
"learning_rate": 0.0001095,
"loss": 7.6379,
"mean_token_accuracy": 0.12373388409614564,
"num_tokens": 550979.0,
"step": 220
},
{
"entropy": 7.7990621566772464,
"epoch": 0.2631578947368421,
"grad_norm": 1.1171875,
"learning_rate": 0.000112,
"loss": 7.6099,
"mean_token_accuracy": 0.12962906658649445,
"num_tokens": 563001.0,
"step": 225
},
{
"entropy": 7.734949684143066,
"epoch": 0.26900584795321636,
"grad_norm": 1.09375,
"learning_rate": 0.0001145,
"loss": 7.508,
"mean_token_accuracy": 0.12917143255472183,
"num_tokens": 573600.0,
"step": 230
},
{
"entropy": 7.779465579986573,
"epoch": 0.27485380116959063,
"grad_norm": 1.1796875,
"learning_rate": 0.00011700000000000001,
"loss": 7.5455,
"mean_token_accuracy": 0.1345866911113262,
"num_tokens": 585287.0,
"step": 235
},
{
"entropy": 7.695637369155884,
"epoch": 0.2807017543859649,
"grad_norm": 1.3359375,
"learning_rate": 0.00011949999999999999,
"loss": 7.5678,
"mean_token_accuracy": 0.12697237282991408,
"num_tokens": 597386.0,
"step": 240
},
{
"entropy": 7.780313444137573,
"epoch": 0.28654970760233917,
"grad_norm": 1.03125,
"learning_rate": 0.000122,
"loss": 7.5723,
"mean_token_accuracy": 0.12808034643530847,
"num_tokens": 609256.0,
"step": 245
},
{
"entropy": 7.76597580909729,
"epoch": 0.29239766081871343,
"grad_norm": 1.15625,
"learning_rate": 0.0001245,
"loss": 7.5533,
"mean_token_accuracy": 0.13153037279844285,
"num_tokens": 621506.0,
"step": 250
},
{
"entropy": 7.672491025924683,
"epoch": 0.2982456140350877,
"grad_norm": 1.25,
"learning_rate": 0.000127,
"loss": 7.4974,
"mean_token_accuracy": 0.13769207671284675,
"num_tokens": 633512.0,
"step": 255
},
{
"entropy": 7.770991230010987,
"epoch": 0.30409356725146197,
"grad_norm": 1.1171875,
"learning_rate": 0.0001295,
"loss": 7.6359,
"mean_token_accuracy": 0.12444958984851837,
"num_tokens": 647098.0,
"step": 260
},
{
"entropy": 7.846727418899536,
"epoch": 0.30994152046783624,
"grad_norm": 1.0625,
"learning_rate": 0.000132,
"loss": 7.6429,
"mean_token_accuracy": 0.12304245010018348,
"num_tokens": 659559.0,
"step": 265
},
{
"entropy": 7.775905132293701,
"epoch": 0.3157894736842105,
"grad_norm": 1.0859375,
"learning_rate": 0.00013450000000000002,
"loss": 7.5725,
"mean_token_accuracy": 0.12677664533257485,
"num_tokens": 672621.0,
"step": 270
},
{
"entropy": 7.75470609664917,
"epoch": 0.3216374269005848,
"grad_norm": 1.0703125,
"learning_rate": 0.00013700000000000002,
"loss": 7.5534,
"mean_token_accuracy": 0.13169768303632737,
"num_tokens": 685652.0,
"step": 275
},
{
"entropy": 7.797368335723877,
"epoch": 0.32748538011695905,
"grad_norm": 1.109375,
"learning_rate": 0.0001395,
"loss": 7.6286,
"mean_token_accuracy": 0.1279862903058529,
"num_tokens": 696910.0,
"step": 280
},
{
"entropy": 7.671004962921143,
"epoch": 0.3333333333333333,
"grad_norm": 1.7265625,
"learning_rate": 0.00014199999999999998,
"loss": 7.4883,
"mean_token_accuracy": 0.13524112179875375,
"num_tokens": 709054.0,
"step": 285
},
{
"entropy": 7.824522686004639,
"epoch": 0.3391812865497076,
"grad_norm": 1.046875,
"learning_rate": 0.0001445,
"loss": 7.4892,
"mean_token_accuracy": 0.13072189688682556,
"num_tokens": 720902.0,
"step": 290
},
{
"entropy": 7.761558198928833,
"epoch": 0.34502923976608185,
"grad_norm": 1.203125,
"learning_rate": 0.000147,
"loss": 7.5921,
"mean_token_accuracy": 0.1260167084634304,
"num_tokens": 733319.0,
"step": 295
},
{
"entropy": 7.650327444076538,
"epoch": 0.3508771929824561,
"grad_norm": 1.0703125,
"learning_rate": 0.0001495,
"loss": 7.4744,
"mean_token_accuracy": 0.13074510917067528,
"num_tokens": 744817.0,
"step": 300
},
{
"entropy": 7.753329420089722,
"epoch": 0.3567251461988304,
"grad_norm": 1.0859375,
"learning_rate": 0.000152,
"loss": 7.5266,
"mean_token_accuracy": 0.125571209192276,
"num_tokens": 756582.0,
"step": 305
},
{
"entropy": 7.731254577636719,
"epoch": 0.36257309941520466,
"grad_norm": 1.125,
"learning_rate": 0.00015450000000000001,
"loss": 7.4451,
"mean_token_accuracy": 0.13850152045488356,
"num_tokens": 769273.0,
"step": 310
},
{
"entropy": 7.6755951881408695,
"epoch": 0.3684210526315789,
"grad_norm": 1.328125,
"learning_rate": 0.000157,
"loss": 7.5185,
"mean_token_accuracy": 0.13168835863471032,
"num_tokens": 782001.0,
"step": 315
},
{
"entropy": 7.720033788681031,
"epoch": 0.3742690058479532,
"grad_norm": 1.1171875,
"learning_rate": 0.0001595,
"loss": 7.5397,
"mean_token_accuracy": 0.13381237834692,
"num_tokens": 794575.0,
"step": 320
},
{
"entropy": 7.720466375350952,
"epoch": 0.38011695906432746,
"grad_norm": 1.0625,
"learning_rate": 0.000162,
"loss": 7.5329,
"mean_token_accuracy": 0.1308806821703911,
"num_tokens": 805583.0,
"step": 325
},
{
"entropy": 7.728007459640503,
"epoch": 0.38596491228070173,
"grad_norm": 1.171875,
"learning_rate": 0.00016450000000000001,
"loss": 7.5653,
"mean_token_accuracy": 0.1306055575609207,
"num_tokens": 817617.0,
"step": 330
},
{
"entropy": 7.779255723953247,
"epoch": 0.391812865497076,
"grad_norm": 1.125,
"learning_rate": 0.00016700000000000002,
"loss": 7.482,
"mean_token_accuracy": 0.13038639426231385,
"num_tokens": 830043.0,
"step": 335
},
{
"entropy": 7.62467098236084,
"epoch": 0.39766081871345027,
"grad_norm": 1.09375,
"learning_rate": 0.00016950000000000003,
"loss": 7.4178,
"mean_token_accuracy": 0.13518237322568893,
"num_tokens": 843200.0,
"step": 340
},
{
"entropy": 7.600871419906616,
"epoch": 0.40350877192982454,
"grad_norm": 1.3515625,
"learning_rate": 0.00017199999999999998,
"loss": 7.4983,
"mean_token_accuracy": 0.13517106473445892,
"num_tokens": 856031.0,
"step": 345
},
{
"entropy": 7.789513301849365,
"epoch": 0.4093567251461988,
"grad_norm": 1.59375,
"learning_rate": 0.00017449999999999999,
"loss": 7.491,
"mean_token_accuracy": 0.13276956006884574,
"num_tokens": 869119.0,
"step": 350
},
{
"entropy": 7.601604795455932,
"epoch": 0.4152046783625731,
"grad_norm": 1.125,
"learning_rate": 0.000177,
"loss": 7.4994,
"mean_token_accuracy": 0.13779703229665757,
"num_tokens": 880802.0,
"step": 355
},
{
"entropy": 7.7842954158782955,
"epoch": 0.42105263157894735,
"grad_norm": 1.1015625,
"learning_rate": 0.0001795,
"loss": 7.5207,
"mean_token_accuracy": 0.13423604667186737,
"num_tokens": 894237.0,
"step": 360
},
{
"entropy": 7.737492561340332,
"epoch": 0.4269005847953216,
"grad_norm": 1.1796875,
"learning_rate": 0.000182,
"loss": 7.5183,
"mean_token_accuracy": 0.13320327475667,
"num_tokens": 906695.0,
"step": 365
},
{
"entropy": 7.619091510772705,
"epoch": 0.4327485380116959,
"grad_norm": 1.15625,
"learning_rate": 0.0001845,
"loss": 7.49,
"mean_token_accuracy": 0.13559429571032525,
"num_tokens": 918955.0,
"step": 370
},
{
"entropy": 7.669496536254883,
"epoch": 0.43859649122807015,
"grad_norm": 1.0390625,
"learning_rate": 0.000187,
"loss": 7.544,
"mean_token_accuracy": 0.1327928401529789,
"num_tokens": 931812.0,
"step": 375
},
{
"entropy": 7.779175567626953,
"epoch": 0.4444444444444444,
"grad_norm": 1.0703125,
"learning_rate": 0.0001895,
"loss": 7.5409,
"mean_token_accuracy": 0.12402037903666496,
"num_tokens": 945261.0,
"step": 380
},
{
"entropy": 7.642067909240723,
"epoch": 0.4502923976608187,
"grad_norm": 1.296875,
"learning_rate": 0.000192,
"loss": 7.4515,
"mean_token_accuracy": 0.13938093855977057,
"num_tokens": 958376.0,
"step": 385
},
{
"entropy": 7.622897481918335,
"epoch": 0.45614035087719296,
"grad_norm": 1.0625,
"learning_rate": 0.0001945,
"loss": 7.4556,
"mean_token_accuracy": 0.1381414406001568,
"num_tokens": 971709.0,
"step": 390
},
{
"entropy": 7.7873311042785645,
"epoch": 0.4619883040935672,
"grad_norm": 1.234375,
"learning_rate": 0.00019700000000000002,
"loss": 7.451,
"mean_token_accuracy": 0.13401587978005408,
"num_tokens": 983847.0,
"step": 395
},
{
"entropy": 7.600602531433106,
"epoch": 0.4678362573099415,
"grad_norm": 1.078125,
"learning_rate": 0.00019950000000000002,
"loss": 7.5516,
"mean_token_accuracy": 0.1232217237353325,
"num_tokens": 997034.0,
"step": 400
},
{
"entropy": 7.722095966339111,
"epoch": 0.47368421052631576,
"grad_norm": 1.1875,
"learning_rate": 0.000202,
"loss": 7.5863,
"mean_token_accuracy": 0.1275957316160202,
"num_tokens": 1010213.0,
"step": 405
},
{
"entropy": 7.782728147506714,
"epoch": 0.47953216374269003,
"grad_norm": 1.234375,
"learning_rate": 0.00020449999999999998,
"loss": 7.4336,
"mean_token_accuracy": 0.13219931200146676,
"num_tokens": 1021618.0,
"step": 410
},
{
"entropy": 7.5872598648071286,
"epoch": 0.4853801169590643,
"grad_norm": 1.1796875,
"learning_rate": 0.000207,
"loss": 7.4806,
"mean_token_accuracy": 0.1273957036435604,
"num_tokens": 1033809.0,
"step": 415
},
{
"entropy": 7.725764799118042,
"epoch": 0.49122807017543857,
"grad_norm": 1.0859375,
"learning_rate": 0.0002095,
"loss": 7.4478,
"mean_token_accuracy": 0.1341485410928726,
"num_tokens": 1045291.0,
"step": 420
},
{
"entropy": 7.599104166030884,
"epoch": 0.49707602339181284,
"grad_norm": 1.0859375,
"learning_rate": 0.000212,
"loss": 7.4463,
"mean_token_accuracy": 0.13240791261196136,
"num_tokens": 1057516.0,
"step": 425
},
{
"entropy": 7.597379302978515,
"epoch": 0.5029239766081871,
"grad_norm": 1.078125,
"learning_rate": 0.0002145,
"loss": 7.4051,
"mean_token_accuracy": 0.14025759547948838,
"num_tokens": 1070675.0,
"step": 430
},
{
"entropy": 7.742543315887451,
"epoch": 0.5087719298245614,
"grad_norm": 1.09375,
"learning_rate": 0.00021700000000000002,
"loss": 7.4834,
"mean_token_accuracy": 0.13848798274993895,
"num_tokens": 1082965.0,
"step": 435
},
{
"entropy": 7.514439630508423,
"epoch": 0.5146198830409356,
"grad_norm": 1.046875,
"learning_rate": 0.0002195,
"loss": 7.4668,
"mean_token_accuracy": 0.12758100926876068,
"num_tokens": 1095922.0,
"step": 440
},
{
"entropy": 7.691053152084351,
"epoch": 0.52046783625731,
"grad_norm": 1.2421875,
"learning_rate": 0.000222,
"loss": 7.445,
"mean_token_accuracy": 0.13450392335653305,
"num_tokens": 1109002.0,
"step": 445
},
{
"entropy": 7.727744150161743,
"epoch": 0.5263157894736842,
"grad_norm": 1.15625,
"learning_rate": 0.0002245,
"loss": 7.4662,
"mean_token_accuracy": 0.12763761729002,
"num_tokens": 1121199.0,
"step": 450
},
{
"entropy": 7.560290861129761,
"epoch": 0.5321637426900585,
"grad_norm": 1.0625,
"learning_rate": 0.00022700000000000002,
"loss": 7.4169,
"mean_token_accuracy": 0.13524916470050813,
"num_tokens": 1134721.0,
"step": 455
},
{
"entropy": 7.655321359634399,
"epoch": 0.5380116959064327,
"grad_norm": 1.1015625,
"learning_rate": 0.00022950000000000002,
"loss": 7.5093,
"mean_token_accuracy": 0.1298431448638439,
"num_tokens": 1147954.0,
"step": 460
},
{
"entropy": 7.60046911239624,
"epoch": 0.543859649122807,
"grad_norm": 1.0859375,
"learning_rate": 0.00023200000000000003,
"loss": 7.4142,
"mean_token_accuracy": 0.13469598144292833,
"num_tokens": 1161208.0,
"step": 465
},
{
"entropy": 7.722092056274414,
"epoch": 0.5497076023391813,
"grad_norm": 1.09375,
"learning_rate": 0.00023449999999999998,
"loss": 7.5282,
"mean_token_accuracy": 0.12658010125160218,
"num_tokens": 1173789.0,
"step": 470
},
{
"entropy": 7.6619044780731205,
"epoch": 0.5555555555555556,
"grad_norm": 1.0390625,
"learning_rate": 0.000237,
"loss": 7.424,
"mean_token_accuracy": 0.13303308114409446,
"num_tokens": 1185765.0,
"step": 475
},
{
"entropy": 7.570261764526367,
"epoch": 0.5614035087719298,
"grad_norm": 1.0703125,
"learning_rate": 0.0002395,
"loss": 7.451,
"mean_token_accuracy": 0.13050716370344162,
"num_tokens": 1197954.0,
"step": 480
},
{
"entropy": 7.640900421142578,
"epoch": 0.5672514619883041,
"grad_norm": 1.0625,
"learning_rate": 0.000242,
"loss": 7.3757,
"mean_token_accuracy": 0.13130198270082474,
"num_tokens": 1211285.0,
"step": 485
},
{
"entropy": 7.608647298812866,
"epoch": 0.5730994152046783,
"grad_norm": 1.1171875,
"learning_rate": 0.0002445,
"loss": 7.4658,
"mean_token_accuracy": 0.12969516664743425,
"num_tokens": 1223197.0,
"step": 490
},
{
"entropy": 7.631179285049439,
"epoch": 0.5789473684210527,
"grad_norm": 1.0546875,
"learning_rate": 0.000247,
"loss": 7.3974,
"mean_token_accuracy": 0.1334150068461895,
"num_tokens": 1234817.0,
"step": 495
},
{
"entropy": 7.633148241043091,
"epoch": 0.5847953216374269,
"grad_norm": 1.15625,
"learning_rate": 0.0002495,
"loss": 7.4529,
"mean_token_accuracy": 0.1385871522128582,
"num_tokens": 1249305.0,
"step": 500
},
{
"epoch": 0.5847953216374269,
"eval_entropy": 7.589141852736121,
"eval_loss": 7.436342716217041,
"eval_mean_token_accuracy": 0.12382437618707573,
"eval_num_tokens": 1249305.0,
"eval_runtime": 2.3017,
"eval_samples_per_second": 1176.107,
"eval_steps_per_second": 147.285,
"step": 500
},
{
"entropy": 7.552242279052734,
"epoch": 0.5906432748538012,
"grad_norm": 1.1953125,
"learning_rate": 0.000252,
"loss": 7.4344,
"mean_token_accuracy": 0.13413268253207206,
"num_tokens": 1262182.0,
"step": 505
},
{
"entropy": 7.569356870651245,
"epoch": 0.5964912280701754,
"grad_norm": 1.046875,
"learning_rate": 0.0002545,
"loss": 7.4669,
"mean_token_accuracy": 0.130877798050642,
"num_tokens": 1274616.0,
"step": 510
},
{
"entropy": 7.56549391746521,
"epoch": 0.6023391812865497,
"grad_norm": 1.1640625,
"learning_rate": 0.000257,
"loss": 7.4634,
"mean_token_accuracy": 0.12909993678331375,
"num_tokens": 1287306.0,
"step": 515
},
{
"entropy": 7.517426347732544,
"epoch": 0.6081871345029239,
"grad_norm": 1.234375,
"learning_rate": 0.0002595,
"loss": 7.342,
"mean_token_accuracy": 0.13786138594150543,
"num_tokens": 1299472.0,
"step": 520
},
{
"entropy": 7.555473279953003,
"epoch": 0.6140350877192983,
"grad_norm": 1.953125,
"learning_rate": 0.000262,
"loss": 7.283,
"mean_token_accuracy": 0.14004315435886383,
"num_tokens": 1312636.0,
"step": 525
},
{
"entropy": 7.51561393737793,
"epoch": 0.6198830409356725,
"grad_norm": 1.265625,
"learning_rate": 0.00026450000000000003,
"loss": 7.3192,
"mean_token_accuracy": 0.14248516857624055,
"num_tokens": 1324058.0,
"step": 530
},
{
"entropy": 7.568233871459961,
"epoch": 0.6257309941520468,
"grad_norm": 1.0703125,
"learning_rate": 0.00026700000000000004,
"loss": 7.4306,
"mean_token_accuracy": 0.12730365842580796,
"num_tokens": 1337605.0,
"step": 535
},
{
"entropy": 7.645518589019775,
"epoch": 0.631578947368421,
"grad_norm": 1.1796875,
"learning_rate": 0.00026950000000000005,
"loss": 7.4333,
"mean_token_accuracy": 0.1309638313949108,
"num_tokens": 1350658.0,
"step": 540
},
{
"entropy": 7.496008110046387,
"epoch": 0.6374269005847953,
"grad_norm": 1.3359375,
"learning_rate": 0.00027200000000000005,
"loss": 7.3572,
"mean_token_accuracy": 0.13723124265670777,
"num_tokens": 1362091.0,
"step": 545
},
{
"entropy": 7.592266511917114,
"epoch": 0.6432748538011696,
"grad_norm": 1.0078125,
"learning_rate": 0.0002745,
"loss": 7.3788,
"mean_token_accuracy": 0.13762461990118027,
"num_tokens": 1374055.0,
"step": 550
},
{
"entropy": 7.408201408386231,
"epoch": 0.6491228070175439,
"grad_norm": 1.28125,
"learning_rate": 0.000277,
"loss": 7.2616,
"mean_token_accuracy": 0.13666822388768196,
"num_tokens": 1385269.0,
"step": 555
},
{
"entropy": 7.505675172805786,
"epoch": 0.6549707602339181,
"grad_norm": 1.0546875,
"learning_rate": 0.0002795,
"loss": 7.3816,
"mean_token_accuracy": 0.13430810645222663,
"num_tokens": 1396559.0,
"step": 560
},
{
"entropy": 7.50752067565918,
"epoch": 0.6608187134502924,
"grad_norm": 1.71875,
"learning_rate": 0.00028199999999999997,
"loss": 7.4251,
"mean_token_accuracy": 0.13418283313512802,
"num_tokens": 1408838.0,
"step": 565
},
{
"entropy": 7.553399276733399,
"epoch": 0.6666666666666666,
"grad_norm": 1.1484375,
"learning_rate": 0.0002845,
"loss": 7.2892,
"mean_token_accuracy": 0.14057099148631097,
"num_tokens": 1421023.0,
"step": 570
},
{
"entropy": 7.452613353729248,
"epoch": 0.672514619883041,
"grad_norm": 1.1953125,
"learning_rate": 0.000287,
"loss": 7.3615,
"mean_token_accuracy": 0.135744871199131,
"num_tokens": 1433912.0,
"step": 575
},
{
"entropy": 7.519769906997681,
"epoch": 0.6783625730994152,
"grad_norm": 1.15625,
"learning_rate": 0.0002895,
"loss": 7.3484,
"mean_token_accuracy": 0.1338112011551857,
"num_tokens": 1445262.0,
"step": 580
},
{
"entropy": 7.566982316970825,
"epoch": 0.6842105263157895,
"grad_norm": 0.9609375,
"learning_rate": 0.000292,
"loss": 7.3174,
"mean_token_accuracy": 0.13909269124269485,
"num_tokens": 1457927.0,
"step": 585
},
{
"entropy": 7.478474092483521,
"epoch": 0.6900584795321637,
"grad_norm": 1.078125,
"learning_rate": 0.0002945,
"loss": 7.3998,
"mean_token_accuracy": 0.13400934338569642,
"num_tokens": 1470116.0,
"step": 590
},
{
"entropy": 7.556169080734253,
"epoch": 0.695906432748538,
"grad_norm": 1.2734375,
"learning_rate": 0.000297,
"loss": 7.2901,
"mean_token_accuracy": 0.13774340003728866,
"num_tokens": 1482067.0,
"step": 595
},
{
"entropy": 7.410288572311401,
"epoch": 0.7017543859649122,
"grad_norm": 1.3125,
"learning_rate": 0.0002995,
"loss": 7.3664,
"mean_token_accuracy": 0.13011181652545928,
"num_tokens": 1493902.0,
"step": 600
},
{
"entropy": 7.562840366363526,
"epoch": 0.7076023391812866,
"grad_norm": 1.09375,
"learning_rate": 0.000302,
"loss": 7.3262,
"mean_token_accuracy": 0.1308012142777443,
"num_tokens": 1505689.0,
"step": 605
},
{
"entropy": 7.458141326904297,
"epoch": 0.7134502923976608,
"grad_norm": 1.1875,
"learning_rate": 0.0003045,
"loss": 7.2804,
"mean_token_accuracy": 0.14782539159059524,
"num_tokens": 1518002.0,
"step": 610
},
{
"entropy": 7.489599323272705,
"epoch": 0.7192982456140351,
"grad_norm": 1.0625,
"learning_rate": 0.000307,
"loss": 7.3166,
"mean_token_accuracy": 0.13400111868977546,
"num_tokens": 1529307.0,
"step": 615
},
{
"entropy": 7.606138706207275,
"epoch": 0.7251461988304093,
"grad_norm": 1.109375,
"learning_rate": 0.0003095,
"loss": 7.4568,
"mean_token_accuracy": 0.12007877230644226,
"num_tokens": 1542351.0,
"step": 620
},
{
"entropy": 7.29756646156311,
"epoch": 0.7309941520467836,
"grad_norm": 1.2109375,
"learning_rate": 0.000312,
"loss": 7.2716,
"mean_token_accuracy": 0.1381143569946289,
"num_tokens": 1556263.0,
"step": 625
},
{
"entropy": 7.489748096466064,
"epoch": 0.7368421052631579,
"grad_norm": 1.265625,
"learning_rate": 0.0003145,
"loss": 7.3245,
"mean_token_accuracy": 0.136425893753767,
"num_tokens": 1568906.0,
"step": 630
},
{
"entropy": 7.391974353790284,
"epoch": 0.7426900584795322,
"grad_norm": 1.0625,
"learning_rate": 0.000317,
"loss": 7.297,
"mean_token_accuracy": 0.14263810962438583,
"num_tokens": 1580631.0,
"step": 635
},
{
"entropy": 7.490023469924926,
"epoch": 0.7485380116959064,
"grad_norm": 1.2265625,
"learning_rate": 0.0003195,
"loss": 7.289,
"mean_token_accuracy": 0.1338388778269291,
"num_tokens": 1592369.0,
"step": 640
},
{
"entropy": 7.392187595367432,
"epoch": 0.7543859649122807,
"grad_norm": 1.1484375,
"learning_rate": 0.000322,
"loss": 7.339,
"mean_token_accuracy": 0.14198264628648757,
"num_tokens": 1604888.0,
"step": 645
},
{
"entropy": 7.5100304126739506,
"epoch": 0.7602339181286549,
"grad_norm": 1.3203125,
"learning_rate": 0.00032450000000000003,
"loss": 7.2629,
"mean_token_accuracy": 0.14026070460677148,
"num_tokens": 1617128.0,
"step": 650
},
{
"entropy": 7.432464599609375,
"epoch": 0.7660818713450293,
"grad_norm": 1.0390625,
"learning_rate": 0.00032700000000000003,
"loss": 7.3001,
"mean_token_accuracy": 0.13010574504733086,
"num_tokens": 1629350.0,
"step": 655
},
{
"entropy": 7.502846670150757,
"epoch": 0.7719298245614035,
"grad_norm": 1.1953125,
"learning_rate": 0.00032950000000000004,
"loss": 7.3361,
"mean_token_accuracy": 0.12987025827169418,
"num_tokens": 1643082.0,
"step": 660
},
{
"entropy": 7.421878957748413,
"epoch": 0.7777777777777778,
"grad_norm": 1.046875,
"learning_rate": 0.00033200000000000005,
"loss": 7.3107,
"mean_token_accuracy": 0.13674438670277594,
"num_tokens": 1654785.0,
"step": 665
},
{
"entropy": 7.4461047649383545,
"epoch": 0.783625730994152,
"grad_norm": 1.1328125,
"learning_rate": 0.00033450000000000005,
"loss": 7.3228,
"mean_token_accuracy": 0.13429757431149483,
"num_tokens": 1667704.0,
"step": 670
},
{
"entropy": 7.477173137664795,
"epoch": 0.7894736842105263,
"grad_norm": 1.03125,
"learning_rate": 0.000337,
"loss": 7.2898,
"mean_token_accuracy": 0.13417055681347848,
"num_tokens": 1680459.0,
"step": 675
},
{
"entropy": 7.343410968780518,
"epoch": 0.7953216374269005,
"grad_norm": 1.3203125,
"learning_rate": 0.0003395,
"loss": 7.2102,
"mean_token_accuracy": 0.13917673528194427,
"num_tokens": 1692912.0,
"step": 680
},
{
"entropy": 7.442608833312988,
"epoch": 0.8011695906432749,
"grad_norm": 2.59375,
"learning_rate": 0.000342,
"loss": 7.236,
"mean_token_accuracy": 0.1385551854968071,
"num_tokens": 1705329.0,
"step": 685
},
{
"entropy": 7.451218032836914,
"epoch": 0.8070175438596491,
"grad_norm": 1.15625,
"learning_rate": 0.00034449999999999997,
"loss": 7.3228,
"mean_token_accuracy": 0.1291360892355442,
"num_tokens": 1717283.0,
"step": 690
},
{
"entropy": 7.335363721847534,
"epoch": 0.8128654970760234,
"grad_norm": 1.1875,
"learning_rate": 0.000347,
"loss": 7.2772,
"mean_token_accuracy": 0.1356094777584076,
"num_tokens": 1730308.0,
"step": 695
},
{
"entropy": 7.4717034816741945,
"epoch": 0.8187134502923976,
"grad_norm": 1.234375,
"learning_rate": 0.0003495,
"loss": 7.3221,
"mean_token_accuracy": 0.1344013422727585,
"num_tokens": 1742716.0,
"step": 700
},
{
"entropy": 7.367938613891601,
"epoch": 0.8245614035087719,
"grad_norm": 1.1640625,
"learning_rate": 0.000352,
"loss": 7.2753,
"mean_token_accuracy": 0.13793661966919898,
"num_tokens": 1754183.0,
"step": 705
},
{
"entropy": 7.361904954910278,
"epoch": 0.8304093567251462,
"grad_norm": 1.1640625,
"learning_rate": 0.0003545,
"loss": 7.2833,
"mean_token_accuracy": 0.13037458211183547,
"num_tokens": 1766601.0,
"step": 710
},
{
"entropy": 7.394190931320191,
"epoch": 0.8362573099415205,
"grad_norm": 1.7578125,
"learning_rate": 0.000357,
"loss": 7.1833,
"mean_token_accuracy": 0.14882892221212388,
"num_tokens": 1778756.0,
"step": 715
},
{
"entropy": 7.174529027938843,
"epoch": 0.8421052631578947,
"grad_norm": 1.1171875,
"learning_rate": 0.0003595,
"loss": 7.2478,
"mean_token_accuracy": 0.1429412543773651,
"num_tokens": 1791485.0,
"step": 720
},
{
"entropy": 7.4221092700958256,
"epoch": 0.847953216374269,
"grad_norm": 1.1875,
"learning_rate": 0.000362,
"loss": 7.2633,
"mean_token_accuracy": 0.1324923552572727,
"num_tokens": 1803712.0,
"step": 725
},
{
"entropy": 7.2948565006256105,
"epoch": 0.8538011695906432,
"grad_norm": 1.0078125,
"learning_rate": 0.0003645,
"loss": 7.2204,
"mean_token_accuracy": 0.1371877059340477,
"num_tokens": 1818387.0,
"step": 730
},
{
"entropy": 7.388651275634766,
"epoch": 0.8596491228070176,
"grad_norm": 1.1171875,
"learning_rate": 0.000367,
"loss": 7.2283,
"mean_token_accuracy": 0.13937100917100906,
"num_tokens": 1830113.0,
"step": 735
},
{
"entropy": 7.378299808502197,
"epoch": 0.8654970760233918,
"grad_norm": 1.1875,
"learning_rate": 0.0003695,
"loss": 7.3293,
"mean_token_accuracy": 0.13593609109520913,
"num_tokens": 1841789.0,
"step": 740
},
{
"entropy": 7.341324663162231,
"epoch": 0.8713450292397661,
"grad_norm": 1.09375,
"learning_rate": 0.000372,
"loss": 7.2004,
"mean_token_accuracy": 0.13838357105851173,
"num_tokens": 1853669.0,
"step": 745
},
{
"entropy": 7.236590003967285,
"epoch": 0.8771929824561403,
"grad_norm": 1.078125,
"learning_rate": 0.0003745,
"loss": 7.2461,
"mean_token_accuracy": 0.136163379997015,
"num_tokens": 1865322.0,
"step": 750
},
{
"entropy": 7.365720558166504,
"epoch": 0.8830409356725146,
"grad_norm": 1.0546875,
"learning_rate": 0.000377,
"loss": 7.1657,
"mean_token_accuracy": 0.14459074139595032,
"num_tokens": 1879039.0,
"step": 755
},
{
"entropy": 7.403848600387573,
"epoch": 0.8888888888888888,
"grad_norm": 1.0390625,
"learning_rate": 0.0003795,
"loss": 7.2886,
"mean_token_accuracy": 0.13285375908017158,
"num_tokens": 1893078.0,
"step": 760
},
{
"entropy": 7.370996904373169,
"epoch": 0.8947368421052632,
"grad_norm": 1.1171875,
"learning_rate": 0.000382,
"loss": 7.3194,
"mean_token_accuracy": 0.13308921158313752,
"num_tokens": 1907399.0,
"step": 765
},
{
"entropy": 7.321315145492553,
"epoch": 0.9005847953216374,
"grad_norm": 1.0,
"learning_rate": 0.0003845,
"loss": 7.2799,
"mean_token_accuracy": 0.13092010468244553,
"num_tokens": 1919290.0,
"step": 770
},
{
"entropy": 7.46008620262146,
"epoch": 0.9064327485380117,
"grad_norm": 1.140625,
"learning_rate": 0.00038700000000000003,
"loss": 7.2349,
"mean_token_accuracy": 0.1373835727572441,
"num_tokens": 1931658.0,
"step": 775
},
{
"entropy": 7.3157525062561035,
"epoch": 0.9122807017543859,
"grad_norm": 1.1171875,
"learning_rate": 0.00038950000000000003,
"loss": 7.2629,
"mean_token_accuracy": 0.13429330810904502,
"num_tokens": 1943881.0,
"step": 780
},
{
"entropy": 7.276627683639527,
"epoch": 0.9181286549707602,
"grad_norm": 1.1171875,
"learning_rate": 0.00039200000000000004,
"loss": 7.1786,
"mean_token_accuracy": 0.1390835165977478,
"num_tokens": 1956288.0,
"step": 785
},
{
"entropy": 7.266375255584717,
"epoch": 0.9239766081871345,
"grad_norm": 1.1171875,
"learning_rate": 0.00039450000000000005,
"loss": 7.0636,
"mean_token_accuracy": 0.14030921161174775,
"num_tokens": 1967265.0,
"step": 790
},
{
"entropy": 7.203761053085327,
"epoch": 0.9298245614035088,
"grad_norm": 1.1484375,
"learning_rate": 0.00039700000000000005,
"loss": 7.1881,
"mean_token_accuracy": 0.13456991314888,
"num_tokens": 1980039.0,
"step": 795
},
{
"entropy": 7.337719869613648,
"epoch": 0.935672514619883,
"grad_norm": 1.1015625,
"learning_rate": 0.0003995,
"loss": 7.1764,
"mean_token_accuracy": 0.13603565096855164,
"num_tokens": 1992735.0,
"step": 800
},
{
"entropy": 7.296339225769043,
"epoch": 0.9415204678362573,
"grad_norm": 1.15625,
"learning_rate": 0.000402,
"loss": 7.2013,
"mean_token_accuracy": 0.13806802779436111,
"num_tokens": 2004545.0,
"step": 805
},
{
"entropy": 7.24875774383545,
"epoch": 0.9473684210526315,
"grad_norm": 1.09375,
"learning_rate": 0.0004045,
"loss": 7.1875,
"mean_token_accuracy": 0.13867619559168815,
"num_tokens": 2017100.0,
"step": 810
},
{
"entropy": 7.204592752456665,
"epoch": 0.9532163742690059,
"grad_norm": 1.1875,
"learning_rate": 0.00040699999999999997,
"loss": 7.1545,
"mean_token_accuracy": 0.14778759256005286,
"num_tokens": 2029868.0,
"step": 815
},
{
"entropy": 7.328850173950196,
"epoch": 0.9590643274853801,
"grad_norm": 1.15625,
"learning_rate": 0.0004095,
"loss": 7.2096,
"mean_token_accuracy": 0.13469874411821364,
"num_tokens": 2041334.0,
"step": 820
},
{
"entropy": 7.2642162322998045,
"epoch": 0.9649122807017544,
"grad_norm": 1.0,
"learning_rate": 0.000412,
"loss": 7.2774,
"mean_token_accuracy": 0.1368864133954048,
"num_tokens": 2054797.0,
"step": 825
},
{
"entropy": 7.313835954666137,
"epoch": 0.9707602339181286,
"grad_norm": 0.98828125,
"learning_rate": 0.0004145,
"loss": 7.1486,
"mean_token_accuracy": 0.13503469973802568,
"num_tokens": 2068337.0,
"step": 830
},
{
"entropy": 7.304871654510498,
"epoch": 0.9766081871345029,
"grad_norm": 1.078125,
"learning_rate": 0.000417,
"loss": 7.2026,
"mean_token_accuracy": 0.14170103818178176,
"num_tokens": 2079984.0,
"step": 835
},
{
"entropy": 7.234012746810913,
"epoch": 0.9824561403508771,
"grad_norm": 1.1484375,
"learning_rate": 0.0004195,
"loss": 7.192,
"mean_token_accuracy": 0.14052301198244094,
"num_tokens": 2093069.0,
"step": 840
},
{
"entropy": 7.241004896163941,
"epoch": 0.9883040935672515,
"grad_norm": 1.109375,
"learning_rate": 0.000422,
"loss": 7.1643,
"mean_token_accuracy": 0.13957486748695375,
"num_tokens": 2104753.0,
"step": 845
},
{
"entropy": 7.209813642501831,
"epoch": 0.9941520467836257,
"grad_norm": 0.98046875,
"learning_rate": 0.0004245,
"loss": 7.165,
"mean_token_accuracy": 0.13409762531518937,
"num_tokens": 2116576.0,
"step": 850
},
{
"entropy": 7.2191554546356205,
"epoch": 1.0,
"grad_norm": 1.1796875,
"learning_rate": 0.000427,
"loss": 7.1645,
"mean_token_accuracy": 0.1336888462305069,
"num_tokens": 2129484.0,
"step": 855
},
{
"entropy": 7.1861903190612795,
"epoch": 1.0058479532163742,
"grad_norm": 1.1484375,
"learning_rate": 0.0004295,
"loss": 6.9777,
"mean_token_accuracy": 0.14769867211580276,
"num_tokens": 2142000.0,
"step": 860
},
{
"entropy": 7.141885423660279,
"epoch": 1.0116959064327484,
"grad_norm": 1.140625,
"learning_rate": 0.000432,
"loss": 6.9415,
"mean_token_accuracy": 0.15276771709322928,
"num_tokens": 2152965.0,
"step": 865
},
{
"entropy": 7.1822991371154785,
"epoch": 1.0175438596491229,
"grad_norm": 1.0078125,
"learning_rate": 0.0004345,
"loss": 7.0209,
"mean_token_accuracy": 0.136310812830925,
"num_tokens": 2165078.0,
"step": 870
},
{
"entropy": 7.08724045753479,
"epoch": 1.023391812865497,
"grad_norm": 1.1484375,
"learning_rate": 0.000437,
"loss": 6.9511,
"mean_token_accuracy": 0.14670970737934114,
"num_tokens": 2176706.0,
"step": 875
},
{
"entropy": 7.297170639038086,
"epoch": 1.0292397660818713,
"grad_norm": 1.1171875,
"learning_rate": 0.0004395,
"loss": 6.9646,
"mean_token_accuracy": 0.1434108167886734,
"num_tokens": 2190157.0,
"step": 880
},
{
"entropy": 7.129994964599609,
"epoch": 1.0350877192982457,
"grad_norm": 1.0625,
"learning_rate": 0.000442,
"loss": 6.9484,
"mean_token_accuracy": 0.1423846922814846,
"num_tokens": 2201821.0,
"step": 885
},
{
"entropy": 7.031896543502808,
"epoch": 1.04093567251462,
"grad_norm": 1.0625,
"learning_rate": 0.0004445,
"loss": 6.8791,
"mean_token_accuracy": 0.14986850768327714,
"num_tokens": 2214504.0,
"step": 890
},
{
"entropy": 7.176162052154541,
"epoch": 1.0467836257309941,
"grad_norm": 1.015625,
"learning_rate": 0.000447,
"loss": 7.0018,
"mean_token_accuracy": 0.14039593786001206,
"num_tokens": 2226318.0,
"step": 895
},
{
"entropy": 7.179743337631225,
"epoch": 1.0526315789473684,
"grad_norm": 1.109375,
"learning_rate": 0.00044950000000000003,
"loss": 7.0454,
"mean_token_accuracy": 0.13472965955734253,
"num_tokens": 2237892.0,
"step": 900
},
{
"entropy": 7.097790002822876,
"epoch": 1.0584795321637426,
"grad_norm": 1.2109375,
"learning_rate": 0.00045200000000000004,
"loss": 6.9788,
"mean_token_accuracy": 0.14418615847826005,
"num_tokens": 2250656.0,
"step": 905
},
{
"entropy": 7.133308935165405,
"epoch": 1.064327485380117,
"grad_norm": 0.9921875,
"learning_rate": 0.00045450000000000004,
"loss": 7.0098,
"mean_token_accuracy": 0.14070403948426247,
"num_tokens": 2263685.0,
"step": 910
},
{
"entropy": 7.104922723770142,
"epoch": 1.0701754385964912,
"grad_norm": 1.015625,
"learning_rate": 0.00045700000000000005,
"loss": 6.9873,
"mean_token_accuracy": 0.13877484947443008,
"num_tokens": 2275413.0,
"step": 915
},
{
"entropy": 7.1140893459320065,
"epoch": 1.0760233918128654,
"grad_norm": 1.0,
"learning_rate": 0.00045950000000000006,
"loss": 6.97,
"mean_token_accuracy": 0.14143042415380477,
"num_tokens": 2288165.0,
"step": 920
},
{
"entropy": 7.182982397079468,
"epoch": 1.0818713450292399,
"grad_norm": 1.0078125,
"learning_rate": 0.000462,
"loss": 6.9813,
"mean_token_accuracy": 0.13951057344675064,
"num_tokens": 2300574.0,
"step": 925
},
{
"entropy": 7.078811120986939,
"epoch": 1.087719298245614,
"grad_norm": 1.2734375,
"learning_rate": 0.0004645,
"loss": 6.9801,
"mean_token_accuracy": 0.1362003989517689,
"num_tokens": 2311631.0,
"step": 930
},
{
"entropy": 7.15027084350586,
"epoch": 1.0935672514619883,
"grad_norm": 1.0859375,
"learning_rate": 0.000467,
"loss": 6.9854,
"mean_token_accuracy": 0.14325082898139954,
"num_tokens": 2324312.0,
"step": 935
},
{
"entropy": 7.1474010944366455,
"epoch": 1.0994152046783625,
"grad_norm": 1.015625,
"learning_rate": 0.0004695,
"loss": 7.0872,
"mean_token_accuracy": 0.13512967377901078,
"num_tokens": 2337747.0,
"step": 940
},
{
"entropy": 7.074398517608643,
"epoch": 1.1052631578947367,
"grad_norm": 0.92578125,
"learning_rate": 0.000472,
"loss": 7.0155,
"mean_token_accuracy": 0.14261559247970582,
"num_tokens": 2349830.0,
"step": 945
},
{
"entropy": 7.10729398727417,
"epoch": 1.1111111111111112,
"grad_norm": 1.21875,
"learning_rate": 0.0004745,
"loss": 6.9669,
"mean_token_accuracy": 0.1496504321694374,
"num_tokens": 2361825.0,
"step": 950
},
{
"entropy": 7.026542329788208,
"epoch": 1.1169590643274854,
"grad_norm": 0.9765625,
"learning_rate": 0.000477,
"loss": 6.9057,
"mean_token_accuracy": 0.14415617287158966,
"num_tokens": 2373963.0,
"step": 955
},
{
"entropy": 7.144531345367431,
"epoch": 1.1228070175438596,
"grad_norm": 1.015625,
"learning_rate": 0.0004795,
"loss": 7.056,
"mean_token_accuracy": 0.13805767446756362,
"num_tokens": 2386348.0,
"step": 960
},
{
"entropy": 7.1077625274658205,
"epoch": 1.128654970760234,
"grad_norm": 1.0234375,
"learning_rate": 0.000482,
"loss": 6.9714,
"mean_token_accuracy": 0.13752444088459015,
"num_tokens": 2398338.0,
"step": 965
},
{
"entropy": 7.0632524490356445,
"epoch": 1.1345029239766082,
"grad_norm": 1.046875,
"learning_rate": 0.0004845,
"loss": 6.9275,
"mean_token_accuracy": 0.14290639460086824,
"num_tokens": 2411855.0,
"step": 970
},
{
"entropy": 7.059965705871582,
"epoch": 1.1403508771929824,
"grad_norm": 1.0703125,
"learning_rate": 0.000487,
"loss": 7.0115,
"mean_token_accuracy": 0.1466517761349678,
"num_tokens": 2424271.0,
"step": 975
},
{
"entropy": 7.02401294708252,
"epoch": 1.1461988304093567,
"grad_norm": 1.0703125,
"learning_rate": 0.0004895,
"loss": 7.0055,
"mean_token_accuracy": 0.1405914232134819,
"num_tokens": 2436668.0,
"step": 980
},
{
"entropy": 7.073333978652954,
"epoch": 1.1520467836257309,
"grad_norm": 1.0625,
"learning_rate": 0.000492,
"loss": 6.9482,
"mean_token_accuracy": 0.1412999577820301,
"num_tokens": 2449451.0,
"step": 985
},
{
"entropy": 7.103520298004151,
"epoch": 1.1578947368421053,
"grad_norm": 1.0234375,
"learning_rate": 0.0004945,
"loss": 7.0655,
"mean_token_accuracy": 0.13647906705737115,
"num_tokens": 2463107.0,
"step": 990
},
{
"entropy": 7.134810161590576,
"epoch": 1.1637426900584795,
"grad_norm": 1.015625,
"learning_rate": 0.000497,
"loss": 7.0015,
"mean_token_accuracy": 0.13996057361364364,
"num_tokens": 2475443.0,
"step": 995
},
{
"entropy": 7.027540588378907,
"epoch": 1.1695906432748537,
"grad_norm": 0.96875,
"learning_rate": 0.0004995,
"loss": 6.9309,
"mean_token_accuracy": 0.14212803095579146,
"num_tokens": 2486884.0,
"step": 1000
},
{
"epoch": 1.1695906432748537,
"eval_entropy": 7.025503185294722,
"eval_loss": 7.157474517822266,
"eval_mean_token_accuracy": 0.13106913304720127,
"eval_num_tokens": 2486884.0,
"eval_runtime": 2.2531,
"eval_samples_per_second": 1201.45,
"eval_steps_per_second": 150.459,
"step": 1000
},
{
"entropy": 6.979807329177857,
"epoch": 1.1754385964912282,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999996875147963,
"loss": 6.892,
"mean_token_accuracy": 0.15283920094370843,
"num_tokens": 2499310.0,
"step": 1005
},
{
"entropy": 7.101342678070068,
"epoch": 1.1812865497076024,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999984180451436,
"loss": 7.0351,
"mean_token_accuracy": 0.1321335293352604,
"num_tokens": 2512230.0,
"step": 1010
},
{
"entropy": 7.048287725448608,
"epoch": 1.1871345029239766,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999961720662221,
"loss": 7.0786,
"mean_token_accuracy": 0.13702130541205407,
"num_tokens": 2525255.0,
"step": 1015
},
{
"entropy": 7.002170991897583,
"epoch": 1.1929824561403508,
"grad_norm": 1.109375,
"learning_rate": 0.0004999929495877796,
"loss": 6.9527,
"mean_token_accuracy": 0.1432950221002102,
"num_tokens": 2537672.0,
"step": 1020
},
{
"entropy": 6.989665222167969,
"epoch": 1.198830409356725,
"grad_norm": 0.9375,
"learning_rate": 0.0004999887506238018,
"loss": 6.9835,
"mean_token_accuracy": 0.14342591613531114,
"num_tokens": 2550371.0,
"step": 1025
},
{
"entropy": 7.050039863586425,
"epoch": 1.2046783625730995,
"grad_norm": 0.9765625,
"learning_rate": 0.0004999835751925124,
"loss": 7.0056,
"mean_token_accuracy": 0.13500169217586516,
"num_tokens": 2563708.0,
"step": 1030
},
{
"entropy": 7.09551887512207,
"epoch": 1.2105263157894737,
"grad_norm": 0.98828125,
"learning_rate": 0.0004999774233163734,
"loss": 6.9677,
"mean_token_accuracy": 0.13617606908082963,
"num_tokens": 2576529.0,
"step": 1035
},
{
"entropy": 7.029808092117309,
"epoch": 1.2163742690058479,
"grad_norm": 0.984375,
"learning_rate": 0.0004999702950220841,
"loss": 6.9991,
"mean_token_accuracy": 0.13568439111113548,
"num_tokens": 2588970.0,
"step": 1040
},
{
"entropy": 7.0275428771972654,
"epoch": 1.2222222222222223,
"grad_norm": 1.03125,
"learning_rate": 0.0004999621903405819,
"loss": 6.9464,
"mean_token_accuracy": 0.1417937010526657,
"num_tokens": 2601134.0,
"step": 1045
},
{
"entropy": 7.014184284210205,
"epoch": 1.2280701754385965,
"grad_norm": 1.0390625,
"learning_rate": 0.000499953109307042,
"loss": 7.0107,
"mean_token_accuracy": 0.13796233609318734,
"num_tokens": 2612670.0,
"step": 1050
},
{
"entropy": 6.969291162490845,
"epoch": 1.2339181286549707,
"grad_norm": 0.94921875,
"learning_rate": 0.0004999430519608762,
"loss": 6.9681,
"mean_token_accuracy": 0.1497281849384308,
"num_tokens": 2625157.0,
"step": 1055
},
{
"entropy": 7.100042295455933,
"epoch": 1.239766081871345,
"grad_norm": 0.9765625,
"learning_rate": 0.0004999320183457346,
"loss": 6.9569,
"mean_token_accuracy": 0.13751377016305924,
"num_tokens": 2638318.0,
"step": 1060
},
{
"entropy": 7.008196592330933,
"epoch": 1.2456140350877192,
"grad_norm": 1.0625,
"learning_rate": 0.0004999200085095035,
"loss": 6.9446,
"mean_token_accuracy": 0.14191727340221405,
"num_tokens": 2651024.0,
"step": 1065
},
{
"entropy": 6.810675716400146,
"epoch": 1.2514619883040936,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999070225043068,
"loss": 6.8837,
"mean_token_accuracy": 0.1498500719666481,
"num_tokens": 2665568.0,
"step": 1070
},
{
"entropy": 7.1478063583374025,
"epoch": 1.2573099415204678,
"grad_norm": 0.96484375,
"learning_rate": 0.0004998930603865042,
"loss": 7.0059,
"mean_token_accuracy": 0.13958486467599868,
"num_tokens": 2678748.0,
"step": 1075
},
{
"entropy": 7.033294582366944,
"epoch": 1.263157894736842,
"grad_norm": 0.9921875,
"learning_rate": 0.0004998781222166929,
"loss": 6.9588,
"mean_token_accuracy": 0.14076827839016914,
"num_tokens": 2690682.0,
"step": 1080
},
{
"entropy": 6.9186183452606205,
"epoch": 1.2690058479532165,
"grad_norm": 0.97265625,
"learning_rate": 0.0004998622080597052,
"loss": 6.9853,
"mean_token_accuracy": 0.13392557948827744,
"num_tokens": 2703077.0,
"step": 1085
},
{
"entropy": 7.00119915008545,
"epoch": 1.2748538011695907,
"grad_norm": 1.0703125,
"learning_rate": 0.0004998453179846098,
"loss": 6.9615,
"mean_token_accuracy": 0.13907949030399322,
"num_tokens": 2714609.0,
"step": 1090
},
{
"entropy": 7.074500322341919,
"epoch": 1.280701754385965,
"grad_norm": 0.97265625,
"learning_rate": 0.0004998274520647109,
"loss": 6.9282,
"mean_token_accuracy": 0.14250491559505463,
"num_tokens": 2726907.0,
"step": 1095
},
{
"entropy": 6.838219261169433,
"epoch": 1.286549707602339,
"grad_norm": 1.1171875,
"learning_rate": 0.0004998086103775478,
"loss": 6.9617,
"mean_token_accuracy": 0.14029049202799798,
"num_tokens": 2739248.0,
"step": 1100
},
{
"entropy": 7.036256265640259,
"epoch": 1.2923976608187133,
"grad_norm": 1.046875,
"learning_rate": 0.0004997887930048947,
"loss": 6.8889,
"mean_token_accuracy": 0.14379191845655442,
"num_tokens": 2753842.0,
"step": 1105
},
{
"entropy": 6.885680103302002,
"epoch": 1.2982456140350878,
"grad_norm": 0.98046875,
"learning_rate": 0.0004997680000327607,
"loss": 6.9546,
"mean_token_accuracy": 0.13402040377259256,
"num_tokens": 2767249.0,
"step": 1110
},
{
"entropy": 7.031604290008545,
"epoch": 1.304093567251462,
"grad_norm": 1.0625,
"learning_rate": 0.0004997462315513884,
"loss": 6.9374,
"mean_token_accuracy": 0.1431535616517067,
"num_tokens": 2779657.0,
"step": 1115
},
{
"entropy": 6.946121454238892,
"epoch": 1.3099415204678362,
"grad_norm": 1.6328125,
"learning_rate": 0.000499723487655255,
"loss": 6.8436,
"mean_token_accuracy": 0.14475864619016648,
"num_tokens": 2792760.0,
"step": 1120
},
{
"entropy": 6.936051940917968,
"epoch": 1.3157894736842106,
"grad_norm": 0.94140625,
"learning_rate": 0.0004996997684430703,
"loss": 6.8776,
"mean_token_accuracy": 0.14265076816082,
"num_tokens": 2805277.0,
"step": 1125
},
{
"entropy": 6.961184453964234,
"epoch": 1.3216374269005848,
"grad_norm": 0.96484375,
"learning_rate": 0.0004996750740177775,
"loss": 7.0026,
"mean_token_accuracy": 0.1412787467241287,
"num_tokens": 2818341.0,
"step": 1130
},
{
"entropy": 7.070265436172486,
"epoch": 1.327485380116959,
"grad_norm": 0.9609375,
"learning_rate": 0.0004996494044865521,
"loss": 6.9404,
"mean_token_accuracy": 0.1430366963148117,
"num_tokens": 2830199.0,
"step": 1135
},
{
"entropy": 6.805611324310303,
"epoch": 1.3333333333333333,
"grad_norm": 1.0,
"learning_rate": 0.0004996227599608018,
"loss": 6.8311,
"mean_token_accuracy": 0.14909116476774215,
"num_tokens": 2842141.0,
"step": 1140
},
{
"entropy": 7.002711963653565,
"epoch": 1.3391812865497075,
"grad_norm": 1.078125,
"learning_rate": 0.0004995951405561658,
"loss": 6.9512,
"mean_token_accuracy": 0.13545295223593712,
"num_tokens": 2854929.0,
"step": 1145
},
{
"entropy": 6.921211051940918,
"epoch": 1.345029239766082,
"grad_norm": 1.09375,
"learning_rate": 0.0004995665463925142,
"loss": 6.9319,
"mean_token_accuracy": 0.1432844214141369,
"num_tokens": 2867034.0,
"step": 1150
},
{
"entropy": 6.984521532058716,
"epoch": 1.3508771929824561,
"grad_norm": 0.91015625,
"learning_rate": 0.0004995369775939478,
"loss": 6.8507,
"mean_token_accuracy": 0.14970757812261581,
"num_tokens": 2881254.0,
"step": 1155
},
{
"entropy": 6.876342678070069,
"epoch": 1.3567251461988303,
"grad_norm": 0.9296875,
"learning_rate": 0.0004995064342887972,
"loss": 6.9515,
"mean_token_accuracy": 0.14343279600143433,
"num_tokens": 2894592.0,
"step": 1160
},
{
"entropy": 6.904094123840332,
"epoch": 1.3625730994152048,
"grad_norm": 1.0390625,
"learning_rate": 0.0004994749166096226,
"loss": 6.8629,
"mean_token_accuracy": 0.14436162114143372,
"num_tokens": 2907196.0,
"step": 1165
},
{
"entropy": 6.960309219360352,
"epoch": 1.368421052631579,
"grad_norm": 0.95703125,
"learning_rate": 0.0004994424246932129,
"loss": 6.9383,
"mean_token_accuracy": 0.1376577876508236,
"num_tokens": 2920398.0,
"step": 1170
},
{
"entropy": 6.878757905960083,
"epoch": 1.3742690058479532,
"grad_norm": 0.98828125,
"learning_rate": 0.0004994089586805856,
"loss": 6.8508,
"mean_token_accuracy": 0.14595492482185363,
"num_tokens": 2932052.0,
"step": 1175
},
{
"entropy": 7.06005220413208,
"epoch": 1.3801169590643274,
"grad_norm": 1.0,
"learning_rate": 0.0004993745187169852,
"loss": 6.9287,
"mean_token_accuracy": 0.1375073656439781,
"num_tokens": 2943620.0,
"step": 1180
},
{
"entropy": 6.855983829498291,
"epoch": 1.3859649122807016,
"grad_norm": 0.984375,
"learning_rate": 0.000499339104951884,
"loss": 6.8938,
"mean_token_accuracy": 0.15123256742954255,
"num_tokens": 2956159.0,
"step": 1185
},
{
"entropy": 6.924196577072143,
"epoch": 1.391812865497076,
"grad_norm": 0.9921875,
"learning_rate": 0.00049930271753898,
"loss": 6.9,
"mean_token_accuracy": 0.14046342819929122,
"num_tokens": 2967332.0,
"step": 1190
},
{
"entropy": 6.889228439331054,
"epoch": 1.3976608187134503,
"grad_norm": 0.98046875,
"learning_rate": 0.0004992653566361974,
"loss": 6.9015,
"mean_token_accuracy": 0.1373600736260414,
"num_tokens": 2980168.0,
"step": 1195
},
{
"entropy": 6.93927321434021,
"epoch": 1.4035087719298245,
"grad_norm": 0.96484375,
"learning_rate": 0.0004992270224056849,
"loss": 6.9826,
"mean_token_accuracy": 0.13944907411932944,
"num_tokens": 2993509.0,
"step": 1200
},
{
"entropy": 6.949486160278321,
"epoch": 1.409356725146199,
"grad_norm": 0.98828125,
"learning_rate": 0.0004991877150138162,
"loss": 6.8945,
"mean_token_accuracy": 0.14218948408961296,
"num_tokens": 3005827.0,
"step": 1205
},
{
"entropy": 6.871033239364624,
"epoch": 1.4152046783625731,
"grad_norm": 0.9140625,
"learning_rate": 0.0004991474346311879,
"loss": 6.8768,
"mean_token_accuracy": 0.14366292506456374,
"num_tokens": 3018968.0,
"step": 1210
},
{
"entropy": 6.929807806015015,
"epoch": 1.4210526315789473,
"grad_norm": 1.046875,
"learning_rate": 0.0004991061814326198,
"loss": 6.9135,
"mean_token_accuracy": 0.13588419929146767,
"num_tokens": 3031892.0,
"step": 1215
},
{
"entropy": 6.92696590423584,
"epoch": 1.4269005847953216,
"grad_norm": 1.6640625,
"learning_rate": 0.0004990639555971539,
"loss": 6.8453,
"mean_token_accuracy": 0.150522081553936,
"num_tokens": 3045994.0,
"step": 1220
},
{
"entropy": 6.801218414306641,
"epoch": 1.4327485380116958,
"grad_norm": 0.91015625,
"learning_rate": 0.0004990207573080533,
"loss": 6.8506,
"mean_token_accuracy": 0.149581628292799,
"num_tokens": 3058304.0,
"step": 1225
},
{
"entropy": 6.902471113204956,
"epoch": 1.4385964912280702,
"grad_norm": 0.9765625,
"learning_rate": 0.0004989765867528014,
"loss": 6.8681,
"mean_token_accuracy": 0.1441365472972393,
"num_tokens": 3071102.0,
"step": 1230
},
{
"entropy": 6.81841254234314,
"epoch": 1.4444444444444444,
"grad_norm": 0.96484375,
"learning_rate": 0.0004989314441231019,
"loss": 6.9074,
"mean_token_accuracy": 0.1388896532356739,
"num_tokens": 3084653.0,
"step": 1235
},
{
"entropy": 6.912723541259766,
"epoch": 1.4502923976608186,
"grad_norm": 1.0,
"learning_rate": 0.000498885329614877,
"loss": 6.9009,
"mean_token_accuracy": 0.14601168483495713,
"num_tokens": 3097425.0,
"step": 1240
},
{
"entropy": 6.887241220474243,
"epoch": 1.456140350877193,
"grad_norm": 1.0,
"learning_rate": 0.0004988382434282668,
"loss": 6.9244,
"mean_token_accuracy": 0.14221828803420067,
"num_tokens": 3111898.0,
"step": 1245
},
{
"entropy": 6.905599069595337,
"epoch": 1.4619883040935673,
"grad_norm": 1.0703125,
"learning_rate": 0.000498790185767629,
"loss": 6.8614,
"mean_token_accuracy": 0.14325395748019218,
"num_tokens": 3124271.0,
"step": 1250
},
{
"entropy": 6.929810428619385,
"epoch": 1.4678362573099415,
"grad_norm": 0.94140625,
"learning_rate": 0.000498741156841537,
"loss": 6.9486,
"mean_token_accuracy": 0.1420268952846527,
"num_tokens": 3136344.0,
"step": 1255
},
{
"entropy": 6.911685276031494,
"epoch": 1.4736842105263157,
"grad_norm": 1.1171875,
"learning_rate": 0.00049869115686278,
"loss": 6.8542,
"mean_token_accuracy": 0.14559442698955535,
"num_tokens": 3148299.0,
"step": 1260
},
{
"entropy": 6.937784719467163,
"epoch": 1.47953216374269,
"grad_norm": 0.8671875,
"learning_rate": 0.0004986401860483615,
"loss": 6.9151,
"mean_token_accuracy": 0.14457187727093695,
"num_tokens": 3161537.0,
"step": 1265
},
{
"entropy": 6.829940843582153,
"epoch": 1.4853801169590644,
"grad_norm": 0.9453125,
"learning_rate": 0.0004985882446194984,
"loss": 6.8594,
"mean_token_accuracy": 0.14513164460659028,
"num_tokens": 3173676.0,
"step": 1270
},
{
"entropy": 6.917054986953735,
"epoch": 1.4912280701754386,
"grad_norm": 1.7890625,
"learning_rate": 0.0004985353328016204,
"loss": 6.819,
"mean_token_accuracy": 0.1553460657596588,
"num_tokens": 3186542.0,
"step": 1275
},
{
"entropy": 6.822628211975098,
"epoch": 1.4970760233918128,
"grad_norm": 1.046875,
"learning_rate": 0.0004984814508243685,
"loss": 6.8335,
"mean_token_accuracy": 0.14759107008576394,
"num_tokens": 3197867.0,
"step": 1280
},
{
"entropy": 6.894301414489746,
"epoch": 1.5029239766081872,
"grad_norm": 1.0625,
"learning_rate": 0.0004984265989215942,
"loss": 6.8249,
"mean_token_accuracy": 0.1432124361395836,
"num_tokens": 3209671.0,
"step": 1285
},
{
"entropy": 6.8511247634887695,
"epoch": 1.5087719298245614,
"grad_norm": 1.09375,
"learning_rate": 0.0004983707773313589,
"loss": 6.8196,
"mean_token_accuracy": 0.15341102480888366,
"num_tokens": 3220707.0,
"step": 1290
},
{
"entropy": 6.715557336807251,
"epoch": 1.5146198830409356,
"grad_norm": 1.015625,
"learning_rate": 0.0004983139862959322,
"loss": 6.7834,
"mean_token_accuracy": 0.154496743530035,
"num_tokens": 3235122.0,
"step": 1295
},
{
"entropy": 6.907053995132446,
"epoch": 1.52046783625731,
"grad_norm": 0.94921875,
"learning_rate": 0.0004982562260617912,
"loss": 6.7848,
"mean_token_accuracy": 0.1419589526951313,
"num_tokens": 3246833.0,
"step": 1300
},
{
"entropy": 6.8496777534484865,
"epoch": 1.526315789473684,
"grad_norm": 0.98046875,
"learning_rate": 0.0004981974968796197,
"loss": 6.8826,
"mean_token_accuracy": 0.1394804447889328,
"num_tokens": 3260493.0,
"step": 1305
},
{
"entropy": 6.791020298004151,
"epoch": 1.5321637426900585,
"grad_norm": 1.078125,
"learning_rate": 0.0004981377990043064,
"loss": 6.8302,
"mean_token_accuracy": 0.14329657703638077,
"num_tokens": 3272445.0,
"step": 1310
},
{
"entropy": 6.856331872940063,
"epoch": 1.5380116959064327,
"grad_norm": 0.98828125,
"learning_rate": 0.0004980771326949442,
"loss": 6.8378,
"mean_token_accuracy": 0.14126284345984458,
"num_tokens": 3284360.0,
"step": 1315
},
{
"entropy": 6.84411244392395,
"epoch": 1.543859649122807,
"grad_norm": 0.92578125,
"learning_rate": 0.0004980154982148296,
"loss": 6.8287,
"mean_token_accuracy": 0.14228586554527284,
"num_tokens": 3296841.0,
"step": 1320
},
{
"entropy": 6.804647207260132,
"epoch": 1.5497076023391814,
"grad_norm": 0.9140625,
"learning_rate": 0.0004979528958314604,
"loss": 6.82,
"mean_token_accuracy": 0.142367422580719,
"num_tokens": 3309393.0,
"step": 1325
},
{
"entropy": 6.910936260223389,
"epoch": 1.5555555555555556,
"grad_norm": 0.99609375,
"learning_rate": 0.0004978893258165354,
"loss": 6.9056,
"mean_token_accuracy": 0.1410933665931225,
"num_tokens": 3322094.0,
"step": 1330
},
{
"entropy": 6.825253486633301,
"epoch": 1.5614035087719298,
"grad_norm": 0.9375,
"learning_rate": 0.0004978247884459532,
"loss": 6.8156,
"mean_token_accuracy": 0.14649758189916612,
"num_tokens": 3334468.0,
"step": 1335
},
{
"entropy": 6.852710199356079,
"epoch": 1.5672514619883042,
"grad_norm": 1.03125,
"learning_rate": 0.0004977592839998105,
"loss": 6.8036,
"mean_token_accuracy": 0.14304608553647996,
"num_tokens": 3346367.0,
"step": 1340
},
{
"entropy": 6.791905450820923,
"epoch": 1.5730994152046782,
"grad_norm": 1.0390625,
"learning_rate": 0.0004976928127624013,
"loss": 6.793,
"mean_token_accuracy": 0.14914015531539918,
"num_tokens": 3357463.0,
"step": 1345
},
{
"entropy": 6.742348051071167,
"epoch": 1.5789473684210527,
"grad_norm": 1.1953125,
"learning_rate": 0.0004976253750222157,
"loss": 6.7969,
"mean_token_accuracy": 0.14524156600236893,
"num_tokens": 3369465.0,
"step": 1350
},
{
"entropy": 6.838311195373535,
"epoch": 1.5847953216374269,
"grad_norm": 0.984375,
"learning_rate": 0.0004975569710719382,
"loss": 6.9109,
"mean_token_accuracy": 0.1383696511387825,
"num_tokens": 3382859.0,
"step": 1355
},
{
"entropy": 6.865691041946411,
"epoch": 1.590643274853801,
"grad_norm": 0.9921875,
"learning_rate": 0.0004974876012084468,
"loss": 6.8452,
"mean_token_accuracy": 0.14151466339826585,
"num_tokens": 3395431.0,
"step": 1360
},
{
"entropy": 6.795973062515259,
"epoch": 1.5964912280701755,
"grad_norm": 1.15625,
"learning_rate": 0.0004974172657328116,
"loss": 6.7491,
"mean_token_accuracy": 0.14309013560414313,
"num_tokens": 3408388.0,
"step": 1365
},
{
"entropy": 6.784850358963013,
"epoch": 1.6023391812865497,
"grad_norm": 1.1953125,
"learning_rate": 0.0004973459649502939,
"loss": 6.7616,
"mean_token_accuracy": 0.14853412210941314,
"num_tokens": 3421275.0,
"step": 1370
},
{
"entropy": 6.804135656356811,
"epoch": 1.608187134502924,
"grad_norm": 0.9765625,
"learning_rate": 0.0004972736991703439,
"loss": 6.8119,
"mean_token_accuracy": 0.14620652347803115,
"num_tokens": 3433537.0,
"step": 1375
},
{
"entropy": 6.811048221588135,
"epoch": 1.6140350877192984,
"grad_norm": 1.2109375,
"learning_rate": 0.0004972004687066005,
"loss": 6.7985,
"mean_token_accuracy": 0.14614380970597268,
"num_tokens": 3445167.0,
"step": 1380
},
{
"entropy": 6.662167978286743,
"epoch": 1.6198830409356724,
"grad_norm": 0.90625,
"learning_rate": 0.000497126273876889,
"loss": 6.7943,
"mean_token_accuracy": 0.15630243718624115,
"num_tokens": 3458529.0,
"step": 1385
},
{
"entropy": 6.829443407058716,
"epoch": 1.6257309941520468,
"grad_norm": 0.97265625,
"learning_rate": 0.0004970511150032205,
"loss": 6.7945,
"mean_token_accuracy": 0.14732301533222197,
"num_tokens": 3470315.0,
"step": 1390
},
{
"entropy": 6.75582447052002,
"epoch": 1.631578947368421,
"grad_norm": 0.9921875,
"learning_rate": 0.0004969749924117896,
"loss": 6.701,
"mean_token_accuracy": 0.14506951123476028,
"num_tokens": 3481899.0,
"step": 1395
},
{
"entropy": 6.743738031387329,
"epoch": 1.6374269005847952,
"grad_norm": 1.0,
"learning_rate": 0.0004968979064329741,
"loss": 6.8369,
"mean_token_accuracy": 0.15120789930224418,
"num_tokens": 3494076.0,
"step": 1400
},
{
"entropy": 6.9596514225006105,
"epoch": 1.6432748538011697,
"grad_norm": 0.93359375,
"learning_rate": 0.0004968198574013323,
"loss": 6.8425,
"mean_token_accuracy": 0.13812664225697519,
"num_tokens": 3507670.0,
"step": 1405
},
{
"entropy": 6.792782258987427,
"epoch": 1.6491228070175439,
"grad_norm": 0.9921875,
"learning_rate": 0.000496740845655603,
"loss": 6.7433,
"mean_token_accuracy": 0.14641203954815865,
"num_tokens": 3519704.0,
"step": 1410
},
{
"entropy": 6.787240839004516,
"epoch": 1.654970760233918,
"grad_norm": 1.0,
"learning_rate": 0.0004966608715387029,
"loss": 6.8121,
"mean_token_accuracy": 0.1471842497587204,
"num_tokens": 3531665.0,
"step": 1415
},
{
"entropy": 6.757086753845215,
"epoch": 1.6608187134502925,
"grad_norm": 0.94140625,
"learning_rate": 0.0004965799353977248,
"loss": 6.818,
"mean_token_accuracy": 0.15390484631061555,
"num_tokens": 3543876.0,
"step": 1420
},
{
"entropy": 6.686848163604736,
"epoch": 1.6666666666666665,
"grad_norm": 0.97265625,
"learning_rate": 0.0004964980375839379,
"loss": 6.7267,
"mean_token_accuracy": 0.15046528726816177,
"num_tokens": 3556517.0,
"step": 1425
},
{
"entropy": 6.811561536788941,
"epoch": 1.672514619883041,
"grad_norm": 0.94921875,
"learning_rate": 0.0004964151784527844,
"loss": 6.7403,
"mean_token_accuracy": 0.15317897498607635,
"num_tokens": 3569378.0,
"step": 1430
},
{
"entropy": 6.782307195663452,
"epoch": 1.6783625730994152,
"grad_norm": 1.2890625,
"learning_rate": 0.000496331358363879,
"loss": 6.742,
"mean_token_accuracy": 0.14681075811386107,
"num_tokens": 3581684.0,
"step": 1435
},
{
"entropy": 6.74970383644104,
"epoch": 1.6842105263157894,
"grad_norm": 0.953125,
"learning_rate": 0.0004962465776810066,
"loss": 6.7141,
"mean_token_accuracy": 0.1596361830830574,
"num_tokens": 3594961.0,
"step": 1440
},
{
"entropy": 6.716477537155152,
"epoch": 1.6900584795321638,
"grad_norm": 0.94140625,
"learning_rate": 0.0004961608367721217,
"loss": 6.7673,
"mean_token_accuracy": 0.1440455324947834,
"num_tokens": 3607197.0,
"step": 1445
},
{
"entropy": 6.746997737884522,
"epoch": 1.695906432748538,
"grad_norm": 0.95703125,
"learning_rate": 0.0004960741360093459,
"loss": 6.7691,
"mean_token_accuracy": 0.14761832132935523,
"num_tokens": 3618728.0,
"step": 1450
},
{
"entropy": 6.850161504745484,
"epoch": 1.7017543859649122,
"grad_norm": 1.015625,
"learning_rate": 0.0004959864757689669,
"loss": 6.8823,
"mean_token_accuracy": 0.13523451015353202,
"num_tokens": 3630872.0,
"step": 1455
},
{
"entropy": 6.786969089508057,
"epoch": 1.7076023391812867,
"grad_norm": 1.046875,
"learning_rate": 0.0004958978564314361,
"loss": 6.7577,
"mean_token_accuracy": 0.1469944402575493,
"num_tokens": 3642657.0,
"step": 1460
},
{
"entropy": 6.790769910812378,
"epoch": 1.7134502923976607,
"grad_norm": 0.97265625,
"learning_rate": 0.000495808278381368,
"loss": 6.7841,
"mean_token_accuracy": 0.14498233646154404,
"num_tokens": 3654505.0,
"step": 1465
},
{
"entropy": 6.619856500625611,
"epoch": 1.719298245614035,
"grad_norm": 0.8984375,
"learning_rate": 0.0004957177420075377,
"loss": 6.6889,
"mean_token_accuracy": 0.16183902323246002,
"num_tokens": 3668554.0,
"step": 1470
},
{
"entropy": 6.789491319656372,
"epoch": 1.7251461988304093,
"grad_norm": 1.0,
"learning_rate": 0.0004956262477028797,
"loss": 6.8489,
"mean_token_accuracy": 0.14220609217882157,
"num_tokens": 3681448.0,
"step": 1475
},
{
"entropy": 6.78629059791565,
"epoch": 1.7309941520467835,
"grad_norm": 1.0546875,
"learning_rate": 0.0004955337958644857,
"loss": 6.7818,
"mean_token_accuracy": 0.14065006971359253,
"num_tokens": 3693677.0,
"step": 1480
},
{
"entropy": 6.67854528427124,
"epoch": 1.736842105263158,
"grad_norm": 1.015625,
"learning_rate": 0.0004954403868936032,
"loss": 6.7757,
"mean_token_accuracy": 0.14981362596154213,
"num_tokens": 3706204.0,
"step": 1485
},
{
"entropy": 6.828735208511352,
"epoch": 1.7426900584795322,
"grad_norm": 1.0078125,
"learning_rate": 0.000495346021195634,
"loss": 6.7136,
"mean_token_accuracy": 0.15087691098451614,
"num_tokens": 3718255.0,
"step": 1490
},
{
"entropy": 6.721991968154907,
"epoch": 1.7485380116959064,
"grad_norm": 1.296875,
"learning_rate": 0.0004952506991801319,
"loss": 6.7482,
"mean_token_accuracy": 0.1513845443725586,
"num_tokens": 3730623.0,
"step": 1495
},
{
"entropy": 6.797335052490235,
"epoch": 1.7543859649122808,
"grad_norm": 1.0234375,
"learning_rate": 0.0004951544212608015,
"loss": 6.7744,
"mean_token_accuracy": 0.14590126723051072,
"num_tokens": 3743636.0,
"step": 1500
},
{
"epoch": 1.7543859649122808,
"eval_entropy": 6.667961138891259,
"eval_loss": 6.917067527770996,
"eval_mean_token_accuracy": 0.13639764849558078,
"eval_num_tokens": 3743636.0,
"eval_runtime": 2.2883,
"eval_samples_per_second": 1182.982,
"eval_steps_per_second": 148.146,
"step": 1500
}
],
"logging_steps": 5,
"max_steps": 8540,
"num_input_tokens_seen": 0,
"num_train_epochs": 10,
"save_steps": 500,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": false
},
"attributes": {}
}
},
"total_flos": 5416136769024000.0,
"train_batch_size": 16,
"trial_name": null,
"trial_params": null
}
Reference in New Issue Copy Permalink