fa35c70621
Model: fpadovani/jpn-jpan-100mb-after-ppt-Dp-10mb-ckpt500_seed3407 Source: Original Platform
1046 lines
28 KiB
JSON
1046 lines
28 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.5847953216374269,
|
|
"eval_steps": 500,
|
|
"global_step": 500,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 7.7522684097290036,
|
|
"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
|
|
}
|
|
],
|
|
"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": 1810059724800000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|