fa35c70621
Model: fpadovani/jpn-jpan-100mb-after-ppt-Dp-10mb-ckpt500_seed3407 Source: Original Platform
2057 lines
54 KiB
JSON
2057 lines
54 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 1.1695906432748537,
|
|
"eval_steps": 500,
|
|
"global_step": 1000,
|
|
"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
|
|
}
|
|
],
|
|
"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": 3593730995712000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|