14677 lines
411 KiB
JSON
14677 lines
411 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 2.0,
|
|
"eval_steps": 500,
|
|
"global_step": 14634,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 1.828515625,
|
|
"epoch": 0.0013667737306088978,
|
|
"grad_norm": 248.49386194698462,
|
|
"learning_rate": 1.0227272727272728e-07,
|
|
"loss": 1.7177,
|
|
"mean_token_accuracy": 0.6351684600114822,
|
|
"num_tokens": 1016869.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 1.834375,
|
|
"epoch": 0.0027335474612177955,
|
|
"grad_norm": 214.04200415353827,
|
|
"learning_rate": 2.1590909090909094e-07,
|
|
"loss": 1.7064,
|
|
"mean_token_accuracy": 0.642715772986412,
|
|
"num_tokens": 1980699.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 1.808984375,
|
|
"epoch": 0.004100321191826693,
|
|
"grad_norm": 216.73854460555742,
|
|
"learning_rate": 3.2954545454545455e-07,
|
|
"loss": 1.7136,
|
|
"mean_token_accuracy": 0.6395897299051285,
|
|
"num_tokens": 2988245.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 1.823046875,
|
|
"epoch": 0.005467094922435591,
|
|
"grad_norm": 192.80752375896805,
|
|
"learning_rate": 4.431818181818182e-07,
|
|
"loss": 1.734,
|
|
"mean_token_accuracy": 0.6381109833717347,
|
|
"num_tokens": 3975995.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 1.73359375,
|
|
"epoch": 0.006833868653044488,
|
|
"grad_norm": 145.00686557537884,
|
|
"learning_rate": 5.568181818181818e-07,
|
|
"loss": 1.6069,
|
|
"mean_token_accuracy": 0.6555261790752411,
|
|
"num_tokens": 4978502.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 1.75625,
|
|
"epoch": 0.008200642383653386,
|
|
"grad_norm": 181.11002442528138,
|
|
"learning_rate": 6.704545454545456e-07,
|
|
"loss": 1.627,
|
|
"mean_token_accuracy": 0.6499351263046265,
|
|
"num_tokens": 6027883.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 1.77109375,
|
|
"epoch": 0.009567416114262284,
|
|
"grad_norm": 115.52868032753295,
|
|
"learning_rate": 7.840909090909092e-07,
|
|
"loss": 1.6449,
|
|
"mean_token_accuracy": 0.6455243289470672,
|
|
"num_tokens": 7061996.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 1.676171875,
|
|
"epoch": 0.010934189844871182,
|
|
"grad_norm": 63.46030563226427,
|
|
"learning_rate": 8.977272727272728e-07,
|
|
"loss": 1.5432,
|
|
"mean_token_accuracy": 0.6612617343664169,
|
|
"num_tokens": 8084402.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 1.56328125,
|
|
"epoch": 0.01230096357548008,
|
|
"grad_norm": 63.84482819861115,
|
|
"learning_rate": 1.0113636363636365e-06,
|
|
"loss": 1.425,
|
|
"mean_token_accuracy": 0.6762066096067428,
|
|
"num_tokens": 9025654.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 1.468359375,
|
|
"epoch": 0.013667737306088976,
|
|
"grad_norm": 55.8279862930327,
|
|
"learning_rate": 1.125e-06,
|
|
"loss": 1.365,
|
|
"mean_token_accuracy": 0.68993439078331,
|
|
"num_tokens": 10036395.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 1.3890625,
|
|
"epoch": 0.015034511036697874,
|
|
"grad_norm": 7.0083321982125195,
|
|
"learning_rate": 1.2386363636363638e-06,
|
|
"loss": 1.2975,
|
|
"mean_token_accuracy": 0.7006600052118301,
|
|
"num_tokens": 11045728.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 1.328125,
|
|
"epoch": 0.016401284767306772,
|
|
"grad_norm": 7.594073220388958,
|
|
"learning_rate": 1.3522727272727273e-06,
|
|
"loss": 1.2708,
|
|
"mean_token_accuracy": 0.7048319876194,
|
|
"num_tokens": 12023582.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 1.2359375,
|
|
"epoch": 0.01776805849791567,
|
|
"grad_norm": 6.587078992248842,
|
|
"learning_rate": 1.465909090909091e-06,
|
|
"loss": 1.2022,
|
|
"mean_token_accuracy": 0.7148551285266876,
|
|
"num_tokens": 13030217.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 1.182421875,
|
|
"epoch": 0.019134832228524568,
|
|
"grad_norm": 2.4194518931082394,
|
|
"learning_rate": 1.5795454545454547e-06,
|
|
"loss": 1.1932,
|
|
"mean_token_accuracy": 0.7160738497972489,
|
|
"num_tokens": 14048615.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 1.1880859375,
|
|
"epoch": 0.020501605959133466,
|
|
"grad_norm": 4.88863584405906,
|
|
"learning_rate": 1.6931818181818182e-06,
|
|
"loss": 1.1909,
|
|
"mean_token_accuracy": 0.7177899748086929,
|
|
"num_tokens": 15078495.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 1.161328125,
|
|
"epoch": 0.021868379689742364,
|
|
"grad_norm": 2.128503985628064,
|
|
"learning_rate": 1.8068181818181822e-06,
|
|
"loss": 1.1557,
|
|
"mean_token_accuracy": 0.7219643980264664,
|
|
"num_tokens": 16103047.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 1.1798828125,
|
|
"epoch": 0.023235153420351262,
|
|
"grad_norm": 1.238174812147721,
|
|
"learning_rate": 1.9204545454545457e-06,
|
|
"loss": 1.1763,
|
|
"mean_token_accuracy": 0.7188547044992447,
|
|
"num_tokens": 17028892.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 1.081640625,
|
|
"epoch": 0.02460192715096016,
|
|
"grad_norm": 1.2509059345042137,
|
|
"learning_rate": 2.034090909090909e-06,
|
|
"loss": 1.0797,
|
|
"mean_token_accuracy": 0.7354021340608596,
|
|
"num_tokens": 18024902.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 1.0771484375,
|
|
"epoch": 0.025968700881569055,
|
|
"grad_norm": 1.0295095748815513,
|
|
"learning_rate": 2.147727272727273e-06,
|
|
"loss": 1.0843,
|
|
"mean_token_accuracy": 0.7327646136283874,
|
|
"num_tokens": 19080401.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 1.12421875,
|
|
"epoch": 0.027335474612177953,
|
|
"grad_norm": 0.937276223825342,
|
|
"learning_rate": 2.2613636363636366e-06,
|
|
"loss": 1.1362,
|
|
"mean_token_accuracy": 0.7275884687900543,
|
|
"num_tokens": 20026588.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 1.1025390625,
|
|
"epoch": 0.02870224834278685,
|
|
"grad_norm": 0.6645589873322946,
|
|
"learning_rate": 2.375e-06,
|
|
"loss": 1.1059,
|
|
"mean_token_accuracy": 0.7286990463733674,
|
|
"num_tokens": 21044783.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 1.101953125,
|
|
"epoch": 0.03006902207339575,
|
|
"grad_norm": 0.7757111991571094,
|
|
"learning_rate": 2.488636363636364e-06,
|
|
"loss": 1.1027,
|
|
"mean_token_accuracy": 0.732992160320282,
|
|
"num_tokens": 22052506.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 1.041796875,
|
|
"epoch": 0.03143579580400465,
|
|
"grad_norm": 0.6399649807563591,
|
|
"learning_rate": 2.6022727272727276e-06,
|
|
"loss": 1.041,
|
|
"mean_token_accuracy": 0.7428334563970566,
|
|
"num_tokens": 23069235.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 1.0814453125,
|
|
"epoch": 0.032802569534613545,
|
|
"grad_norm": 0.5223507534203075,
|
|
"learning_rate": 2.715909090909091e-06,
|
|
"loss": 1.0781,
|
|
"mean_token_accuracy": 0.7352965503931046,
|
|
"num_tokens": 24079191.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 1.0794921875,
|
|
"epoch": 0.03416934326522244,
|
|
"grad_norm": 0.8118138245512392,
|
|
"learning_rate": 2.829545454545455e-06,
|
|
"loss": 1.0717,
|
|
"mean_token_accuracy": 0.7389876693487167,
|
|
"num_tokens": 25079457.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 1.043359375,
|
|
"epoch": 0.03553611699583134,
|
|
"grad_norm": 0.6052869352254817,
|
|
"learning_rate": 2.9431818181818185e-06,
|
|
"loss": 1.0425,
|
|
"mean_token_accuracy": 0.7422428488731384,
|
|
"num_tokens": 26091186.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 1.0390625,
|
|
"epoch": 0.036902890726440235,
|
|
"grad_norm": 0.9922384727388065,
|
|
"learning_rate": 3.056818181818182e-06,
|
|
"loss": 1.0396,
|
|
"mean_token_accuracy": 0.74297536611557,
|
|
"num_tokens": 27112339.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 1.0412109375,
|
|
"epoch": 0.038269664457049136,
|
|
"grad_norm": 0.6815472994631442,
|
|
"learning_rate": 3.1704545454545456e-06,
|
|
"loss": 1.0321,
|
|
"mean_token_accuracy": 0.744473534822464,
|
|
"num_tokens": 28134048.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 1.06640625,
|
|
"epoch": 0.03963643818765803,
|
|
"grad_norm": 0.5059089945862689,
|
|
"learning_rate": 3.2840909090909095e-06,
|
|
"loss": 1.0516,
|
|
"mean_token_accuracy": 0.7405686616897583,
|
|
"num_tokens": 29136473.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 1.0416015625,
|
|
"epoch": 0.04100321191826693,
|
|
"grad_norm": 0.3879504767453129,
|
|
"learning_rate": 3.397727272727273e-06,
|
|
"loss": 1.043,
|
|
"mean_token_accuracy": 0.7430653601884842,
|
|
"num_tokens": 30137323.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 1.0708984375,
|
|
"epoch": 0.04236998564887583,
|
|
"grad_norm": 0.5281891626743758,
|
|
"learning_rate": 3.5113636363636365e-06,
|
|
"loss": 1.0654,
|
|
"mean_token_accuracy": 0.7410249888896943,
|
|
"num_tokens": 31163454.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 1.0537109375,
|
|
"epoch": 0.04373675937948473,
|
|
"grad_norm": 0.5867282923964702,
|
|
"learning_rate": 3.625e-06,
|
|
"loss": 1.0381,
|
|
"mean_token_accuracy": 0.7411713182926178,
|
|
"num_tokens": 32141812.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 1.120703125,
|
|
"epoch": 0.04510353311009362,
|
|
"grad_norm": 0.47193272354603105,
|
|
"learning_rate": 3.7386363636363635e-06,
|
|
"loss": 1.1063,
|
|
"mean_token_accuracy": 0.7320977449417114,
|
|
"num_tokens": 33126442.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 1.01171875,
|
|
"epoch": 0.046470306840702524,
|
|
"grad_norm": 0.5693187595515118,
|
|
"learning_rate": 3.852272727272728e-06,
|
|
"loss": 1.0026,
|
|
"mean_token_accuracy": 0.7509290158748627,
|
|
"num_tokens": 34104354.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 0.9912109375,
|
|
"epoch": 0.04783708057131142,
|
|
"grad_norm": 0.4016392900736782,
|
|
"learning_rate": 3.965909090909091e-06,
|
|
"loss": 0.9767,
|
|
"mean_token_accuracy": 0.7551457583904266,
|
|
"num_tokens": 35132692.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 1.05,
|
|
"epoch": 0.04920385430192032,
|
|
"grad_norm": 0.5130063868187086,
|
|
"learning_rate": 4.079545454545455e-06,
|
|
"loss": 1.04,
|
|
"mean_token_accuracy": 0.740850493311882,
|
|
"num_tokens": 36158630.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 1.0673828125,
|
|
"epoch": 0.050570628032529215,
|
|
"grad_norm": 0.5261474227773227,
|
|
"learning_rate": 4.193181818181819e-06,
|
|
"loss": 1.0553,
|
|
"mean_token_accuracy": 0.7408947557210922,
|
|
"num_tokens": 37163234.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 1.078515625,
|
|
"epoch": 0.05193740176313811,
|
|
"grad_norm": 0.48267956047503413,
|
|
"learning_rate": 4.306818181818182e-06,
|
|
"loss": 1.0711,
|
|
"mean_token_accuracy": 0.7371064454317093,
|
|
"num_tokens": 38146361.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 1.0048828125,
|
|
"epoch": 0.05330417549374701,
|
|
"grad_norm": 0.3366497546524972,
|
|
"learning_rate": 4.420454545454546e-06,
|
|
"loss": 1.0106,
|
|
"mean_token_accuracy": 0.7476702988147735,
|
|
"num_tokens": 39174804.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 1.0083984375,
|
|
"epoch": 0.054670949224355905,
|
|
"grad_norm": 0.4153372279648267,
|
|
"learning_rate": 4.53409090909091e-06,
|
|
"loss": 0.9904,
|
|
"mean_token_accuracy": 0.7537746489048004,
|
|
"num_tokens": 40139777.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 1.0759765625,
|
|
"epoch": 0.05603772295496481,
|
|
"grad_norm": 0.5880303320130282,
|
|
"learning_rate": 4.647727272727273e-06,
|
|
"loss": 1.0732,
|
|
"mean_token_accuracy": 0.7351951003074646,
|
|
"num_tokens": 41114574.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 1.0345703125,
|
|
"epoch": 0.0574044966855737,
|
|
"grad_norm": 0.3367576873269941,
|
|
"learning_rate": 4.761363636363637e-06,
|
|
"loss": 1.0385,
|
|
"mean_token_accuracy": 0.7445290088653564,
|
|
"num_tokens": 42124866.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 0.9923828125,
|
|
"epoch": 0.0587712704161826,
|
|
"grad_norm": 0.4356800736990072,
|
|
"learning_rate": 4.875e-06,
|
|
"loss": 0.9867,
|
|
"mean_token_accuracy": 0.7531208842992783,
|
|
"num_tokens": 43150980.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 1.0193359375,
|
|
"epoch": 0.0601380441467915,
|
|
"grad_norm": 0.3482776502198438,
|
|
"learning_rate": 4.988636363636364e-06,
|
|
"loss": 1.0175,
|
|
"mean_token_accuracy": 0.747447994351387,
|
|
"num_tokens": 44191349.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 1.0642578125,
|
|
"epoch": 0.0615048178774004,
|
|
"grad_norm": 0.4171529511515502,
|
|
"learning_rate": 4.996829646329435e-06,
|
|
"loss": 1.0467,
|
|
"mean_token_accuracy": 0.740112054347992,
|
|
"num_tokens": 45193679.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 1.084765625,
|
|
"epoch": 0.0628715916080093,
|
|
"grad_norm": 0.3346217251778849,
|
|
"learning_rate": 4.993307031139919e-06,
|
|
"loss": 1.0813,
|
|
"mean_token_accuracy": 0.7364024251699448,
|
|
"num_tokens": 46212953.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 1.0412109375,
|
|
"epoch": 0.0642383653386182,
|
|
"grad_norm": 0.34306496651994967,
|
|
"learning_rate": 4.989784415950402e-06,
|
|
"loss": 1.0377,
|
|
"mean_token_accuracy": 0.7437080591917038,
|
|
"num_tokens": 47223409.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 1.08515625,
|
|
"epoch": 0.06560513906922709,
|
|
"grad_norm": 0.37178662647589306,
|
|
"learning_rate": 4.986261800760885e-06,
|
|
"loss": 1.081,
|
|
"mean_token_accuracy": 0.7353532493114472,
|
|
"num_tokens": 48219206.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 1.0232421875,
|
|
"epoch": 0.06697191279983598,
|
|
"grad_norm": 0.3063455297910031,
|
|
"learning_rate": 4.9827391855713685e-06,
|
|
"loss": 1.0195,
|
|
"mean_token_accuracy": 0.749971580505371,
|
|
"num_tokens": 49254735.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 1.0412109375,
|
|
"epoch": 0.06833868653044488,
|
|
"grad_norm": 0.37002405909778624,
|
|
"learning_rate": 4.979216570381852e-06,
|
|
"loss": 1.0253,
|
|
"mean_token_accuracy": 0.7451900005340576,
|
|
"num_tokens": 50255750.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 1.0427734375,
|
|
"epoch": 0.06970546026105379,
|
|
"grad_norm": 0.34457856774705153,
|
|
"learning_rate": 4.975693955192335e-06,
|
|
"loss": 1.0409,
|
|
"mean_token_accuracy": 0.7405862122774124,
|
|
"num_tokens": 51239612.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 1.052734375,
|
|
"epoch": 0.07107223399166268,
|
|
"grad_norm": 0.41137457468914235,
|
|
"learning_rate": 4.972171340002819e-06,
|
|
"loss": 1.0476,
|
|
"mean_token_accuracy": 0.7416637241840363,
|
|
"num_tokens": 52276744.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 1.0587890625,
|
|
"epoch": 0.07243900772227158,
|
|
"grad_norm": 0.32446644730671503,
|
|
"learning_rate": 4.968648724813302e-06,
|
|
"loss": 1.0565,
|
|
"mean_token_accuracy": 0.73836929500103,
|
|
"num_tokens": 53273673.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 1.0357421875,
|
|
"epoch": 0.07380578145288047,
|
|
"grad_norm": 0.38454340717735397,
|
|
"learning_rate": 4.965126109623785e-06,
|
|
"loss": 1.0309,
|
|
"mean_token_accuracy": 0.745680570602417,
|
|
"num_tokens": 54252824.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 1.04765625,
|
|
"epoch": 0.07517255518348938,
|
|
"grad_norm": 0.35232754434264374,
|
|
"learning_rate": 4.961603494434268e-06,
|
|
"loss": 1.0479,
|
|
"mean_token_accuracy": 0.7423206478357315,
|
|
"num_tokens": 55242317.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 1.0134765625,
|
|
"epoch": 0.07653932891409827,
|
|
"grad_norm": 0.34585679245698997,
|
|
"learning_rate": 4.958080879244752e-06,
|
|
"loss": 1.0115,
|
|
"mean_token_accuracy": 0.7478535562753678,
|
|
"num_tokens": 56262198.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 1.0021484375,
|
|
"epoch": 0.07790610264470717,
|
|
"grad_norm": 0.32826218279439967,
|
|
"learning_rate": 4.954558264055234e-06,
|
|
"loss": 1.0021,
|
|
"mean_token_accuracy": 0.7482786238193512,
|
|
"num_tokens": 57224893.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 1.009765625,
|
|
"epoch": 0.07927287637531606,
|
|
"grad_norm": 0.36389160818331273,
|
|
"learning_rate": 4.951035648865719e-06,
|
|
"loss": 0.9973,
|
|
"mean_token_accuracy": 0.7509441614151001,
|
|
"num_tokens": 58192939.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 1.0126953125,
|
|
"epoch": 0.08063965010592497,
|
|
"grad_norm": 0.3966179596422678,
|
|
"learning_rate": 4.9475130336762015e-06,
|
|
"loss": 1.0165,
|
|
"mean_token_accuracy": 0.7467415660619736,
|
|
"num_tokens": 59173681.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 1.0453125,
|
|
"epoch": 0.08200642383653386,
|
|
"grad_norm": 0.3277406652623828,
|
|
"learning_rate": 4.943990418486685e-06,
|
|
"loss": 1.0381,
|
|
"mean_token_accuracy": 0.7408169627189636,
|
|
"num_tokens": 60217834.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 1.013671875,
|
|
"epoch": 0.08337319756714276,
|
|
"grad_norm": 0.3064690563470391,
|
|
"learning_rate": 4.9404678032971685e-06,
|
|
"loss": 1.0115,
|
|
"mean_token_accuracy": 0.7492232412099838,
|
|
"num_tokens": 61203226.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 1.0025390625,
|
|
"epoch": 0.08473997129775165,
|
|
"grad_norm": 0.32456962311457466,
|
|
"learning_rate": 4.936945188107651e-06,
|
|
"loss": 1.0078,
|
|
"mean_token_accuracy": 0.7502179771661759,
|
|
"num_tokens": 62218471.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 0.972265625,
|
|
"epoch": 0.08610674502836055,
|
|
"grad_norm": 0.3839206985964994,
|
|
"learning_rate": 4.933422572918135e-06,
|
|
"loss": 0.9633,
|
|
"mean_token_accuracy": 0.7586764991283417,
|
|
"num_tokens": 63196352.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 0.99375,
|
|
"epoch": 0.08747351875896946,
|
|
"grad_norm": 0.38759614592859476,
|
|
"learning_rate": 4.929899957728618e-06,
|
|
"loss": 0.9824,
|
|
"mean_token_accuracy": 0.7531021177768707,
|
|
"num_tokens": 64213947.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 1.003515625,
|
|
"epoch": 0.08884029248957835,
|
|
"grad_norm": 0.34260534392215114,
|
|
"learning_rate": 4.926377342539102e-06,
|
|
"loss": 0.9937,
|
|
"mean_token_accuracy": 0.7506791681051255,
|
|
"num_tokens": 65212813.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 0.990625,
|
|
"epoch": 0.09020706622018725,
|
|
"grad_norm": 0.32433543620286626,
|
|
"learning_rate": 4.922854727349585e-06,
|
|
"loss": 0.9784,
|
|
"mean_token_accuracy": 0.7541175156831741,
|
|
"num_tokens": 66207188.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 1.02578125,
|
|
"epoch": 0.09157383995079614,
|
|
"grad_norm": 0.4032347615121775,
|
|
"learning_rate": 4.919332112160068e-06,
|
|
"loss": 1.0187,
|
|
"mean_token_accuracy": 0.746381652355194,
|
|
"num_tokens": 67233532.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 0.9984375,
|
|
"epoch": 0.09294061368140505,
|
|
"grad_norm": 0.33481452388968924,
|
|
"learning_rate": 4.915809496970551e-06,
|
|
"loss": 0.9938,
|
|
"mean_token_accuracy": 0.7513839036226273,
|
|
"num_tokens": 68263201.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 1.066015625,
|
|
"epoch": 0.09430738741201394,
|
|
"grad_norm": 0.30845656553374545,
|
|
"learning_rate": 4.912286881781035e-06,
|
|
"loss": 1.0755,
|
|
"mean_token_accuracy": 0.7349037498235702,
|
|
"num_tokens": 69260237.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 0.996484375,
|
|
"epoch": 0.09567416114262284,
|
|
"grad_norm": 0.3046711570452363,
|
|
"learning_rate": 4.908764266591518e-06,
|
|
"loss": 0.9907,
|
|
"mean_token_accuracy": 0.7518613755702972,
|
|
"num_tokens": 70257011.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 1.025390625,
|
|
"epoch": 0.09704093487323173,
|
|
"grad_norm": 0.30280659131608345,
|
|
"learning_rate": 4.9052416514020015e-06,
|
|
"loss": 1.0187,
|
|
"mean_token_accuracy": 0.74702889919281,
|
|
"num_tokens": 71242071.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 0.9697265625,
|
|
"epoch": 0.09840770860384064,
|
|
"grad_norm": 0.400031004456019,
|
|
"learning_rate": 4.901719036212484e-06,
|
|
"loss": 0.9694,
|
|
"mean_token_accuracy": 0.7547564297914505,
|
|
"num_tokens": 72300678.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 0.9958984375,
|
|
"epoch": 0.09977448233444954,
|
|
"grad_norm": 0.3162760979545829,
|
|
"learning_rate": 4.898196421022968e-06,
|
|
"loss": 0.9957,
|
|
"mean_token_accuracy": 0.7535901069641113,
|
|
"num_tokens": 73295969.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 0.9875,
|
|
"epoch": 0.10114125606505843,
|
|
"grad_norm": 0.333441187984322,
|
|
"learning_rate": 4.894673805833451e-06,
|
|
"loss": 0.9727,
|
|
"mean_token_accuracy": 0.7554042369127274,
|
|
"num_tokens": 74224367.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 1.0017578125,
|
|
"epoch": 0.10250802979566732,
|
|
"grad_norm": 0.3051102161156666,
|
|
"learning_rate": 4.891151190643935e-06,
|
|
"loss": 0.9995,
|
|
"mean_token_accuracy": 0.750582268834114,
|
|
"num_tokens": 75220927.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 1.003515625,
|
|
"epoch": 0.10387480352627622,
|
|
"grad_norm": 0.2975128300614858,
|
|
"learning_rate": 4.8876285754544175e-06,
|
|
"loss": 0.9965,
|
|
"mean_token_accuracy": 0.7500815868377686,
|
|
"num_tokens": 76264459.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 1.0515625,
|
|
"epoch": 0.10524157725688513,
|
|
"grad_norm": 0.34995617646124555,
|
|
"learning_rate": 4.884105960264901e-06,
|
|
"loss": 1.051,
|
|
"mean_token_accuracy": 0.7411069720983505,
|
|
"num_tokens": 77237576.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 1.0041015625,
|
|
"epoch": 0.10660835098749402,
|
|
"grad_norm": 0.2861456823743024,
|
|
"learning_rate": 4.880583345075385e-06,
|
|
"loss": 0.9896,
|
|
"mean_token_accuracy": 0.7522629499435425,
|
|
"num_tokens": 78233874.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 0.9744140625,
|
|
"epoch": 0.10797512471810292,
|
|
"grad_norm": 0.29353721867219945,
|
|
"learning_rate": 4.877060729885867e-06,
|
|
"loss": 0.9737,
|
|
"mean_token_accuracy": 0.7533451318740845,
|
|
"num_tokens": 79226628.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 0.994921875,
|
|
"epoch": 0.10934189844871181,
|
|
"grad_norm": 0.35617400395727106,
|
|
"learning_rate": 4.873538114696351e-06,
|
|
"loss": 0.9897,
|
|
"mean_token_accuracy": 0.7519093364477157,
|
|
"num_tokens": 80216871.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 1.003515625,
|
|
"epoch": 0.11070867217932072,
|
|
"grad_norm": 0.3328960527563896,
|
|
"learning_rate": 4.870015499506834e-06,
|
|
"loss": 0.9999,
|
|
"mean_token_accuracy": 0.7509783357381821,
|
|
"num_tokens": 81209897.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 0.9884765625,
|
|
"epoch": 0.11207544590992961,
|
|
"grad_norm": 0.2750925332993394,
|
|
"learning_rate": 4.866492884317318e-06,
|
|
"loss": 0.9817,
|
|
"mean_token_accuracy": 0.7512441456317902,
|
|
"num_tokens": 82222143.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 0.9802734375,
|
|
"epoch": 0.11344221964053851,
|
|
"grad_norm": 0.3432128059376145,
|
|
"learning_rate": 4.862970269127801e-06,
|
|
"loss": 0.9856,
|
|
"mean_token_accuracy": 0.7553154408931733,
|
|
"num_tokens": 83210601.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 1.011328125,
|
|
"epoch": 0.1148089933711474,
|
|
"grad_norm": 0.3152296714969121,
|
|
"learning_rate": 4.859447653938284e-06,
|
|
"loss": 1.0016,
|
|
"mean_token_accuracy": 0.7494343310594559,
|
|
"num_tokens": 84189098.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 1.0064453125,
|
|
"epoch": 0.11617576710175631,
|
|
"grad_norm": 0.3240776882076449,
|
|
"learning_rate": 4.855925038748768e-06,
|
|
"loss": 1.0048,
|
|
"mean_token_accuracy": 0.7507609277963638,
|
|
"num_tokens": 85278005.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 0.9994140625,
|
|
"epoch": 0.1175425408323652,
|
|
"grad_norm": 0.3141321673048378,
|
|
"learning_rate": 4.852402423559251e-06,
|
|
"loss": 0.9981,
|
|
"mean_token_accuracy": 0.7510209083557129,
|
|
"num_tokens": 86314347.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 1.039453125,
|
|
"epoch": 0.1189093145629741,
|
|
"grad_norm": 0.31703971870300707,
|
|
"learning_rate": 4.848879808369734e-06,
|
|
"loss": 1.0402,
|
|
"mean_token_accuracy": 0.7419896274805069,
|
|
"num_tokens": 87295027.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 0.9912109375,
|
|
"epoch": 0.120276088293583,
|
|
"grad_norm": 0.30483863787810844,
|
|
"learning_rate": 4.8453571931802175e-06,
|
|
"loss": 0.9732,
|
|
"mean_token_accuracy": 0.7565240651369095,
|
|
"num_tokens": 88315349.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 0.957421875,
|
|
"epoch": 0.12164286202419189,
|
|
"grad_norm": 0.3253358757682383,
|
|
"learning_rate": 4.8418345779907e-06,
|
|
"loss": 0.9579,
|
|
"mean_token_accuracy": 0.758674755692482,
|
|
"num_tokens": 89285381.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 1.0396484375,
|
|
"epoch": 0.1230096357548008,
|
|
"grad_norm": 0.39533044836141,
|
|
"learning_rate": 4.838311962801184e-06,
|
|
"loss": 1.0358,
|
|
"mean_token_accuracy": 0.7446712881326676,
|
|
"num_tokens": 90306461.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 1.0033203125,
|
|
"epoch": 0.12437640948540969,
|
|
"grad_norm": 0.3843051257694837,
|
|
"learning_rate": 4.834789347611667e-06,
|
|
"loss": 0.9979,
|
|
"mean_token_accuracy": 0.7510333359241486,
|
|
"num_tokens": 91255724.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 1.00703125,
|
|
"epoch": 0.1257431832160186,
|
|
"grad_norm": 0.44733855288319,
|
|
"learning_rate": 4.831266732422151e-06,
|
|
"loss": 0.9938,
|
|
"mean_token_accuracy": 0.7501683712005616,
|
|
"num_tokens": 92220294.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 0.94296875,
|
|
"epoch": 0.1271099569466275,
|
|
"grad_norm": 0.40353239333502466,
|
|
"learning_rate": 4.827744117232634e-06,
|
|
"loss": 0.9292,
|
|
"mean_token_accuracy": 0.763161239027977,
|
|
"num_tokens": 93189250.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 0.9896484375,
|
|
"epoch": 0.1284767306772364,
|
|
"grad_norm": 0.3553768874559467,
|
|
"learning_rate": 4.824221502043117e-06,
|
|
"loss": 0.9739,
|
|
"mean_token_accuracy": 0.7562126576900482,
|
|
"num_tokens": 94155449.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 1.001953125,
|
|
"epoch": 0.12984350440784528,
|
|
"grad_norm": 0.3986923255013217,
|
|
"learning_rate": 4.820698886853601e-06,
|
|
"loss": 0.9968,
|
|
"mean_token_accuracy": 0.7483080506324769,
|
|
"num_tokens": 95222208.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 1.0138671875,
|
|
"epoch": 0.13121027813845418,
|
|
"grad_norm": 0.3921397643535812,
|
|
"learning_rate": 4.817176271664084e-06,
|
|
"loss": 1.0037,
|
|
"mean_token_accuracy": 0.7480872094631195,
|
|
"num_tokens": 96240732.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 0.96875,
|
|
"epoch": 0.13257705186906307,
|
|
"grad_norm": 0.31982117343230304,
|
|
"learning_rate": 4.813653656474567e-06,
|
|
"loss": 0.956,
|
|
"mean_token_accuracy": 0.7574336767196655,
|
|
"num_tokens": 97189622.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 0.9939453125,
|
|
"epoch": 0.13394382559967197,
|
|
"grad_norm": 0.39972104198571723,
|
|
"learning_rate": 4.8101310412850505e-06,
|
|
"loss": 0.9966,
|
|
"mean_token_accuracy": 0.7489461094141007,
|
|
"num_tokens": 98193657.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 0.969140625,
|
|
"epoch": 0.13531059933028086,
|
|
"grad_norm": 0.3559811904163674,
|
|
"learning_rate": 4.806608426095534e-06,
|
|
"loss": 0.9732,
|
|
"mean_token_accuracy": 0.7563803851604461,
|
|
"num_tokens": 99244929.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 0.9666015625,
|
|
"epoch": 0.13667737306088976,
|
|
"grad_norm": 0.29022578699498336,
|
|
"learning_rate": 4.803085810906017e-06,
|
|
"loss": 0.9522,
|
|
"mean_token_accuracy": 0.760769784450531,
|
|
"num_tokens": 100253866.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 1.0333984375,
|
|
"epoch": 0.13804414679149868,
|
|
"grad_norm": 0.28603139617636497,
|
|
"learning_rate": 4.7995631957165e-06,
|
|
"loss": 1.0275,
|
|
"mean_token_accuracy": 0.7431843400001525,
|
|
"num_tokens": 101284311.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 0.95234375,
|
|
"epoch": 0.13941092052210757,
|
|
"grad_norm": 0.34555526630968375,
|
|
"learning_rate": 4.796040580526984e-06,
|
|
"loss": 0.9493,
|
|
"mean_token_accuracy": 0.7592301309108734,
|
|
"num_tokens": 102289462.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 0.9822265625,
|
|
"epoch": 0.14077769425271647,
|
|
"grad_norm": 0.3714880479672673,
|
|
"learning_rate": 4.792517965337467e-06,
|
|
"loss": 0.9755,
|
|
"mean_token_accuracy": 0.7548005282878876,
|
|
"num_tokens": 103255021.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 0.965234375,
|
|
"epoch": 0.14214446798332536,
|
|
"grad_norm": 0.38420996770419047,
|
|
"learning_rate": 4.78899535014795e-06,
|
|
"loss": 0.9651,
|
|
"mean_token_accuracy": 0.7577881664037704,
|
|
"num_tokens": 104261992.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 0.96796875,
|
|
"epoch": 0.14351124171393426,
|
|
"grad_norm": 0.2791036813890957,
|
|
"learning_rate": 4.785472734958434e-06,
|
|
"loss": 0.9571,
|
|
"mean_token_accuracy": 0.7569888591766357,
|
|
"num_tokens": 105292243.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 0.98515625,
|
|
"epoch": 0.14487801544454315,
|
|
"grad_norm": 0.4436913712119186,
|
|
"learning_rate": 4.781950119768916e-06,
|
|
"loss": 0.9865,
|
|
"mean_token_accuracy": 0.7512211173772811,
|
|
"num_tokens": 106305001.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 0.9640625,
|
|
"epoch": 0.14624478917515205,
|
|
"grad_norm": 0.28792410267317814,
|
|
"learning_rate": 4.778427504579401e-06,
|
|
"loss": 0.9389,
|
|
"mean_token_accuracy": 0.7600586295127869,
|
|
"num_tokens": 107316342.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 0.934765625,
|
|
"epoch": 0.14761156290576094,
|
|
"grad_norm": 0.2688247247386612,
|
|
"learning_rate": 4.774904889389883e-06,
|
|
"loss": 0.9284,
|
|
"mean_token_accuracy": 0.7632242769002915,
|
|
"num_tokens": 108361044.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 0.9779296875,
|
|
"epoch": 0.14897833663636986,
|
|
"grad_norm": 0.3469429485203149,
|
|
"learning_rate": 4.771382274200367e-06,
|
|
"loss": 0.9722,
|
|
"mean_token_accuracy": 0.7554498761892319,
|
|
"num_tokens": 109356718.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 1.009375,
|
|
"epoch": 0.15034511036697876,
|
|
"grad_norm": 0.27563852157817575,
|
|
"learning_rate": 4.76785965901085e-06,
|
|
"loss": 0.9978,
|
|
"mean_token_accuracy": 0.7485051840543747,
|
|
"num_tokens": 110398127.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 1.0330078125,
|
|
"epoch": 0.15171188409758765,
|
|
"grad_norm": 0.3226058821436729,
|
|
"learning_rate": 4.764337043821333e-06,
|
|
"loss": 1.0318,
|
|
"mean_token_accuracy": 0.742601403594017,
|
|
"num_tokens": 111382725.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 1.01875,
|
|
"epoch": 0.15307865782819655,
|
|
"grad_norm": 0.3241860199478388,
|
|
"learning_rate": 4.760814428631817e-06,
|
|
"loss": 1.006,
|
|
"mean_token_accuracy": 0.74780875146389,
|
|
"num_tokens": 112369399.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 0.9939453125,
|
|
"epoch": 0.15444543155880544,
|
|
"grad_norm": 0.3477772461150356,
|
|
"learning_rate": 4.7572918134423e-06,
|
|
"loss": 0.9768,
|
|
"mean_token_accuracy": 0.7526664793491363,
|
|
"num_tokens": 113312860.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 0.933203125,
|
|
"epoch": 0.15581220528941434,
|
|
"grad_norm": 0.28363759905106295,
|
|
"learning_rate": 4.753769198252783e-06,
|
|
"loss": 0.9331,
|
|
"mean_token_accuracy": 0.7623846232891083,
|
|
"num_tokens": 114320706.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 1.03046875,
|
|
"epoch": 0.15717897902002323,
|
|
"grad_norm": 0.3623247127500949,
|
|
"learning_rate": 4.7502465830632665e-06,
|
|
"loss": 1.0205,
|
|
"mean_token_accuracy": 0.7430798053741455,
|
|
"num_tokens": 115361624.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 0.95234375,
|
|
"epoch": 0.15854575275063212,
|
|
"grad_norm": 0.30706844250726917,
|
|
"learning_rate": 4.74672396787375e-06,
|
|
"loss": 0.939,
|
|
"mean_token_accuracy": 0.7628292888402939,
|
|
"num_tokens": 116373525.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 0.9654296875,
|
|
"epoch": 0.15991252648124102,
|
|
"grad_norm": 0.2782112782808599,
|
|
"learning_rate": 4.743201352684233e-06,
|
|
"loss": 0.954,
|
|
"mean_token_accuracy": 0.7582858383655549,
|
|
"num_tokens": 117386597.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 0.98515625,
|
|
"epoch": 0.16127930021184994,
|
|
"grad_norm": 0.3775045455368937,
|
|
"learning_rate": 4.739678737494716e-06,
|
|
"loss": 0.9755,
|
|
"mean_token_accuracy": 0.7560882836580276,
|
|
"num_tokens": 118360572.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 0.9908203125,
|
|
"epoch": 0.16264607394245884,
|
|
"grad_norm": 0.29968778278761704,
|
|
"learning_rate": 4.7361561223052e-06,
|
|
"loss": 0.9778,
|
|
"mean_token_accuracy": 0.7543432176113128,
|
|
"num_tokens": 119348661.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 0.9603515625,
|
|
"epoch": 0.16401284767306773,
|
|
"grad_norm": 0.3277586027307556,
|
|
"learning_rate": 4.7326335071156834e-06,
|
|
"loss": 0.949,
|
|
"mean_token_accuracy": 0.7595767050981521,
|
|
"num_tokens": 120333852.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 0.9767578125,
|
|
"epoch": 0.16537962140367662,
|
|
"grad_norm": 0.303808794210055,
|
|
"learning_rate": 4.729110891926166e-06,
|
|
"loss": 0.9726,
|
|
"mean_token_accuracy": 0.7543101370334625,
|
|
"num_tokens": 121367501.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 0.973828125,
|
|
"epoch": 0.16674639513428552,
|
|
"grad_norm": 0.3795846575645162,
|
|
"learning_rate": 4.72558827673665e-06,
|
|
"loss": 0.9705,
|
|
"mean_token_accuracy": 0.7550342530012131,
|
|
"num_tokens": 122322265.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 0.981640625,
|
|
"epoch": 0.1681131688648944,
|
|
"grad_norm": 0.4822639111366193,
|
|
"learning_rate": 4.722065661547132e-06,
|
|
"loss": 0.9786,
|
|
"mean_token_accuracy": 0.7559493571519852,
|
|
"num_tokens": 123326121.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 0.989453125,
|
|
"epoch": 0.1694799425955033,
|
|
"grad_norm": 0.31918415139918876,
|
|
"learning_rate": 4.718543046357617e-06,
|
|
"loss": 0.977,
|
|
"mean_token_accuracy": 0.7527957916259765,
|
|
"num_tokens": 124379656.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 0.9462890625,
|
|
"epoch": 0.1708467163261122,
|
|
"grad_norm": 0.3387052830688386,
|
|
"learning_rate": 4.7150204311680995e-06,
|
|
"loss": 0.9383,
|
|
"mean_token_accuracy": 0.7608568161725998,
|
|
"num_tokens": 125327601.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 0.99375,
|
|
"epoch": 0.1722134900567211,
|
|
"grad_norm": 0.32809540728611875,
|
|
"learning_rate": 4.711497815978583e-06,
|
|
"loss": 0.9795,
|
|
"mean_token_accuracy": 0.7533856093883514,
|
|
"num_tokens": 126310987.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 0.99921875,
|
|
"epoch": 0.17358026378733002,
|
|
"grad_norm": 0.4127737542878994,
|
|
"learning_rate": 4.707975200789066e-06,
|
|
"loss": 0.9995,
|
|
"mean_token_accuracy": 0.7486604481935502,
|
|
"num_tokens": 127351646.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 0.9908203125,
|
|
"epoch": 0.1749470375179389,
|
|
"grad_norm": 0.4276960492489289,
|
|
"learning_rate": 4.704452585599549e-06,
|
|
"loss": 0.9811,
|
|
"mean_token_accuracy": 0.7516766130924225,
|
|
"num_tokens": 128345444.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 0.97578125,
|
|
"epoch": 0.1763138112485478,
|
|
"grad_norm": 0.29490894184411554,
|
|
"learning_rate": 4.700929970410033e-06,
|
|
"loss": 0.9654,
|
|
"mean_token_accuracy": 0.7571633994579315,
|
|
"num_tokens": 129339497.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 0.9521484375,
|
|
"epoch": 0.1776805849791567,
|
|
"grad_norm": 0.29971596144367135,
|
|
"learning_rate": 4.697407355220516e-06,
|
|
"loss": 0.9352,
|
|
"mean_token_accuracy": 0.7615818947553634,
|
|
"num_tokens": 130382964.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 0.9712890625,
|
|
"epoch": 0.1790473587097656,
|
|
"grad_norm": 0.34903416214055705,
|
|
"learning_rate": 4.693884740030999e-06,
|
|
"loss": 0.955,
|
|
"mean_token_accuracy": 0.7553958714008331,
|
|
"num_tokens": 131333722.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 0.9953125,
|
|
"epoch": 0.1804141324403745,
|
|
"grad_norm": 0.31496327614902697,
|
|
"learning_rate": 4.690362124841483e-06,
|
|
"loss": 0.9838,
|
|
"mean_token_accuracy": 0.7519759356975555,
|
|
"num_tokens": 132332448.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 0.98359375,
|
|
"epoch": 0.1817809061709834,
|
|
"grad_norm": 0.29700245154180827,
|
|
"learning_rate": 4.686839509651966e-06,
|
|
"loss": 0.9675,
|
|
"mean_token_accuracy": 0.7550416022539139,
|
|
"num_tokens": 133339343.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 0.9796875,
|
|
"epoch": 0.18314767990159228,
|
|
"grad_norm": 0.31855489039424034,
|
|
"learning_rate": 4.683316894462449e-06,
|
|
"loss": 0.9545,
|
|
"mean_token_accuracy": 0.7569522559642792,
|
|
"num_tokens": 134361824.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 0.9625,
|
|
"epoch": 0.18451445363220118,
|
|
"grad_norm": 0.3596367939780353,
|
|
"learning_rate": 4.679794279272933e-06,
|
|
"loss": 0.9648,
|
|
"mean_token_accuracy": 0.7547035336494445,
|
|
"num_tokens": 135366570.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 0.978515625,
|
|
"epoch": 0.1858812273628101,
|
|
"grad_norm": 0.3316590028679052,
|
|
"learning_rate": 4.676271664083416e-06,
|
|
"loss": 0.972,
|
|
"mean_token_accuracy": 0.7538880079984664,
|
|
"num_tokens": 136388166.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 1.025,
|
|
"epoch": 0.187248001093419,
|
|
"grad_norm": 0.4276077187755448,
|
|
"learning_rate": 4.6727490488938995e-06,
|
|
"loss": 1.0107,
|
|
"mean_token_accuracy": 0.7464945942163468,
|
|
"num_tokens": 137334309.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 0.9728515625,
|
|
"epoch": 0.1886147748240279,
|
|
"grad_norm": 0.4018131901551473,
|
|
"learning_rate": 4.669226433704382e-06,
|
|
"loss": 0.9559,
|
|
"mean_token_accuracy": 0.7586585968732834,
|
|
"num_tokens": 138333216.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 0.953125,
|
|
"epoch": 0.18998154855463678,
|
|
"grad_norm": 0.2912282487811292,
|
|
"learning_rate": 4.665703818514866e-06,
|
|
"loss": 0.939,
|
|
"mean_token_accuracy": 0.7628429263830185,
|
|
"num_tokens": 139302001.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 0.9615234375,
|
|
"epoch": 0.19134832228524568,
|
|
"grad_norm": 0.28049174293992296,
|
|
"learning_rate": 4.6621812033253484e-06,
|
|
"loss": 0.9485,
|
|
"mean_token_accuracy": 0.7601240664720536,
|
|
"num_tokens": 140300151.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 0.89609375,
|
|
"epoch": 0.19271509601585457,
|
|
"grad_norm": 0.35452928230963265,
|
|
"learning_rate": 4.658658588135833e-06,
|
|
"loss": 0.8798,
|
|
"mean_token_accuracy": 0.7727393597364426,
|
|
"num_tokens": 141291771.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 0.9462890625,
|
|
"epoch": 0.19408186974646346,
|
|
"grad_norm": 0.3326180101001367,
|
|
"learning_rate": 4.6551359729463155e-06,
|
|
"loss": 0.9214,
|
|
"mean_token_accuracy": 0.7641403555870057,
|
|
"num_tokens": 142345426.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 0.998046875,
|
|
"epoch": 0.19544864347707236,
|
|
"grad_norm": 0.30501602475288836,
|
|
"learning_rate": 4.651613357756799e-06,
|
|
"loss": 0.9961,
|
|
"mean_token_accuracy": 0.7479186505079269,
|
|
"num_tokens": 143350038.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 0.99609375,
|
|
"epoch": 0.19681541720768128,
|
|
"grad_norm": 0.3457853186165989,
|
|
"learning_rate": 4.648090742567283e-06,
|
|
"loss": 0.9958,
|
|
"mean_token_accuracy": 0.7519662231206894,
|
|
"num_tokens": 144364790.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 0.9724609375,
|
|
"epoch": 0.19818219093829018,
|
|
"grad_norm": 0.34204731637379826,
|
|
"learning_rate": 4.644568127377765e-06,
|
|
"loss": 0.9628,
|
|
"mean_token_accuracy": 0.7552968800067902,
|
|
"num_tokens": 145393023.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 0.9626953125,
|
|
"epoch": 0.19954896466889907,
|
|
"grad_norm": 0.32698550100344864,
|
|
"learning_rate": 4.641045512188249e-06,
|
|
"loss": 0.9564,
|
|
"mean_token_accuracy": 0.7580635607242584,
|
|
"num_tokens": 146402141.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 0.9275390625,
|
|
"epoch": 0.20091573839950796,
|
|
"grad_norm": 0.48550793094087297,
|
|
"learning_rate": 4.6375228969987324e-06,
|
|
"loss": 0.9205,
|
|
"mean_token_accuracy": 0.7626504272222518,
|
|
"num_tokens": 147429117.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 0.980078125,
|
|
"epoch": 0.20228251213011686,
|
|
"grad_norm": 0.3355739542478802,
|
|
"learning_rate": 4.634000281809216e-06,
|
|
"loss": 0.963,
|
|
"mean_token_accuracy": 0.7554794758558273,
|
|
"num_tokens": 148449641.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 0.9451171875,
|
|
"epoch": 0.20364928586072575,
|
|
"grad_norm": 0.29785352301342516,
|
|
"learning_rate": 4.630477666619699e-06,
|
|
"loss": 0.9252,
|
|
"mean_token_accuracy": 0.762885183095932,
|
|
"num_tokens": 149435614.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 0.939453125,
|
|
"epoch": 0.20501605959133465,
|
|
"grad_norm": 0.2901571041915355,
|
|
"learning_rate": 4.626955051430182e-06,
|
|
"loss": 0.925,
|
|
"mean_token_accuracy": 0.7650482535362244,
|
|
"num_tokens": 150463227.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 0.9994140625,
|
|
"epoch": 0.20638283332194354,
|
|
"grad_norm": 0.32786251592684346,
|
|
"learning_rate": 4.623432436240665e-06,
|
|
"loss": 0.9806,
|
|
"mean_token_accuracy": 0.7527417212724685,
|
|
"num_tokens": 151395436.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 0.954296875,
|
|
"epoch": 0.20774960705255244,
|
|
"grad_norm": 0.31015181690459587,
|
|
"learning_rate": 4.619909821051149e-06,
|
|
"loss": 0.9337,
|
|
"mean_token_accuracy": 0.7607152670621872,
|
|
"num_tokens": 152324730.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 0.947265625,
|
|
"epoch": 0.20911638078316136,
|
|
"grad_norm": 0.3361464316075905,
|
|
"learning_rate": 4.616387205861632e-06,
|
|
"loss": 0.9398,
|
|
"mean_token_accuracy": 0.7583794295787811,
|
|
"num_tokens": 153320864.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 0.9333984375,
|
|
"epoch": 0.21048315451377025,
|
|
"grad_norm": 0.31081309655964995,
|
|
"learning_rate": 4.6128645906721156e-06,
|
|
"loss": 0.9231,
|
|
"mean_token_accuracy": 0.7635580271482467,
|
|
"num_tokens": 154278191.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 0.9873046875,
|
|
"epoch": 0.21184992824437915,
|
|
"grad_norm": 0.32188647858927183,
|
|
"learning_rate": 4.609341975482598e-06,
|
|
"loss": 0.9782,
|
|
"mean_token_accuracy": 0.75376156270504,
|
|
"num_tokens": 155275095.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 0.96328125,
|
|
"epoch": 0.21321670197498804,
|
|
"grad_norm": 0.3398452166320509,
|
|
"learning_rate": 4.605819360293082e-06,
|
|
"loss": 0.9501,
|
|
"mean_token_accuracy": 0.7576699465513229,
|
|
"num_tokens": 156313420.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 0.9224609375,
|
|
"epoch": 0.21458347570559694,
|
|
"grad_norm": 0.282029226003175,
|
|
"learning_rate": 4.602296745103565e-06,
|
|
"loss": 0.9157,
|
|
"mean_token_accuracy": 0.7656923651695251,
|
|
"num_tokens": 157298105.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 0.9736328125,
|
|
"epoch": 0.21595024943620583,
|
|
"grad_norm": 0.3081291129976526,
|
|
"learning_rate": 4.598774129914049e-06,
|
|
"loss": 0.9784,
|
|
"mean_token_accuracy": 0.75456602871418,
|
|
"num_tokens": 158304186.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 0.977734375,
|
|
"epoch": 0.21731702316681473,
|
|
"grad_norm": 0.33957851032498043,
|
|
"learning_rate": 4.595251514724532e-06,
|
|
"loss": 0.9533,
|
|
"mean_token_accuracy": 0.754729500412941,
|
|
"num_tokens": 159286903.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 0.9537109375,
|
|
"epoch": 0.21868379689742362,
|
|
"grad_norm": 0.2653670127218828,
|
|
"learning_rate": 4.591728899535015e-06,
|
|
"loss": 0.9295,
|
|
"mean_token_accuracy": 0.7615975379943848,
|
|
"num_tokens": 160287992.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 0.91953125,
|
|
"epoch": 0.22005057062803252,
|
|
"grad_norm": 0.29614446799111116,
|
|
"learning_rate": 4.588206284345499e-06,
|
|
"loss": 0.9103,
|
|
"mean_token_accuracy": 0.7669180631637573,
|
|
"num_tokens": 161293325.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 1.01015625,
|
|
"epoch": 0.22141734435864144,
|
|
"grad_norm": 0.3274363014210417,
|
|
"learning_rate": 4.584683669155981e-06,
|
|
"loss": 1.004,
|
|
"mean_token_accuracy": 0.7480843275785446,
|
|
"num_tokens": 162313205.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 0.9798828125,
|
|
"epoch": 0.22278411808925033,
|
|
"grad_norm": 0.3179683798525132,
|
|
"learning_rate": 4.581161053966465e-06,
|
|
"loss": 0.9703,
|
|
"mean_token_accuracy": 0.7566818684339524,
|
|
"num_tokens": 163369684.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 1.015234375,
|
|
"epoch": 0.22415089181985923,
|
|
"grad_norm": 0.38088812333786903,
|
|
"learning_rate": 4.5776384387769485e-06,
|
|
"loss": 1.0166,
|
|
"mean_token_accuracy": 0.7461527228355408,
|
|
"num_tokens": 164375021.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 0.9455078125,
|
|
"epoch": 0.22551766555046812,
|
|
"grad_norm": 0.31973378911572853,
|
|
"learning_rate": 4.574115823587432e-06,
|
|
"loss": 0.9323,
|
|
"mean_token_accuracy": 0.7633433759212493,
|
|
"num_tokens": 165385723.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 0.9431640625,
|
|
"epoch": 0.22688443928107702,
|
|
"grad_norm": 0.3836364070922413,
|
|
"learning_rate": 4.570593208397915e-06,
|
|
"loss": 0.942,
|
|
"mean_token_accuracy": 0.7617446511983872,
|
|
"num_tokens": 166386777.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 0.962890625,
|
|
"epoch": 0.2282512130116859,
|
|
"grad_norm": 0.28230277293182665,
|
|
"learning_rate": 4.567070593208398e-06,
|
|
"loss": 0.9504,
|
|
"mean_token_accuracy": 0.7588118046522141,
|
|
"num_tokens": 167364188.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 0.952734375,
|
|
"epoch": 0.2296179867422948,
|
|
"grad_norm": 0.29925323739507664,
|
|
"learning_rate": 4.563547978018882e-06,
|
|
"loss": 0.9454,
|
|
"mean_token_accuracy": 0.7578962236642838,
|
|
"num_tokens": 168389909.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 0.9373046875,
|
|
"epoch": 0.2309847604729037,
|
|
"grad_norm": 0.32578913369569557,
|
|
"learning_rate": 4.560025362829365e-06,
|
|
"loss": 0.9292,
|
|
"mean_token_accuracy": 0.7616907387971878,
|
|
"num_tokens": 169381885.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 0.94921875,
|
|
"epoch": 0.23235153420351262,
|
|
"grad_norm": 0.2761050721594914,
|
|
"learning_rate": 4.556502747639848e-06,
|
|
"loss": 0.9469,
|
|
"mean_token_accuracy": 0.7592941969633102,
|
|
"num_tokens": 170399072.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 0.9646484375,
|
|
"epoch": 0.23371830793412152,
|
|
"grad_norm": 0.27154225849709596,
|
|
"learning_rate": 4.552980132450332e-06,
|
|
"loss": 0.9539,
|
|
"mean_token_accuracy": 0.7556976228952408,
|
|
"num_tokens": 171449606.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 0.9482421875,
|
|
"epoch": 0.2350850816647304,
|
|
"grad_norm": 0.2561935446067893,
|
|
"learning_rate": 4.549457517260814e-06,
|
|
"loss": 0.941,
|
|
"mean_token_accuracy": 0.761197954416275,
|
|
"num_tokens": 172478002.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 0.9333984375,
|
|
"epoch": 0.2364518553953393,
|
|
"grad_norm": 0.3056767044592072,
|
|
"learning_rate": 4.545934902071298e-06,
|
|
"loss": 0.9311,
|
|
"mean_token_accuracy": 0.7614845365285874,
|
|
"num_tokens": 173536624.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 0.986328125,
|
|
"epoch": 0.2378186291259482,
|
|
"grad_norm": 0.31027278155642884,
|
|
"learning_rate": 4.5424122868817814e-06,
|
|
"loss": 0.975,
|
|
"mean_token_accuracy": 0.7521440953016281,
|
|
"num_tokens": 174499171.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 0.9265625,
|
|
"epoch": 0.2391854028565571,
|
|
"grad_norm": 0.32827648253263775,
|
|
"learning_rate": 4.538889671692265e-06,
|
|
"loss": 0.9184,
|
|
"mean_token_accuracy": 0.7650352537631988,
|
|
"num_tokens": 175471599.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 0.938671875,
|
|
"epoch": 0.240552176587166,
|
|
"grad_norm": 0.310396659107655,
|
|
"learning_rate": 4.535367056502748e-06,
|
|
"loss": 0.9301,
|
|
"mean_token_accuracy": 0.7610342681407929,
|
|
"num_tokens": 176498065.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 0.9669921875,
|
|
"epoch": 0.24191895031777488,
|
|
"grad_norm": 0.3395247958982919,
|
|
"learning_rate": 4.531844441313231e-06,
|
|
"loss": 0.9544,
|
|
"mean_token_accuracy": 0.7576577603816986,
|
|
"num_tokens": 177499643.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 0.91875,
|
|
"epoch": 0.24328572404838378,
|
|
"grad_norm": 0.31207132316598857,
|
|
"learning_rate": 4.528321826123715e-06,
|
|
"loss": 0.9135,
|
|
"mean_token_accuracy": 0.7665604412555694,
|
|
"num_tokens": 178477704.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 0.9494140625,
|
|
"epoch": 0.2446524977789927,
|
|
"grad_norm": 0.26419996144469937,
|
|
"learning_rate": 4.524799210934198e-06,
|
|
"loss": 0.9427,
|
|
"mean_token_accuracy": 0.7618486493825912,
|
|
"num_tokens": 179484701.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 0.95546875,
|
|
"epoch": 0.2460192715096016,
|
|
"grad_norm": 0.29510346091384676,
|
|
"learning_rate": 4.521276595744681e-06,
|
|
"loss": 0.937,
|
|
"mean_token_accuracy": 0.7628296583890914,
|
|
"num_tokens": 180460840.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 0.96171875,
|
|
"epoch": 0.2473860452402105,
|
|
"grad_norm": 0.29015453899170235,
|
|
"learning_rate": 4.5177539805551646e-06,
|
|
"loss": 0.9635,
|
|
"mean_token_accuracy": 0.7570092082023621,
|
|
"num_tokens": 181453261.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 0.9396484375,
|
|
"epoch": 0.24875281897081938,
|
|
"grad_norm": 0.30916874680159206,
|
|
"learning_rate": 4.514231365365648e-06,
|
|
"loss": 0.9397,
|
|
"mean_token_accuracy": 0.7599998772144317,
|
|
"num_tokens": 182417078.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 0.9513671875,
|
|
"epoch": 0.2501195927014283,
|
|
"grad_norm": 0.37932523605185847,
|
|
"learning_rate": 4.510708750176131e-06,
|
|
"loss": 0.9303,
|
|
"mean_token_accuracy": 0.761031374335289,
|
|
"num_tokens": 183391495.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 0.933984375,
|
|
"epoch": 0.2514863664320372,
|
|
"grad_norm": 0.2769363331542965,
|
|
"learning_rate": 4.507186134986614e-06,
|
|
"loss": 0.9232,
|
|
"mean_token_accuracy": 0.7640673398971558,
|
|
"num_tokens": 184444423.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 0.92109375,
|
|
"epoch": 0.2528531401626461,
|
|
"grad_norm": 0.272686402550579,
|
|
"learning_rate": 4.503663519797098e-06,
|
|
"loss": 0.9188,
|
|
"mean_token_accuracy": 0.7638040632009506,
|
|
"num_tokens": 185464205.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 0.9421875,
|
|
"epoch": 0.254219913893255,
|
|
"grad_norm": 0.3757573193177148,
|
|
"learning_rate": 4.5001409046075814e-06,
|
|
"loss": 0.9374,
|
|
"mean_token_accuracy": 0.7614442974328994,
|
|
"num_tokens": 186450985.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 0.92421875,
|
|
"epoch": 0.2555866876238639,
|
|
"grad_norm": 0.2667688860681897,
|
|
"learning_rate": 4.496618289418064e-06,
|
|
"loss": 0.9069,
|
|
"mean_token_accuracy": 0.7671888709068299,
|
|
"num_tokens": 187494172.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 0.935546875,
|
|
"epoch": 0.2569534613544728,
|
|
"grad_norm": 0.3061960113204895,
|
|
"learning_rate": 4.493095674228548e-06,
|
|
"loss": 0.9256,
|
|
"mean_token_accuracy": 0.7635408282279968,
|
|
"num_tokens": 188466326.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 0.9685546875,
|
|
"epoch": 0.2583202350850817,
|
|
"grad_norm": 0.2726450401156821,
|
|
"learning_rate": 4.48957305903903e-06,
|
|
"loss": 0.9637,
|
|
"mean_token_accuracy": 0.7560075342655181,
|
|
"num_tokens": 189486495.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 0.9607421875,
|
|
"epoch": 0.25968700881569057,
|
|
"grad_norm": 0.35145021931972986,
|
|
"learning_rate": 4.486050443849515e-06,
|
|
"loss": 0.9546,
|
|
"mean_token_accuracy": 0.7585060834884644,
|
|
"num_tokens": 190471394.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 0.938671875,
|
|
"epoch": 0.26105378254629946,
|
|
"grad_norm": 0.28605167115292196,
|
|
"learning_rate": 4.4825278286599975e-06,
|
|
"loss": 0.9399,
|
|
"mean_token_accuracy": 0.761422461271286,
|
|
"num_tokens": 191482976.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 0.9111328125,
|
|
"epoch": 0.26242055627690836,
|
|
"grad_norm": 0.2993221372808091,
|
|
"learning_rate": 4.479005213470481e-06,
|
|
"loss": 0.9067,
|
|
"mean_token_accuracy": 0.7677863001823425,
|
|
"num_tokens": 192487003.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 0.957421875,
|
|
"epoch": 0.26378733000751725,
|
|
"grad_norm": 0.3645210702804304,
|
|
"learning_rate": 4.475482598280964e-06,
|
|
"loss": 0.9442,
|
|
"mean_token_accuracy": 0.7549897313117981,
|
|
"num_tokens": 193484373.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 0.925,
|
|
"epoch": 0.26515410373812615,
|
|
"grad_norm": 0.2960391550344042,
|
|
"learning_rate": 4.471959983091447e-06,
|
|
"loss": 0.9205,
|
|
"mean_token_accuracy": 0.7654795587062836,
|
|
"num_tokens": 194489877.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 0.953125,
|
|
"epoch": 0.26652087746873504,
|
|
"grad_norm": 0.30220047057914484,
|
|
"learning_rate": 4.468437367901931e-06,
|
|
"loss": 0.9556,
|
|
"mean_token_accuracy": 0.7583772629499436,
|
|
"num_tokens": 195497240.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 0.9724609375,
|
|
"epoch": 0.26788765119934393,
|
|
"grad_norm": 0.3306653340056146,
|
|
"learning_rate": 4.464914752712414e-06,
|
|
"loss": 0.9674,
|
|
"mean_token_accuracy": 0.7543736875057221,
|
|
"num_tokens": 196488733.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 0.9025390625,
|
|
"epoch": 0.26925442492995283,
|
|
"grad_norm": 0.30778576486069625,
|
|
"learning_rate": 4.461392137522897e-06,
|
|
"loss": 0.8983,
|
|
"mean_token_accuracy": 0.7678695738315582,
|
|
"num_tokens": 197524439.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 0.9009765625,
|
|
"epoch": 0.2706211986605617,
|
|
"grad_norm": 0.2733907539535985,
|
|
"learning_rate": 4.457869522333381e-06,
|
|
"loss": 0.889,
|
|
"mean_token_accuracy": 0.7703852742910385,
|
|
"num_tokens": 198500937.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 0.9328125,
|
|
"epoch": 0.2719879723911706,
|
|
"grad_norm": 0.3122340692559591,
|
|
"learning_rate": 4.454346907143864e-06,
|
|
"loss": 0.93,
|
|
"mean_token_accuracy": 0.7640798091888428,
|
|
"num_tokens": 199495588.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 0.9728515625,
|
|
"epoch": 0.2733547461217795,
|
|
"grad_norm": 0.33663316317227865,
|
|
"learning_rate": 4.450824291954347e-06,
|
|
"loss": 0.9666,
|
|
"mean_token_accuracy": 0.7563201248645782,
|
|
"num_tokens": 200506609.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 0.909375,
|
|
"epoch": 0.27472151985238846,
|
|
"grad_norm": 0.24502305024188809,
|
|
"learning_rate": 4.44730167676483e-06,
|
|
"loss": 0.9074,
|
|
"mean_token_accuracy": 0.7684537768363953,
|
|
"num_tokens": 201514736.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 0.9427734375,
|
|
"epoch": 0.27608829358299736,
|
|
"grad_norm": 0.3089393628536112,
|
|
"learning_rate": 4.443779061575314e-06,
|
|
"loss": 0.9311,
|
|
"mean_token_accuracy": 0.7607100427150726,
|
|
"num_tokens": 202528916.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 0.9017578125,
|
|
"epoch": 0.27745506731360625,
|
|
"grad_norm": 0.3428980195144158,
|
|
"learning_rate": 4.4402564463857975e-06,
|
|
"loss": 0.8985,
|
|
"mean_token_accuracy": 0.7674260437488556,
|
|
"num_tokens": 203493505.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 0.9412109375,
|
|
"epoch": 0.27882184104421515,
|
|
"grad_norm": 0.30288832527215853,
|
|
"learning_rate": 4.43673383119628e-06,
|
|
"loss": 0.9488,
|
|
"mean_token_accuracy": 0.7589116185903549,
|
|
"num_tokens": 204472127.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 0.937890625,
|
|
"epoch": 0.28018861477482404,
|
|
"grad_norm": 0.3428391498807461,
|
|
"learning_rate": 4.433211216006764e-06,
|
|
"loss": 0.9305,
|
|
"mean_token_accuracy": 0.7606789976358413,
|
|
"num_tokens": 205444690.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 0.9203125,
|
|
"epoch": 0.28155538850543294,
|
|
"grad_norm": 0.35777072170146934,
|
|
"learning_rate": 4.4296886008172465e-06,
|
|
"loss": 0.9292,
|
|
"mean_token_accuracy": 0.7623257040977478,
|
|
"num_tokens": 206433420.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 0.9353515625,
|
|
"epoch": 0.28292216223604183,
|
|
"grad_norm": 0.39099832131051093,
|
|
"learning_rate": 4.426165985627731e-06,
|
|
"loss": 0.9317,
|
|
"mean_token_accuracy": 0.7633480459451676,
|
|
"num_tokens": 207474650.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 0.9068359375,
|
|
"epoch": 0.2842889359666507,
|
|
"grad_norm": 0.31994985810613813,
|
|
"learning_rate": 4.4226433704382136e-06,
|
|
"loss": 0.8987,
|
|
"mean_token_accuracy": 0.7690522521734238,
|
|
"num_tokens": 208449182.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 0.919140625,
|
|
"epoch": 0.2856557096972596,
|
|
"grad_norm": 0.2813270828176761,
|
|
"learning_rate": 4.419120755248697e-06,
|
|
"loss": 0.913,
|
|
"mean_token_accuracy": 0.7653984934091568,
|
|
"num_tokens": 209450690.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 0.994140625,
|
|
"epoch": 0.2870224834278685,
|
|
"grad_norm": 0.26262753306770587,
|
|
"learning_rate": 4.41559814005918e-06,
|
|
"loss": 0.9968,
|
|
"mean_token_accuracy": 0.7493891686201095,
|
|
"num_tokens": 210469459.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 0.9265625,
|
|
"epoch": 0.2883892571584774,
|
|
"grad_norm": 0.2811734035372985,
|
|
"learning_rate": 4.412075524869663e-06,
|
|
"loss": 0.921,
|
|
"mean_token_accuracy": 0.7654306083917618,
|
|
"num_tokens": 211422588.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 0.953515625,
|
|
"epoch": 0.2897560308890863,
|
|
"grad_norm": 0.36834915893380393,
|
|
"learning_rate": 4.408552909680147e-06,
|
|
"loss": 0.9488,
|
|
"mean_token_accuracy": 0.7591090857982635,
|
|
"num_tokens": 212401656.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 0.93515625,
|
|
"epoch": 0.2911228046196952,
|
|
"grad_norm": 0.3158012087797059,
|
|
"learning_rate": 4.4050302944906304e-06,
|
|
"loss": 0.9345,
|
|
"mean_token_accuracy": 0.76079081594944,
|
|
"num_tokens": 213418240.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 0.9251953125,
|
|
"epoch": 0.2924895783503041,
|
|
"grad_norm": 0.3023084222618906,
|
|
"learning_rate": 4.401507679301113e-06,
|
|
"loss": 0.9162,
|
|
"mean_token_accuracy": 0.7640172690153122,
|
|
"num_tokens": 214387238.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 0.9943359375,
|
|
"epoch": 0.293856352080913,
|
|
"grad_norm": 0.2985369767866622,
|
|
"learning_rate": 4.397985064111597e-06,
|
|
"loss": 0.9972,
|
|
"mean_token_accuracy": 0.7492607742547989,
|
|
"num_tokens": 215412774.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 0.947265625,
|
|
"epoch": 0.2952231258115219,
|
|
"grad_norm": 0.2619152275637926,
|
|
"learning_rate": 4.39446244892208e-06,
|
|
"loss": 0.9366,
|
|
"mean_token_accuracy": 0.7617646634578705,
|
|
"num_tokens": 216404116.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 0.9326171875,
|
|
"epoch": 0.2965898995421308,
|
|
"grad_norm": 0.3153089800032464,
|
|
"learning_rate": 4.390939833732563e-06,
|
|
"loss": 0.9293,
|
|
"mean_token_accuracy": 0.7611831247806549,
|
|
"num_tokens": 217329569.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 0.933984375,
|
|
"epoch": 0.2979566732727397,
|
|
"grad_norm": 0.30664671076704003,
|
|
"learning_rate": 4.387417218543047e-06,
|
|
"loss": 0.9312,
|
|
"mean_token_accuracy": 0.7612697809934617,
|
|
"num_tokens": 218348233.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 0.9078125,
|
|
"epoch": 0.2993234470033486,
|
|
"grad_norm": 0.27792193290941086,
|
|
"learning_rate": 4.38389460335353e-06,
|
|
"loss": 0.89,
|
|
"mean_token_accuracy": 0.7713354051113128,
|
|
"num_tokens": 219347779.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 0.93984375,
|
|
"epoch": 0.3006902207339575,
|
|
"grad_norm": 0.25345921071240696,
|
|
"learning_rate": 4.380371988164014e-06,
|
|
"loss": 0.9343,
|
|
"mean_token_accuracy": 0.7596987873315811,
|
|
"num_tokens": 220403938.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 0.926953125,
|
|
"epoch": 0.3020569944645664,
|
|
"grad_norm": 0.2595387497599974,
|
|
"learning_rate": 4.376849372974496e-06,
|
|
"loss": 0.9219,
|
|
"mean_token_accuracy": 0.7655311495065689,
|
|
"num_tokens": 221435776.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 0.9078125,
|
|
"epoch": 0.3034237681951753,
|
|
"grad_norm": 0.2862737412846619,
|
|
"learning_rate": 4.37332675778498e-06,
|
|
"loss": 0.9055,
|
|
"mean_token_accuracy": 0.766112095117569,
|
|
"num_tokens": 222450369.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 0.930078125,
|
|
"epoch": 0.3047905419257842,
|
|
"grad_norm": 0.30469957442568424,
|
|
"learning_rate": 4.369804142595463e-06,
|
|
"loss": 0.9341,
|
|
"mean_token_accuracy": 0.7627186626195908,
|
|
"num_tokens": 223430185.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 0.9447265625,
|
|
"epoch": 0.3061573156563931,
|
|
"grad_norm": 0.3329986196200335,
|
|
"learning_rate": 4.366281527405947e-06,
|
|
"loss": 0.9419,
|
|
"mean_token_accuracy": 0.7587600290775299,
|
|
"num_tokens": 224476704.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 0.8986328125,
|
|
"epoch": 0.307524089387002,
|
|
"grad_norm": 0.2942175585535601,
|
|
"learning_rate": 4.36275891221643e-06,
|
|
"loss": 0.89,
|
|
"mean_token_accuracy": 0.7681857466697692,
|
|
"num_tokens": 225491846.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 0.8955078125,
|
|
"epoch": 0.3088908631176109,
|
|
"grad_norm": 0.5528492114628702,
|
|
"learning_rate": 4.359236297026913e-06,
|
|
"loss": 0.8946,
|
|
"mean_token_accuracy": 0.7701717555522919,
|
|
"num_tokens": 226484127.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 0.882421875,
|
|
"epoch": 0.3102576368482198,
|
|
"grad_norm": 0.3623594792560495,
|
|
"learning_rate": 4.355713681837396e-06,
|
|
"loss": 0.8822,
|
|
"mean_token_accuracy": 0.7718490809202194,
|
|
"num_tokens": 227474908.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 0.91015625,
|
|
"epoch": 0.31162441057882867,
|
|
"grad_norm": 0.28902063920642923,
|
|
"learning_rate": 4.352191066647879e-06,
|
|
"loss": 0.9124,
|
|
"mean_token_accuracy": 0.7665034621953964,
|
|
"num_tokens": 228463177.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 0.9255859375,
|
|
"epoch": 0.31299118430943756,
|
|
"grad_norm": 0.3057569436106022,
|
|
"learning_rate": 4.348668451458363e-06,
|
|
"loss": 0.9247,
|
|
"mean_token_accuracy": 0.7635338574647903,
|
|
"num_tokens": 229469712.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 0.940234375,
|
|
"epoch": 0.31435795804004646,
|
|
"grad_norm": 0.29083951433840005,
|
|
"learning_rate": 4.3451458362688465e-06,
|
|
"loss": 0.9355,
|
|
"mean_token_accuracy": 0.7611991822719574,
|
|
"num_tokens": 230474108.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 0.948046875,
|
|
"epoch": 0.31572473177065535,
|
|
"grad_norm": 0.3285007953140542,
|
|
"learning_rate": 4.34162322107933e-06,
|
|
"loss": 0.933,
|
|
"mean_token_accuracy": 0.7632004916667938,
|
|
"num_tokens": 231443208.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 0.934375,
|
|
"epoch": 0.31709150550126425,
|
|
"grad_norm": 0.26799766761092875,
|
|
"learning_rate": 4.338100605889813e-06,
|
|
"loss": 0.9258,
|
|
"mean_token_accuracy": 0.7627386420965194,
|
|
"num_tokens": 232407706.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 0.8953125,
|
|
"epoch": 0.31845827923187314,
|
|
"grad_norm": 0.30334509191488035,
|
|
"learning_rate": 4.334577990700296e-06,
|
|
"loss": 0.8794,
|
|
"mean_token_accuracy": 0.7740000426769257,
|
|
"num_tokens": 233367033.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 0.9630859375,
|
|
"epoch": 0.31982505296248204,
|
|
"grad_norm": 0.4385872568435246,
|
|
"learning_rate": 4.33105537551078e-06,
|
|
"loss": 0.9595,
|
|
"mean_token_accuracy": 0.7544190734624863,
|
|
"num_tokens": 234337713.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 0.9009765625,
|
|
"epoch": 0.32119182669309093,
|
|
"grad_norm": 0.3156990261245654,
|
|
"learning_rate": 4.327532760321263e-06,
|
|
"loss": 0.8839,
|
|
"mean_token_accuracy": 0.7722537904977799,
|
|
"num_tokens": 235293773.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 0.926171875,
|
|
"epoch": 0.3225586004236999,
|
|
"grad_norm": 0.2943722786989206,
|
|
"learning_rate": 4.324010145131746e-06,
|
|
"loss": 0.9173,
|
|
"mean_token_accuracy": 0.7627825796604156,
|
|
"num_tokens": 236278533.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 0.93125,
|
|
"epoch": 0.3239253741543088,
|
|
"grad_norm": 0.27444957312602447,
|
|
"learning_rate": 4.32048752994223e-06,
|
|
"loss": 0.9292,
|
|
"mean_token_accuracy": 0.7617819726467132,
|
|
"num_tokens": 237320727.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 0.9462890625,
|
|
"epoch": 0.32529214788491767,
|
|
"grad_norm": 0.3286094468389035,
|
|
"learning_rate": 4.316964914752712e-06,
|
|
"loss": 0.9412,
|
|
"mean_token_accuracy": 0.7603401362895965,
|
|
"num_tokens": 238313682.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 0.9470703125,
|
|
"epoch": 0.32665892161552657,
|
|
"grad_norm": 0.3081628694094939,
|
|
"learning_rate": 4.313442299563196e-06,
|
|
"loss": 0.9436,
|
|
"mean_token_accuracy": 0.7608725935220718,
|
|
"num_tokens": 239272562.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 0.92265625,
|
|
"epoch": 0.32802569534613546,
|
|
"grad_norm": 0.31692357672255234,
|
|
"learning_rate": 4.3099196843736794e-06,
|
|
"loss": 0.9209,
|
|
"mean_token_accuracy": 0.7637060642242431,
|
|
"num_tokens": 240247837.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 0.93125,
|
|
"epoch": 0.32939246907674435,
|
|
"grad_norm": 0.2745074925828302,
|
|
"learning_rate": 4.306397069184163e-06,
|
|
"loss": 0.9192,
|
|
"mean_token_accuracy": 0.7622980296611785,
|
|
"num_tokens": 241244375.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 0.933984375,
|
|
"epoch": 0.33075924280735325,
|
|
"grad_norm": 0.33580746082056157,
|
|
"learning_rate": 4.302874453994646e-06,
|
|
"loss": 0.9277,
|
|
"mean_token_accuracy": 0.761882197856903,
|
|
"num_tokens": 242235921.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 0.914453125,
|
|
"epoch": 0.33212601653796214,
|
|
"grad_norm": 0.29832514291335344,
|
|
"learning_rate": 4.299351838805129e-06,
|
|
"loss": 0.9123,
|
|
"mean_token_accuracy": 0.766022315621376,
|
|
"num_tokens": 243226532.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 0.9482421875,
|
|
"epoch": 0.33349279026857104,
|
|
"grad_norm": 0.2783838246994064,
|
|
"learning_rate": 4.295829223615613e-06,
|
|
"loss": 0.9447,
|
|
"mean_token_accuracy": 0.7601381093263626,
|
|
"num_tokens": 244230951.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 0.9416015625,
|
|
"epoch": 0.33485956399917993,
|
|
"grad_norm": 0.3789637666056786,
|
|
"learning_rate": 4.292306608426096e-06,
|
|
"loss": 0.9311,
|
|
"mean_token_accuracy": 0.7623953968286514,
|
|
"num_tokens": 245216118.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 0.897265625,
|
|
"epoch": 0.3362263377297888,
|
|
"grad_norm": 0.2832146283406114,
|
|
"learning_rate": 4.288783993236579e-06,
|
|
"loss": 0.8898,
|
|
"mean_token_accuracy": 0.7709452748298645,
|
|
"num_tokens": 246233429.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 0.951953125,
|
|
"epoch": 0.3375931114603977,
|
|
"grad_norm": 0.31823535679679504,
|
|
"learning_rate": 4.285261378047063e-06,
|
|
"loss": 0.9435,
|
|
"mean_token_accuracy": 0.7603794276714325,
|
|
"num_tokens": 247241545.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 0.926171875,
|
|
"epoch": 0.3389598851910066,
|
|
"grad_norm": 0.2747534617180832,
|
|
"learning_rate": 4.281738762857546e-06,
|
|
"loss": 0.9245,
|
|
"mean_token_accuracy": 0.7616445660591126,
|
|
"num_tokens": 248252703.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 0.9701171875,
|
|
"epoch": 0.3403266589216155,
|
|
"grad_norm": 0.29156403290226923,
|
|
"learning_rate": 4.278216147668029e-06,
|
|
"loss": 0.9614,
|
|
"mean_token_accuracy": 0.755089259147644,
|
|
"num_tokens": 249259960.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 0.9373046875,
|
|
"epoch": 0.3416934326522244,
|
|
"grad_norm": 0.301281259792644,
|
|
"learning_rate": 4.274693532478512e-06,
|
|
"loss": 0.9346,
|
|
"mean_token_accuracy": 0.7594863951206208,
|
|
"num_tokens": 250256940.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 0.8943359375,
|
|
"epoch": 0.3430602063828333,
|
|
"grad_norm": 0.25762504512852,
|
|
"learning_rate": 4.271170917288996e-06,
|
|
"loss": 0.8885,
|
|
"mean_token_accuracy": 0.7710637956857681,
|
|
"num_tokens": 251256658.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 0.919140625,
|
|
"epoch": 0.3444269801134422,
|
|
"grad_norm": 0.3650536172586655,
|
|
"learning_rate": 4.2676483020994795e-06,
|
|
"loss": 0.9148,
|
|
"mean_token_accuracy": 0.7654563993215561,
|
|
"num_tokens": 252264359.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 0.933203125,
|
|
"epoch": 0.34579375384405114,
|
|
"grad_norm": 0.28166744426292845,
|
|
"learning_rate": 4.264125686909962e-06,
|
|
"loss": 0.9376,
|
|
"mean_token_accuracy": 0.7620678514242172,
|
|
"num_tokens": 253309857.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 0.9376953125,
|
|
"epoch": 0.34716052757466004,
|
|
"grad_norm": 0.2809892007554737,
|
|
"learning_rate": 4.260603071720446e-06,
|
|
"loss": 0.9395,
|
|
"mean_token_accuracy": 0.7618267595767975,
|
|
"num_tokens": 254290565.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 0.9625,
|
|
"epoch": 0.34852730130526893,
|
|
"grad_norm": 0.37655128039296537,
|
|
"learning_rate": 4.257080456530928e-06,
|
|
"loss": 0.9587,
|
|
"mean_token_accuracy": 0.7552110880613327,
|
|
"num_tokens": 255281492.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 0.8908203125,
|
|
"epoch": 0.3498940750358778,
|
|
"grad_norm": 0.29406521599617585,
|
|
"learning_rate": 4.253557841341413e-06,
|
|
"loss": 0.893,
|
|
"mean_token_accuracy": 0.770569309592247,
|
|
"num_tokens": 256296392.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 0.9078125,
|
|
"epoch": 0.3512608487664867,
|
|
"grad_norm": 0.297128545558957,
|
|
"learning_rate": 4.2500352261518955e-06,
|
|
"loss": 0.9066,
|
|
"mean_token_accuracy": 0.7656499564647674,
|
|
"num_tokens": 257302063.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 0.8744140625,
|
|
"epoch": 0.3526276224970956,
|
|
"grad_norm": 0.28049796929227744,
|
|
"learning_rate": 4.246512610962379e-06,
|
|
"loss": 0.8714,
|
|
"mean_token_accuracy": 0.7744253695011138,
|
|
"num_tokens": 258304554.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 0.94140625,
|
|
"epoch": 0.3539943962277045,
|
|
"grad_norm": 0.3062655798859269,
|
|
"learning_rate": 4.242989995772862e-06,
|
|
"loss": 0.9398,
|
|
"mean_token_accuracy": 0.7610694408416748,
|
|
"num_tokens": 259259663.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 0.93515625,
|
|
"epoch": 0.3553611699583134,
|
|
"grad_norm": 0.2992611254957664,
|
|
"learning_rate": 4.239467380583345e-06,
|
|
"loss": 0.934,
|
|
"mean_token_accuracy": 0.7617115825414658,
|
|
"num_tokens": 260318431.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 0.9021484375,
|
|
"epoch": 0.3567279436889223,
|
|
"grad_norm": 0.30534660842961575,
|
|
"learning_rate": 4.235944765393829e-06,
|
|
"loss": 0.891,
|
|
"mean_token_accuracy": 0.7696558982133865,
|
|
"num_tokens": 261314415.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 0.926171875,
|
|
"epoch": 0.3580947174195312,
|
|
"grad_norm": 0.38769784030267906,
|
|
"learning_rate": 4.232422150204312e-06,
|
|
"loss": 0.9281,
|
|
"mean_token_accuracy": 0.7606382101774216,
|
|
"num_tokens": 262280825.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 0.9720703125,
|
|
"epoch": 0.3594614911501401,
|
|
"grad_norm": 0.28731278654038556,
|
|
"learning_rate": 4.228899535014795e-06,
|
|
"loss": 0.9673,
|
|
"mean_token_accuracy": 0.7543341994285584,
|
|
"num_tokens": 263248393.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 0.933203125,
|
|
"epoch": 0.360828264880749,
|
|
"grad_norm": 0.2875256233367535,
|
|
"learning_rate": 4.225376919825279e-06,
|
|
"loss": 0.9356,
|
|
"mean_token_accuracy": 0.7605664879083633,
|
|
"num_tokens": 264237945.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 0.917578125,
|
|
"epoch": 0.3621950386113579,
|
|
"grad_norm": 0.2677328704492588,
|
|
"learning_rate": 4.221854304635762e-06,
|
|
"loss": 0.9073,
|
|
"mean_token_accuracy": 0.7652874290943146,
|
|
"num_tokens": 265242177.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 0.9625,
|
|
"epoch": 0.3635618123419668,
|
|
"grad_norm": 0.28963435789769504,
|
|
"learning_rate": 4.218331689446245e-06,
|
|
"loss": 0.9645,
|
|
"mean_token_accuracy": 0.7548456132411957,
|
|
"num_tokens": 266261360.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 0.8986328125,
|
|
"epoch": 0.36492858607257567,
|
|
"grad_norm": 0.30060490912654064,
|
|
"learning_rate": 4.2148090742567284e-06,
|
|
"loss": 0.8842,
|
|
"mean_token_accuracy": 0.7725124448537827,
|
|
"num_tokens": 267257220.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 0.91796875,
|
|
"epoch": 0.36629535980318456,
|
|
"grad_norm": 0.4955740099599833,
|
|
"learning_rate": 4.211286459067212e-06,
|
|
"loss": 0.9292,
|
|
"mean_token_accuracy": 0.7605182558298111,
|
|
"num_tokens": 268317830.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 0.891796875,
|
|
"epoch": 0.36766213353379346,
|
|
"grad_norm": 0.31898235053155405,
|
|
"learning_rate": 4.2077638438776955e-06,
|
|
"loss": 0.8746,
|
|
"mean_token_accuracy": 0.7731282651424408,
|
|
"num_tokens": 269283864.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 0.8953125,
|
|
"epoch": 0.36902890726440235,
|
|
"grad_norm": 0.28639370444109286,
|
|
"learning_rate": 4.204241228688178e-06,
|
|
"loss": 0.8954,
|
|
"mean_token_accuracy": 0.7700981378555298,
|
|
"num_tokens": 270373918.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 0.90625,
|
|
"epoch": 0.3703956809950113,
|
|
"grad_norm": 0.3021704258000859,
|
|
"learning_rate": 4.200718613498662e-06,
|
|
"loss": 0.9052,
|
|
"mean_token_accuracy": 0.7674299240112304,
|
|
"num_tokens": 271394623.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 0.9185546875,
|
|
"epoch": 0.3717624547256202,
|
|
"grad_norm": 0.28403044555416956,
|
|
"learning_rate": 4.1971959983091445e-06,
|
|
"loss": 0.9061,
|
|
"mean_token_accuracy": 0.7673598289489746,
|
|
"num_tokens": 272366138.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 0.8900390625,
|
|
"epoch": 0.3731292284562291,
|
|
"grad_norm": 0.28374617472041497,
|
|
"learning_rate": 4.193673383119629e-06,
|
|
"loss": 0.8757,
|
|
"mean_token_accuracy": 0.7730923235416413,
|
|
"num_tokens": 273393329.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 0.9658203125,
|
|
"epoch": 0.374496002186838,
|
|
"grad_norm": 0.2869234185066581,
|
|
"learning_rate": 4.190150767930112e-06,
|
|
"loss": 0.9553,
|
|
"mean_token_accuracy": 0.7574404925107956,
|
|
"num_tokens": 274456082.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 0.951171875,
|
|
"epoch": 0.3758627759174469,
|
|
"grad_norm": 0.3538090426090921,
|
|
"learning_rate": 4.186628152740595e-06,
|
|
"loss": 0.9477,
|
|
"mean_token_accuracy": 0.7590755224227905,
|
|
"num_tokens": 275482962.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 0.91640625,
|
|
"epoch": 0.3772295496480558,
|
|
"grad_norm": 0.29968986473036036,
|
|
"learning_rate": 4.183105537551078e-06,
|
|
"loss": 0.903,
|
|
"mean_token_accuracy": 0.7671309024095535,
|
|
"num_tokens": 276499414.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 0.9484375,
|
|
"epoch": 0.37859632337866467,
|
|
"grad_norm": 0.306155501450588,
|
|
"learning_rate": 4.179582922361561e-06,
|
|
"loss": 0.9501,
|
|
"mean_token_accuracy": 0.7561418175697326,
|
|
"num_tokens": 277471208.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 0.939453125,
|
|
"epoch": 0.37996309710927356,
|
|
"grad_norm": 0.28713424436718427,
|
|
"learning_rate": 4.176060307172045e-06,
|
|
"loss": 0.9324,
|
|
"mean_token_accuracy": 0.7604744791984558,
|
|
"num_tokens": 278492364.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 0.9189453125,
|
|
"epoch": 0.38132987083988246,
|
|
"grad_norm": 0.2956466799896635,
|
|
"learning_rate": 4.1725376919825285e-06,
|
|
"loss": 0.9091,
|
|
"mean_token_accuracy": 0.7669297009706497,
|
|
"num_tokens": 279461129.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 0.93125,
|
|
"epoch": 0.38269664457049135,
|
|
"grad_norm": 0.27359949913382975,
|
|
"learning_rate": 4.169015076793011e-06,
|
|
"loss": 0.9298,
|
|
"mean_token_accuracy": 0.7624536782503128,
|
|
"num_tokens": 280432345.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 0.902734375,
|
|
"epoch": 0.38406341830110025,
|
|
"grad_norm": 0.29160565747243483,
|
|
"learning_rate": 4.165492461603495e-06,
|
|
"loss": 0.8964,
|
|
"mean_token_accuracy": 0.7682175546884537,
|
|
"num_tokens": 281431868.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 0.911328125,
|
|
"epoch": 0.38543019203170914,
|
|
"grad_norm": 0.3223803267948713,
|
|
"learning_rate": 4.161969846413978e-06,
|
|
"loss": 0.9088,
|
|
"mean_token_accuracy": 0.7671536356210709,
|
|
"num_tokens": 282396063.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 0.864453125,
|
|
"epoch": 0.38679696576231803,
|
|
"grad_norm": 0.2865326970527864,
|
|
"learning_rate": 4.158447231224461e-06,
|
|
"loss": 0.8635,
|
|
"mean_token_accuracy": 0.775470769405365,
|
|
"num_tokens": 283462179.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 0.921484375,
|
|
"epoch": 0.38816373949292693,
|
|
"grad_norm": 0.2653300383086452,
|
|
"learning_rate": 4.1549246160349445e-06,
|
|
"loss": 0.9131,
|
|
"mean_token_accuracy": 0.7646975249052048,
|
|
"num_tokens": 284512317.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 0.909765625,
|
|
"epoch": 0.3895305132235358,
|
|
"grad_norm": 0.2779423641941073,
|
|
"learning_rate": 4.151402000845428e-06,
|
|
"loss": 0.9012,
|
|
"mean_token_accuracy": 0.7653542637825013,
|
|
"num_tokens": 285546366.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 0.89609375,
|
|
"epoch": 0.3908972869541447,
|
|
"grad_norm": 0.26371166693848863,
|
|
"learning_rate": 4.147879385655912e-06,
|
|
"loss": 0.8853,
|
|
"mean_token_accuracy": 0.7725135117769242,
|
|
"num_tokens": 286536086.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 0.90390625,
|
|
"epoch": 0.3922640606847536,
|
|
"grad_norm": 0.36427640099872605,
|
|
"learning_rate": 4.144356770466394e-06,
|
|
"loss": 0.8932,
|
|
"mean_token_accuracy": 0.7669319033622741,
|
|
"num_tokens": 287506225.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 0.984765625,
|
|
"epoch": 0.39363083441536256,
|
|
"grad_norm": 0.3060313402129116,
|
|
"learning_rate": 4.140834155276878e-06,
|
|
"loss": 0.978,
|
|
"mean_token_accuracy": 0.7523733794689178,
|
|
"num_tokens": 288517766.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 0.9439453125,
|
|
"epoch": 0.39499760814597146,
|
|
"grad_norm": 0.31245376249791473,
|
|
"learning_rate": 4.137311540087361e-06,
|
|
"loss": 0.942,
|
|
"mean_token_accuracy": 0.7609674423933029,
|
|
"num_tokens": 289575438.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 0.9509765625,
|
|
"epoch": 0.39636438187658035,
|
|
"grad_norm": 0.3202631884402749,
|
|
"learning_rate": 4.133788924897845e-06,
|
|
"loss": 0.9391,
|
|
"mean_token_accuracy": 0.7609072357416153,
|
|
"num_tokens": 290579778.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 0.883984375,
|
|
"epoch": 0.39773115560718925,
|
|
"grad_norm": 0.28615243424517095,
|
|
"learning_rate": 4.130266309708328e-06,
|
|
"loss": 0.8817,
|
|
"mean_token_accuracy": 0.7752098679542542,
|
|
"num_tokens": 291573891.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 0.9447265625,
|
|
"epoch": 0.39909792933779814,
|
|
"grad_norm": 0.28764932430911655,
|
|
"learning_rate": 4.126743694518811e-06,
|
|
"loss": 0.9344,
|
|
"mean_token_accuracy": 0.7606434047222137,
|
|
"num_tokens": 292603953.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 0.913671875,
|
|
"epoch": 0.40046470306840704,
|
|
"grad_norm": 0.3021445783371929,
|
|
"learning_rate": 4.123221079329294e-06,
|
|
"loss": 0.9033,
|
|
"mean_token_accuracy": 0.7675307899713516,
|
|
"num_tokens": 293541021.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 0.9306640625,
|
|
"epoch": 0.40183147679901593,
|
|
"grad_norm": 0.3038759345912456,
|
|
"learning_rate": 4.1196984641397774e-06,
|
|
"loss": 0.9296,
|
|
"mean_token_accuracy": 0.7612747132778168,
|
|
"num_tokens": 294528087.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 0.8943359375,
|
|
"epoch": 0.4031982505296248,
|
|
"grad_norm": 0.3021180135668986,
|
|
"learning_rate": 4.116175848950261e-06,
|
|
"loss": 0.8887,
|
|
"mean_token_accuracy": 0.7717967629432678,
|
|
"num_tokens": 295510553.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 0.895703125,
|
|
"epoch": 0.4045650242602337,
|
|
"grad_norm": 0.3098685140093233,
|
|
"learning_rate": 4.1126532337607445e-06,
|
|
"loss": 0.8846,
|
|
"mean_token_accuracy": 0.7714555919170379,
|
|
"num_tokens": 296428358.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 0.9443359375,
|
|
"epoch": 0.4059317979908426,
|
|
"grad_norm": 0.3417766044799608,
|
|
"learning_rate": 4.109130618571227e-06,
|
|
"loss": 0.9381,
|
|
"mean_token_accuracy": 0.7620276689529419,
|
|
"num_tokens": 297444594.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 0.925,
|
|
"epoch": 0.4072985717214515,
|
|
"grad_norm": 0.40888704763853734,
|
|
"learning_rate": 4.105608003381711e-06,
|
|
"loss": 0.9143,
|
|
"mean_token_accuracy": 0.7631752490997314,
|
|
"num_tokens": 298427763.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 0.8853515625,
|
|
"epoch": 0.4086653454520604,
|
|
"grad_norm": 0.3090178834907218,
|
|
"learning_rate": 4.102085388192194e-06,
|
|
"loss": 0.8788,
|
|
"mean_token_accuracy": 0.7729545652866363,
|
|
"num_tokens": 299412943.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 0.924609375,
|
|
"epoch": 0.4100321191826693,
|
|
"grad_norm": 0.33952359728557235,
|
|
"learning_rate": 4.098562773002678e-06,
|
|
"loss": 0.9223,
|
|
"mean_token_accuracy": 0.7615919768810272,
|
|
"num_tokens": 300389659.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"entropy": 0.8890625,
|
|
"epoch": 0.4113988929132782,
|
|
"grad_norm": 0.28720692507384105,
|
|
"learning_rate": 4.095040157813161e-06,
|
|
"loss": 0.8826,
|
|
"mean_token_accuracy": 0.7720259130001068,
|
|
"num_tokens": 301433498.0,
|
|
"step": 3010
|
|
},
|
|
{
|
|
"entropy": 0.933203125,
|
|
"epoch": 0.4127656666438871,
|
|
"grad_norm": 0.3692135290187806,
|
|
"learning_rate": 4.091517542623644e-06,
|
|
"loss": 0.9427,
|
|
"mean_token_accuracy": 0.7586299151182174,
|
|
"num_tokens": 302413081.0,
|
|
"step": 3020
|
|
},
|
|
{
|
|
"entropy": 0.9697265625,
|
|
"epoch": 0.414132440374496,
|
|
"grad_norm": 0.2936159795184501,
|
|
"learning_rate": 4.087994927434128e-06,
|
|
"loss": 0.9733,
|
|
"mean_token_accuracy": 0.7533062040805817,
|
|
"num_tokens": 303402573.0,
|
|
"step": 3030
|
|
},
|
|
{
|
|
"entropy": 0.935546875,
|
|
"epoch": 0.4154992141051049,
|
|
"grad_norm": 0.3210223632701201,
|
|
"learning_rate": 4.08447231224461e-06,
|
|
"loss": 0.9313,
|
|
"mean_token_accuracy": 0.7607459932565689,
|
|
"num_tokens": 304370910.0,
|
|
"step": 3040
|
|
},
|
|
{
|
|
"entropy": 0.8724609375,
|
|
"epoch": 0.4168659878357138,
|
|
"grad_norm": 0.3082527055872067,
|
|
"learning_rate": 4.080949697055094e-06,
|
|
"loss": 0.8675,
|
|
"mean_token_accuracy": 0.7739057034254074,
|
|
"num_tokens": 305334775.0,
|
|
"step": 3050
|
|
},
|
|
{
|
|
"entropy": 0.9025390625,
|
|
"epoch": 0.4182327615663227,
|
|
"grad_norm": 0.2912526359869142,
|
|
"learning_rate": 4.0774270818655775e-06,
|
|
"loss": 0.8973,
|
|
"mean_token_accuracy": 0.769287970662117,
|
|
"num_tokens": 306353872.0,
|
|
"step": 3060
|
|
},
|
|
{
|
|
"entropy": 0.9541015625,
|
|
"epoch": 0.4195995352969316,
|
|
"grad_norm": 0.3124195318058036,
|
|
"learning_rate": 4.073904466676061e-06,
|
|
"loss": 0.9492,
|
|
"mean_token_accuracy": 0.7570004671812057,
|
|
"num_tokens": 307331806.0,
|
|
"step": 3070
|
|
},
|
|
{
|
|
"entropy": 0.9330078125,
|
|
"epoch": 0.4209663090275405,
|
|
"grad_norm": 0.3177999189422952,
|
|
"learning_rate": 4.070381851486544e-06,
|
|
"loss": 0.9331,
|
|
"mean_token_accuracy": 0.7624543905258179,
|
|
"num_tokens": 308368576.0,
|
|
"step": 3080
|
|
},
|
|
{
|
|
"entropy": 0.9126953125,
|
|
"epoch": 0.4223330827581494,
|
|
"grad_norm": 0.2831146712076681,
|
|
"learning_rate": 4.066859236297027e-06,
|
|
"loss": 0.9124,
|
|
"mean_token_accuracy": 0.7666915744543076,
|
|
"num_tokens": 309387885.0,
|
|
"step": 3090
|
|
},
|
|
{
|
|
"entropy": 0.9154296875,
|
|
"epoch": 0.4236998564887583,
|
|
"grad_norm": 0.3687169744121827,
|
|
"learning_rate": 4.06333662110751e-06,
|
|
"loss": 0.9103,
|
|
"mean_token_accuracy": 0.7664415001869201,
|
|
"num_tokens": 310360880.0,
|
|
"step": 3100
|
|
},
|
|
{
|
|
"entropy": 0.9615234375,
|
|
"epoch": 0.4250666302193672,
|
|
"grad_norm": 0.3491129810537241,
|
|
"learning_rate": 4.059814005917994e-06,
|
|
"loss": 0.9673,
|
|
"mean_token_accuracy": 0.7564496964216232,
|
|
"num_tokens": 311413015.0,
|
|
"step": 3110
|
|
},
|
|
{
|
|
"entropy": 0.9125,
|
|
"epoch": 0.4264334039499761,
|
|
"grad_norm": 0.2654894073200961,
|
|
"learning_rate": 4.056291390728477e-06,
|
|
"loss": 0.9034,
|
|
"mean_token_accuracy": 0.7673636674880981,
|
|
"num_tokens": 312411848.0,
|
|
"step": 3120
|
|
},
|
|
{
|
|
"entropy": 0.9642578125,
|
|
"epoch": 0.427800177680585,
|
|
"grad_norm": 0.3159702303791009,
|
|
"learning_rate": 4.052768775538961e-06,
|
|
"loss": 0.9652,
|
|
"mean_token_accuracy": 0.7575671195983886,
|
|
"num_tokens": 313432995.0,
|
|
"step": 3130
|
|
},
|
|
{
|
|
"entropy": 0.9298828125,
|
|
"epoch": 0.4291669514111939,
|
|
"grad_norm": 0.32320266766009936,
|
|
"learning_rate": 4.049246160349444e-06,
|
|
"loss": 0.9191,
|
|
"mean_token_accuracy": 0.7659081518650055,
|
|
"num_tokens": 314460323.0,
|
|
"step": 3140
|
|
},
|
|
{
|
|
"entropy": 0.89921875,
|
|
"epoch": 0.43053372514180277,
|
|
"grad_norm": 0.2719682787549965,
|
|
"learning_rate": 4.045723545159927e-06,
|
|
"loss": 0.8957,
|
|
"mean_token_accuracy": 0.7690840870141983,
|
|
"num_tokens": 315425698.0,
|
|
"step": 3150
|
|
},
|
|
{
|
|
"entropy": 0.87578125,
|
|
"epoch": 0.43190049887241166,
|
|
"grad_norm": 0.2717765836905755,
|
|
"learning_rate": 4.04220092997041e-06,
|
|
"loss": 0.8779,
|
|
"mean_token_accuracy": 0.7739171862602234,
|
|
"num_tokens": 316440825.0,
|
|
"step": 3160
|
|
},
|
|
{
|
|
"entropy": 0.902734375,
|
|
"epoch": 0.43326727260302056,
|
|
"grad_norm": 0.27048054692537815,
|
|
"learning_rate": 4.038678314780894e-06,
|
|
"loss": 0.8917,
|
|
"mean_token_accuracy": 0.7704970121383667,
|
|
"num_tokens": 317499361.0,
|
|
"step": 3170
|
|
},
|
|
{
|
|
"entropy": 0.9228515625,
|
|
"epoch": 0.43463404633362945,
|
|
"grad_norm": 0.28223739979384294,
|
|
"learning_rate": 4.0351556995913775e-06,
|
|
"loss": 0.9203,
|
|
"mean_token_accuracy": 0.763028958439827,
|
|
"num_tokens": 318535159.0,
|
|
"step": 3180
|
|
},
|
|
{
|
|
"entropy": 0.903515625,
|
|
"epoch": 0.43600082006423835,
|
|
"grad_norm": 0.2567379283827537,
|
|
"learning_rate": 4.03163308440186e-06,
|
|
"loss": 0.9064,
|
|
"mean_token_accuracy": 0.7676450103521347,
|
|
"num_tokens": 319525232.0,
|
|
"step": 3190
|
|
},
|
|
{
|
|
"entropy": 0.9263671875,
|
|
"epoch": 0.43736759379484724,
|
|
"grad_norm": 0.3045259790606339,
|
|
"learning_rate": 4.028110469212344e-06,
|
|
"loss": 0.9261,
|
|
"mean_token_accuracy": 0.765043631196022,
|
|
"num_tokens": 320474887.0,
|
|
"step": 3200
|
|
},
|
|
{
|
|
"entropy": 0.9083984375,
|
|
"epoch": 0.43873436752545614,
|
|
"grad_norm": 0.5762279912761675,
|
|
"learning_rate": 4.0245878540228264e-06,
|
|
"loss": 0.9012,
|
|
"mean_token_accuracy": 0.767872828245163,
|
|
"num_tokens": 321463665.0,
|
|
"step": 3210
|
|
},
|
|
{
|
|
"entropy": 0.9525390625,
|
|
"epoch": 0.44010114125606503,
|
|
"grad_norm": 0.2986699604006375,
|
|
"learning_rate": 4.02106523883331e-06,
|
|
"loss": 0.9495,
|
|
"mean_token_accuracy": 0.7605759352445602,
|
|
"num_tokens": 322430041.0,
|
|
"step": 3220
|
|
},
|
|
{
|
|
"entropy": 0.8884765625,
|
|
"epoch": 0.441467914986674,
|
|
"grad_norm": 0.350383925217884,
|
|
"learning_rate": 4.0175426236437935e-06,
|
|
"loss": 0.8932,
|
|
"mean_token_accuracy": 0.770888340473175,
|
|
"num_tokens": 323393790.0,
|
|
"step": 3230
|
|
},
|
|
{
|
|
"entropy": 0.8935546875,
|
|
"epoch": 0.4428346887172829,
|
|
"grad_norm": 0.3330721352278246,
|
|
"learning_rate": 4.014020008454277e-06,
|
|
"loss": 0.8927,
|
|
"mean_token_accuracy": 0.7704477548599243,
|
|
"num_tokens": 324396436.0,
|
|
"step": 3240
|
|
},
|
|
{
|
|
"entropy": 0.900390625,
|
|
"epoch": 0.44420146244789177,
|
|
"grad_norm": 0.26531983470929826,
|
|
"learning_rate": 4.01049739326476e-06,
|
|
"loss": 0.8867,
|
|
"mean_token_accuracy": 0.7691289752721786,
|
|
"num_tokens": 325343170.0,
|
|
"step": 3250
|
|
},
|
|
{
|
|
"entropy": 0.911328125,
|
|
"epoch": 0.44556823617850067,
|
|
"grad_norm": 0.26918611016941413,
|
|
"learning_rate": 4.006974778075243e-06,
|
|
"loss": 0.9095,
|
|
"mean_token_accuracy": 0.7677451461553574,
|
|
"num_tokens": 326355307.0,
|
|
"step": 3260
|
|
},
|
|
{
|
|
"entropy": 0.93515625,
|
|
"epoch": 0.44693500990910956,
|
|
"grad_norm": 0.26872342979742314,
|
|
"learning_rate": 4.003452162885727e-06,
|
|
"loss": 0.9347,
|
|
"mean_token_accuracy": 0.7598981887102128,
|
|
"num_tokens": 327388298.0,
|
|
"step": 3270
|
|
},
|
|
{
|
|
"entropy": 0.8767578125,
|
|
"epoch": 0.44830178363971845,
|
|
"grad_norm": 0.30723749478546475,
|
|
"learning_rate": 3.99992954769621e-06,
|
|
"loss": 0.8797,
|
|
"mean_token_accuracy": 0.7726427286863327,
|
|
"num_tokens": 328389066.0,
|
|
"step": 3280
|
|
},
|
|
{
|
|
"entropy": 0.8916015625,
|
|
"epoch": 0.44966855737032735,
|
|
"grad_norm": 0.3508044221956045,
|
|
"learning_rate": 3.996406932506693e-06,
|
|
"loss": 0.8869,
|
|
"mean_token_accuracy": 0.7703417629003525,
|
|
"num_tokens": 329429521.0,
|
|
"step": 3290
|
|
},
|
|
{
|
|
"entropy": 0.8775390625,
|
|
"epoch": 0.45103533110093624,
|
|
"grad_norm": 0.5898278330946807,
|
|
"learning_rate": 3.992884317317177e-06,
|
|
"loss": 0.8798,
|
|
"mean_token_accuracy": 0.7717438817024231,
|
|
"num_tokens": 330482625.0,
|
|
"step": 3300
|
|
},
|
|
{
|
|
"entropy": 0.9412109375,
|
|
"epoch": 0.45240210483154514,
|
|
"grad_norm": 0.33778745904828883,
|
|
"learning_rate": 3.98936170212766e-06,
|
|
"loss": 0.9403,
|
|
"mean_token_accuracy": 0.7613541930913925,
|
|
"num_tokens": 331506313.0,
|
|
"step": 3310
|
|
},
|
|
{
|
|
"entropy": 0.879296875,
|
|
"epoch": 0.45376887856215403,
|
|
"grad_norm": 0.322235984712991,
|
|
"learning_rate": 3.985839086938143e-06,
|
|
"loss": 0.8704,
|
|
"mean_token_accuracy": 0.7748452395200729,
|
|
"num_tokens": 332494855.0,
|
|
"step": 3320
|
|
},
|
|
{
|
|
"entropy": 0.909375,
|
|
"epoch": 0.4551356522927629,
|
|
"grad_norm": 0.29557364954459264,
|
|
"learning_rate": 3.9823164717486265e-06,
|
|
"loss": 0.9083,
|
|
"mean_token_accuracy": 0.766100424528122,
|
|
"num_tokens": 333547164.0,
|
|
"step": 3330
|
|
},
|
|
{
|
|
"entropy": 0.9150390625,
|
|
"epoch": 0.4565024260233718,
|
|
"grad_norm": 0.30776055657610957,
|
|
"learning_rate": 3.97879385655911e-06,
|
|
"loss": 0.9123,
|
|
"mean_token_accuracy": 0.7660103142261505,
|
|
"num_tokens": 334529077.0,
|
|
"step": 3340
|
|
},
|
|
{
|
|
"entropy": 0.93359375,
|
|
"epoch": 0.4578691997539807,
|
|
"grad_norm": 0.4670815725698831,
|
|
"learning_rate": 3.9752712413695936e-06,
|
|
"loss": 0.9331,
|
|
"mean_token_accuracy": 0.7621869534254074,
|
|
"num_tokens": 335516657.0,
|
|
"step": 3350
|
|
},
|
|
{
|
|
"entropy": 0.958984375,
|
|
"epoch": 0.4592359734845896,
|
|
"grad_norm": 0.29560525392777487,
|
|
"learning_rate": 3.971748626180076e-06,
|
|
"loss": 0.9599,
|
|
"mean_token_accuracy": 0.7563399374485016,
|
|
"num_tokens": 336529399.0,
|
|
"step": 3360
|
|
},
|
|
{
|
|
"entropy": 0.9201171875,
|
|
"epoch": 0.4606027472151985,
|
|
"grad_norm": 0.2813656122426702,
|
|
"learning_rate": 3.96822601099056e-06,
|
|
"loss": 0.9132,
|
|
"mean_token_accuracy": 0.7650532245635986,
|
|
"num_tokens": 337575192.0,
|
|
"step": 3370
|
|
},
|
|
{
|
|
"entropy": 0.9158203125,
|
|
"epoch": 0.4619695209458074,
|
|
"grad_norm": 0.32506254958522635,
|
|
"learning_rate": 3.9647033958010425e-06,
|
|
"loss": 0.9247,
|
|
"mean_token_accuracy": 0.7626821786165238,
|
|
"num_tokens": 338561351.0,
|
|
"step": 3380
|
|
},
|
|
{
|
|
"entropy": 0.9046875,
|
|
"epoch": 0.4633362946764163,
|
|
"grad_norm": 0.30033828048940725,
|
|
"learning_rate": 3.961180780611527e-06,
|
|
"loss": 0.9038,
|
|
"mean_token_accuracy": 0.7658067494630814,
|
|
"num_tokens": 339557736.0,
|
|
"step": 3390
|
|
},
|
|
{
|
|
"entropy": 0.9392578125,
|
|
"epoch": 0.46470306840702524,
|
|
"grad_norm": 0.2719536380683746,
|
|
"learning_rate": 3.95765816542201e-06,
|
|
"loss": 0.9332,
|
|
"mean_token_accuracy": 0.7626032143831253,
|
|
"num_tokens": 340603014.0,
|
|
"step": 3400
|
|
},
|
|
{
|
|
"entropy": 0.88203125,
|
|
"epoch": 0.46606984213763414,
|
|
"grad_norm": 0.27726076362756935,
|
|
"learning_rate": 3.954135550232493e-06,
|
|
"loss": 0.8721,
|
|
"mean_token_accuracy": 0.7719032555818558,
|
|
"num_tokens": 341584764.0,
|
|
"step": 3410
|
|
},
|
|
{
|
|
"entropy": 0.9244140625,
|
|
"epoch": 0.46743661586824303,
|
|
"grad_norm": 0.2881208513751079,
|
|
"learning_rate": 3.950612935042976e-06,
|
|
"loss": 0.9177,
|
|
"mean_token_accuracy": 0.7642891764640808,
|
|
"num_tokens": 342590031.0,
|
|
"step": 3420
|
|
},
|
|
{
|
|
"entropy": 0.887109375,
|
|
"epoch": 0.4688033895988519,
|
|
"grad_norm": 0.3376843720491647,
|
|
"learning_rate": 3.947090319853459e-06,
|
|
"loss": 0.8711,
|
|
"mean_token_accuracy": 0.7733410805463791,
|
|
"num_tokens": 343580633.0,
|
|
"step": 3430
|
|
},
|
|
{
|
|
"entropy": 0.9587890625,
|
|
"epoch": 0.4701701633294608,
|
|
"grad_norm": 0.4670715341207451,
|
|
"learning_rate": 3.943567704663943e-06,
|
|
"loss": 0.9503,
|
|
"mean_token_accuracy": 0.7599731296300888,
|
|
"num_tokens": 344555481.0,
|
|
"step": 3440
|
|
},
|
|
{
|
|
"entropy": 0.9046875,
|
|
"epoch": 0.4715369370600697,
|
|
"grad_norm": 0.34543298270774614,
|
|
"learning_rate": 3.9400450894744265e-06,
|
|
"loss": 0.8989,
|
|
"mean_token_accuracy": 0.7671310931444169,
|
|
"num_tokens": 345513015.0,
|
|
"step": 3450
|
|
},
|
|
{
|
|
"entropy": 0.9556640625,
|
|
"epoch": 0.4729037107906786,
|
|
"grad_norm": 0.39784582280505437,
|
|
"learning_rate": 3.936522474284909e-06,
|
|
"loss": 0.9707,
|
|
"mean_token_accuracy": 0.7558092594146728,
|
|
"num_tokens": 346519872.0,
|
|
"step": 3460
|
|
},
|
|
{
|
|
"entropy": 0.90625,
|
|
"epoch": 0.4742704845212875,
|
|
"grad_norm": 0.3527718639413328,
|
|
"learning_rate": 3.932999859095393e-06,
|
|
"loss": 0.9055,
|
|
"mean_token_accuracy": 0.7675750434398652,
|
|
"num_tokens": 347593024.0,
|
|
"step": 3470
|
|
},
|
|
{
|
|
"entropy": 0.8951171875,
|
|
"epoch": 0.4756372582518964,
|
|
"grad_norm": 0.29611904384393717,
|
|
"learning_rate": 3.929477243905876e-06,
|
|
"loss": 0.8898,
|
|
"mean_token_accuracy": 0.7714505791664124,
|
|
"num_tokens": 348594832.0,
|
|
"step": 3480
|
|
},
|
|
{
|
|
"entropy": 0.93125,
|
|
"epoch": 0.4770040319825053,
|
|
"grad_norm": 0.3531373374894184,
|
|
"learning_rate": 3.925954628716359e-06,
|
|
"loss": 0.9245,
|
|
"mean_token_accuracy": 0.7656036585569381,
|
|
"num_tokens": 349593190.0,
|
|
"step": 3490
|
|
},
|
|
{
|
|
"entropy": 0.9375,
|
|
"epoch": 0.4783708057131142,
|
|
"grad_norm": 0.2972079335434535,
|
|
"learning_rate": 3.9224320135268425e-06,
|
|
"loss": 0.9291,
|
|
"mean_token_accuracy": 0.7614673733711242,
|
|
"num_tokens": 350561915.0,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"entropy": 0.9197265625,
|
|
"epoch": 0.4797375794437231,
|
|
"grad_norm": 0.34684490750505553,
|
|
"learning_rate": 3.918909398337326e-06,
|
|
"loss": 0.9155,
|
|
"mean_token_accuracy": 0.7654666513204574,
|
|
"num_tokens": 351565428.0,
|
|
"step": 3510
|
|
},
|
|
{
|
|
"entropy": 0.890625,
|
|
"epoch": 0.481104353174332,
|
|
"grad_norm": 0.2976836949972579,
|
|
"learning_rate": 3.91538678314781e-06,
|
|
"loss": 0.882,
|
|
"mean_token_accuracy": 0.773875018954277,
|
|
"num_tokens": 352587603.0,
|
|
"step": 3520
|
|
},
|
|
{
|
|
"entropy": 0.9287109375,
|
|
"epoch": 0.4824711269049409,
|
|
"grad_norm": 0.29355319569326654,
|
|
"learning_rate": 3.911864167958292e-06,
|
|
"loss": 0.9142,
|
|
"mean_token_accuracy": 0.7662282645702362,
|
|
"num_tokens": 353602912.0,
|
|
"step": 3530
|
|
},
|
|
{
|
|
"entropy": 0.907421875,
|
|
"epoch": 0.48383790063554977,
|
|
"grad_norm": 0.27664719434026136,
|
|
"learning_rate": 3.908341552768776e-06,
|
|
"loss": 0.8994,
|
|
"mean_token_accuracy": 0.7678870558738708,
|
|
"num_tokens": 354606364.0,
|
|
"step": 3540
|
|
},
|
|
{
|
|
"entropy": 0.8970703125,
|
|
"epoch": 0.48520467436615866,
|
|
"grad_norm": 0.29386242961408415,
|
|
"learning_rate": 3.9048189375792586e-06,
|
|
"loss": 0.8971,
|
|
"mean_token_accuracy": 0.7671164095401763,
|
|
"num_tokens": 355587589.0,
|
|
"step": 3550
|
|
},
|
|
{
|
|
"entropy": 0.946484375,
|
|
"epoch": 0.48657144809676756,
|
|
"grad_norm": 0.28761693274827965,
|
|
"learning_rate": 3.901296322389743e-06,
|
|
"loss": 0.9421,
|
|
"mean_token_accuracy": 0.7598811119794846,
|
|
"num_tokens": 356558152.0,
|
|
"step": 3560
|
|
},
|
|
{
|
|
"entropy": 0.92890625,
|
|
"epoch": 0.48793822182737645,
|
|
"grad_norm": 0.37622698126682486,
|
|
"learning_rate": 3.897773707200226e-06,
|
|
"loss": 0.9174,
|
|
"mean_token_accuracy": 0.7629639774560928,
|
|
"num_tokens": 357540541.0,
|
|
"step": 3570
|
|
},
|
|
{
|
|
"entropy": 0.9275390625,
|
|
"epoch": 0.4893049955579854,
|
|
"grad_norm": 0.29484667502836026,
|
|
"learning_rate": 3.894251092010709e-06,
|
|
"loss": 0.9086,
|
|
"mean_token_accuracy": 0.7645941734313965,
|
|
"num_tokens": 358570092.0,
|
|
"step": 3580
|
|
},
|
|
{
|
|
"entropy": 0.875390625,
|
|
"epoch": 0.4906717692885943,
|
|
"grad_norm": 0.3150979787909053,
|
|
"learning_rate": 3.890728476821192e-06,
|
|
"loss": 0.864,
|
|
"mean_token_accuracy": 0.7765456914901734,
|
|
"num_tokens": 359534391.0,
|
|
"step": 3590
|
|
},
|
|
{
|
|
"entropy": 0.8564453125,
|
|
"epoch": 0.4920385430192032,
|
|
"grad_norm": 0.3271723916577464,
|
|
"learning_rate": 3.8872058616316755e-06,
|
|
"loss": 0.8473,
|
|
"mean_token_accuracy": 0.7804722160100936,
|
|
"num_tokens": 360497311.0,
|
|
"step": 3600
|
|
},
|
|
{
|
|
"entropy": 0.9048828125,
|
|
"epoch": 0.4934053167498121,
|
|
"grad_norm": 0.2908443079528101,
|
|
"learning_rate": 3.883683246442159e-06,
|
|
"loss": 0.8938,
|
|
"mean_token_accuracy": 0.769324392080307,
|
|
"num_tokens": 361488754.0,
|
|
"step": 3610
|
|
},
|
|
{
|
|
"entropy": 0.927734375,
|
|
"epoch": 0.494772090480421,
|
|
"grad_norm": 0.2840820548805122,
|
|
"learning_rate": 3.8801606312526426e-06,
|
|
"loss": 0.9214,
|
|
"mean_token_accuracy": 0.7649717539548874,
|
|
"num_tokens": 362539500.0,
|
|
"step": 3620
|
|
},
|
|
{
|
|
"entropy": 0.8404296875,
|
|
"epoch": 0.4961388642110299,
|
|
"grad_norm": 0.26597414201223707,
|
|
"learning_rate": 3.876638016063125e-06,
|
|
"loss": 0.8278,
|
|
"mean_token_accuracy": 0.7843117862939835,
|
|
"num_tokens": 363566701.0,
|
|
"step": 3630
|
|
},
|
|
{
|
|
"entropy": 0.880078125,
|
|
"epoch": 0.49750563794163877,
|
|
"grad_norm": 0.30158477915151066,
|
|
"learning_rate": 3.873115400873609e-06,
|
|
"loss": 0.874,
|
|
"mean_token_accuracy": 0.7740333557128907,
|
|
"num_tokens": 364529201.0,
|
|
"step": 3640
|
|
},
|
|
{
|
|
"entropy": 0.9208984375,
|
|
"epoch": 0.49887241167224766,
|
|
"grad_norm": 0.29139520921460155,
|
|
"learning_rate": 3.869592785684092e-06,
|
|
"loss": 0.9178,
|
|
"mean_token_accuracy": 0.7660344988107681,
|
|
"num_tokens": 365535910.0,
|
|
"step": 3650
|
|
},
|
|
{
|
|
"entropy": 0.91953125,
|
|
"epoch": 0.5002391854028566,
|
|
"grad_norm": 0.4901625442978379,
|
|
"learning_rate": 3.866070170494575e-06,
|
|
"loss": 0.9173,
|
|
"mean_token_accuracy": 0.7665181666612625,
|
|
"num_tokens": 366511634.0,
|
|
"step": 3660
|
|
},
|
|
{
|
|
"entropy": 0.912109375,
|
|
"epoch": 0.5016059591334655,
|
|
"grad_norm": 0.3483016106773495,
|
|
"learning_rate": 3.862547555305059e-06,
|
|
"loss": 0.9153,
|
|
"mean_token_accuracy": 0.7640544503927231,
|
|
"num_tokens": 367546812.0,
|
|
"step": 3670
|
|
},
|
|
{
|
|
"entropy": 0.914453125,
|
|
"epoch": 0.5029727328640744,
|
|
"grad_norm": 0.3139030521478235,
|
|
"learning_rate": 3.859024940115542e-06,
|
|
"loss": 0.9187,
|
|
"mean_token_accuracy": 0.7653339385986329,
|
|
"num_tokens": 368552370.0,
|
|
"step": 3680
|
|
},
|
|
{
|
|
"entropy": 0.880859375,
|
|
"epoch": 0.5043395065946833,
|
|
"grad_norm": 0.26817678690381996,
|
|
"learning_rate": 3.855502324926026e-06,
|
|
"loss": 0.8747,
|
|
"mean_token_accuracy": 0.776081171631813,
|
|
"num_tokens": 369603613.0,
|
|
"step": 3690
|
|
},
|
|
{
|
|
"entropy": 0.914453125,
|
|
"epoch": 0.5057062803252922,
|
|
"grad_norm": 0.2895912714845701,
|
|
"learning_rate": 3.851979709736508e-06,
|
|
"loss": 0.9137,
|
|
"mean_token_accuracy": 0.7688512921333313,
|
|
"num_tokens": 370579772.0,
|
|
"step": 3700
|
|
},
|
|
{
|
|
"entropy": 0.921875,
|
|
"epoch": 0.5070730540559011,
|
|
"grad_norm": 0.3040254031945483,
|
|
"learning_rate": 3.848457094546992e-06,
|
|
"loss": 0.9072,
|
|
"mean_token_accuracy": 0.7678411036729813,
|
|
"num_tokens": 371569052.0,
|
|
"step": 3710
|
|
},
|
|
{
|
|
"entropy": 0.88828125,
|
|
"epoch": 0.50843982778651,
|
|
"grad_norm": 0.30301854479617346,
|
|
"learning_rate": 3.8449344793574755e-06,
|
|
"loss": 0.8896,
|
|
"mean_token_accuracy": 0.7716005593538284,
|
|
"num_tokens": 372600551.0,
|
|
"step": 3720
|
|
},
|
|
{
|
|
"entropy": 0.916796875,
|
|
"epoch": 0.5098066015171189,
|
|
"grad_norm": 0.25999640640849536,
|
|
"learning_rate": 3.841411864167959e-06,
|
|
"loss": 0.9131,
|
|
"mean_token_accuracy": 0.7656413465738297,
|
|
"num_tokens": 373584656.0,
|
|
"step": 3730
|
|
},
|
|
{
|
|
"entropy": 0.919921875,
|
|
"epoch": 0.5111733752477278,
|
|
"grad_norm": 0.28751830933615685,
|
|
"learning_rate": 3.837889248978442e-06,
|
|
"loss": 0.9193,
|
|
"mean_token_accuracy": 0.7643132716417312,
|
|
"num_tokens": 374563340.0,
|
|
"step": 3740
|
|
},
|
|
{
|
|
"entropy": 0.9107421875,
|
|
"epoch": 0.5125401489783367,
|
|
"grad_norm": 0.28939719621059695,
|
|
"learning_rate": 3.834366633788925e-06,
|
|
"loss": 0.9078,
|
|
"mean_token_accuracy": 0.7688408941030502,
|
|
"num_tokens": 375567281.0,
|
|
"step": 3750
|
|
},
|
|
{
|
|
"entropy": 0.9224609375,
|
|
"epoch": 0.5139069227089456,
|
|
"grad_norm": 0.31672422192856,
|
|
"learning_rate": 3.830844018599408e-06,
|
|
"loss": 0.9185,
|
|
"mean_token_accuracy": 0.763445594906807,
|
|
"num_tokens": 376511811.0,
|
|
"step": 3760
|
|
},
|
|
{
|
|
"entropy": 0.93125,
|
|
"epoch": 0.5152736964395545,
|
|
"grad_norm": 0.2893205841844033,
|
|
"learning_rate": 3.8273214034098915e-06,
|
|
"loss": 0.9252,
|
|
"mean_token_accuracy": 0.7645682662725448,
|
|
"num_tokens": 377512831.0,
|
|
"step": 3770
|
|
},
|
|
{
|
|
"entropy": 0.931640625,
|
|
"epoch": 0.5166404701701633,
|
|
"grad_norm": 0.2836316616423504,
|
|
"learning_rate": 3.823798788220375e-06,
|
|
"loss": 0.9174,
|
|
"mean_token_accuracy": 0.7634561747312546,
|
|
"num_tokens": 378478492.0,
|
|
"step": 3780
|
|
},
|
|
{
|
|
"entropy": 0.8923828125,
|
|
"epoch": 0.5180072439007722,
|
|
"grad_norm": 0.3350512260123635,
|
|
"learning_rate": 3.820276173030859e-06,
|
|
"loss": 0.8892,
|
|
"mean_token_accuracy": 0.7698444634675979,
|
|
"num_tokens": 379444691.0,
|
|
"step": 3790
|
|
},
|
|
{
|
|
"entropy": 0.9126953125,
|
|
"epoch": 0.5193740176313811,
|
|
"grad_norm": 0.30811632113979176,
|
|
"learning_rate": 3.816753557841341e-06,
|
|
"loss": 0.9116,
|
|
"mean_token_accuracy": 0.7656842708587647,
|
|
"num_tokens": 380423660.0,
|
|
"step": 3800
|
|
},
|
|
{
|
|
"entropy": 0.9181640625,
|
|
"epoch": 0.52074079136199,
|
|
"grad_norm": 0.2875646252045547,
|
|
"learning_rate": 3.813230942651825e-06,
|
|
"loss": 0.9132,
|
|
"mean_token_accuracy": 0.7665803611278534,
|
|
"num_tokens": 381365233.0,
|
|
"step": 3810
|
|
},
|
|
{
|
|
"entropy": 0.9099609375,
|
|
"epoch": 0.5221075650925989,
|
|
"grad_norm": 0.27371240320180107,
|
|
"learning_rate": 3.809708327462308e-06,
|
|
"loss": 0.9022,
|
|
"mean_token_accuracy": 0.7669348120689392,
|
|
"num_tokens": 382338629.0,
|
|
"step": 3820
|
|
},
|
|
{
|
|
"entropy": 0.9173828125,
|
|
"epoch": 0.5234743388232078,
|
|
"grad_norm": 0.36357628320781726,
|
|
"learning_rate": 3.806185712272792e-06,
|
|
"loss": 0.9142,
|
|
"mean_token_accuracy": 0.7653904914855957,
|
|
"num_tokens": 383353565.0,
|
|
"step": 3830
|
|
},
|
|
{
|
|
"entropy": 0.9861328125,
|
|
"epoch": 0.5248411125538167,
|
|
"grad_norm": 0.29290348456649723,
|
|
"learning_rate": 3.802663097083275e-06,
|
|
"loss": 0.9929,
|
|
"mean_token_accuracy": 0.7494483321905137,
|
|
"num_tokens": 384284378.0,
|
|
"step": 3840
|
|
},
|
|
{
|
|
"entropy": 0.916015625,
|
|
"epoch": 0.5262078862844256,
|
|
"grad_norm": 0.3128394897747115,
|
|
"learning_rate": 3.799140481893758e-06,
|
|
"loss": 0.9058,
|
|
"mean_token_accuracy": 0.76614408493042,
|
|
"num_tokens": 385243988.0,
|
|
"step": 3850
|
|
},
|
|
{
|
|
"entropy": 0.90234375,
|
|
"epoch": 0.5275746600150345,
|
|
"grad_norm": 0.30860425160063615,
|
|
"learning_rate": 3.7956178667042413e-06,
|
|
"loss": 0.8911,
|
|
"mean_token_accuracy": 0.7717981785535812,
|
|
"num_tokens": 386204705.0,
|
|
"step": 3860
|
|
},
|
|
{
|
|
"entropy": 0.8796875,
|
|
"epoch": 0.5289414337456434,
|
|
"grad_norm": 0.2742486297676268,
|
|
"learning_rate": 3.792095251514725e-06,
|
|
"loss": 0.8745,
|
|
"mean_token_accuracy": 0.7722369939088821,
|
|
"num_tokens": 387221442.0,
|
|
"step": 3870
|
|
},
|
|
{
|
|
"entropy": 0.8775390625,
|
|
"epoch": 0.5303082074762523,
|
|
"grad_norm": 0.27965510424043366,
|
|
"learning_rate": 3.788572636325208e-06,
|
|
"loss": 0.8765,
|
|
"mean_token_accuracy": 0.7734510958194732,
|
|
"num_tokens": 388252567.0,
|
|
"step": 3880
|
|
},
|
|
{
|
|
"entropy": 0.8791015625,
|
|
"epoch": 0.5316749812068612,
|
|
"grad_norm": 0.26885519391200596,
|
|
"learning_rate": 3.7850500211356916e-06,
|
|
"loss": 0.8746,
|
|
"mean_token_accuracy": 0.7730431228876113,
|
|
"num_tokens": 389259487.0,
|
|
"step": 3890
|
|
},
|
|
{
|
|
"entropy": 0.914453125,
|
|
"epoch": 0.5330417549374701,
|
|
"grad_norm": 0.29481424392489175,
|
|
"learning_rate": 3.7815274059461747e-06,
|
|
"loss": 0.9096,
|
|
"mean_token_accuracy": 0.7653242528438569,
|
|
"num_tokens": 390253163.0,
|
|
"step": 3900
|
|
},
|
|
{
|
|
"entropy": 0.8955078125,
|
|
"epoch": 0.534408528668079,
|
|
"grad_norm": 0.24637575529874342,
|
|
"learning_rate": 3.7780047907566582e-06,
|
|
"loss": 0.9014,
|
|
"mean_token_accuracy": 0.7680971264839173,
|
|
"num_tokens": 391269104.0,
|
|
"step": 3910
|
|
},
|
|
{
|
|
"entropy": 0.893359375,
|
|
"epoch": 0.5357753023986879,
|
|
"grad_norm": 0.2800280696644179,
|
|
"learning_rate": 3.7744821755671413e-06,
|
|
"loss": 0.8885,
|
|
"mean_token_accuracy": 0.77311290204525,
|
|
"num_tokens": 392288204.0,
|
|
"step": 3920
|
|
},
|
|
{
|
|
"entropy": 0.905859375,
|
|
"epoch": 0.5371420761292968,
|
|
"grad_norm": 0.27809411682324964,
|
|
"learning_rate": 3.7709595603776245e-06,
|
|
"loss": 0.9207,
|
|
"mean_token_accuracy": 0.7649626791477203,
|
|
"num_tokens": 393303667.0,
|
|
"step": 3930
|
|
},
|
|
{
|
|
"entropy": 0.9306640625,
|
|
"epoch": 0.5385088498599057,
|
|
"grad_norm": 0.30494466582618396,
|
|
"learning_rate": 3.767436945188108e-06,
|
|
"loss": 0.9312,
|
|
"mean_token_accuracy": 0.7618204414844513,
|
|
"num_tokens": 394294972.0,
|
|
"step": 3940
|
|
},
|
|
{
|
|
"entropy": 0.9001953125,
|
|
"epoch": 0.5398756235905146,
|
|
"grad_norm": 0.3026341651225417,
|
|
"learning_rate": 3.7639143299985916e-06,
|
|
"loss": 0.8962,
|
|
"mean_token_accuracy": 0.7696307480335236,
|
|
"num_tokens": 395281144.0,
|
|
"step": 3950
|
|
},
|
|
{
|
|
"entropy": 0.93203125,
|
|
"epoch": 0.5412423973211234,
|
|
"grad_norm": 0.3330408479442847,
|
|
"learning_rate": 3.7603917148090747e-06,
|
|
"loss": 0.9343,
|
|
"mean_token_accuracy": 0.7613282024860382,
|
|
"num_tokens": 396282458.0,
|
|
"step": 3960
|
|
},
|
|
{
|
|
"entropy": 0.88515625,
|
|
"epoch": 0.5426091710517323,
|
|
"grad_norm": 0.2723431285793291,
|
|
"learning_rate": 3.756869099619558e-06,
|
|
"loss": 0.8885,
|
|
"mean_token_accuracy": 0.7719673722982406,
|
|
"num_tokens": 397299233.0,
|
|
"step": 3970
|
|
},
|
|
{
|
|
"entropy": 0.88046875,
|
|
"epoch": 0.5439759447823412,
|
|
"grad_norm": 0.2675805469489099,
|
|
"learning_rate": 3.753346484430041e-06,
|
|
"loss": 0.8775,
|
|
"mean_token_accuracy": 0.7731725662946701,
|
|
"num_tokens": 398281717.0,
|
|
"step": 3980
|
|
},
|
|
{
|
|
"entropy": 0.8861328125,
|
|
"epoch": 0.5453427185129501,
|
|
"grad_norm": 0.2874667045933724,
|
|
"learning_rate": 3.749823869240524e-06,
|
|
"loss": 0.8822,
|
|
"mean_token_accuracy": 0.772596788406372,
|
|
"num_tokens": 399265146.0,
|
|
"step": 3990
|
|
},
|
|
{
|
|
"entropy": 0.8703125,
|
|
"epoch": 0.546709492243559,
|
|
"grad_norm": 0.2970701919354301,
|
|
"learning_rate": 3.746301254051008e-06,
|
|
"loss": 0.8667,
|
|
"mean_token_accuracy": 0.7752142727375031,
|
|
"num_tokens": 400256118.0,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"entropy": 0.869140625,
|
|
"epoch": 0.548076265974168,
|
|
"grad_norm": 0.2846231590904809,
|
|
"learning_rate": 3.742778638861491e-06,
|
|
"loss": 0.8671,
|
|
"mean_token_accuracy": 0.7744826197624206,
|
|
"num_tokens": 401251718.0,
|
|
"step": 4010
|
|
},
|
|
{
|
|
"entropy": 0.9041015625,
|
|
"epoch": 0.5494430397047769,
|
|
"grad_norm": 0.3057980224969252,
|
|
"learning_rate": 3.7392560236719743e-06,
|
|
"loss": 0.8956,
|
|
"mean_token_accuracy": 0.7677527666091919,
|
|
"num_tokens": 402208932.0,
|
|
"step": 4020
|
|
},
|
|
{
|
|
"entropy": 0.864453125,
|
|
"epoch": 0.5508098134353858,
|
|
"grad_norm": 0.291725772018906,
|
|
"learning_rate": 3.7357334084824574e-06,
|
|
"loss": 0.8612,
|
|
"mean_token_accuracy": 0.7776931524276733,
|
|
"num_tokens": 403201031.0,
|
|
"step": 4030
|
|
},
|
|
{
|
|
"entropy": 0.8779296875,
|
|
"epoch": 0.5521765871659947,
|
|
"grad_norm": 0.3782384589791917,
|
|
"learning_rate": 3.732210793292941e-06,
|
|
"loss": 0.8627,
|
|
"mean_token_accuracy": 0.7764532834291458,
|
|
"num_tokens": 404208352.0,
|
|
"step": 4040
|
|
},
|
|
{
|
|
"entropy": 0.915625,
|
|
"epoch": 0.5535433608966036,
|
|
"grad_norm": 0.30776785464068485,
|
|
"learning_rate": 3.7286881781034245e-06,
|
|
"loss": 0.9137,
|
|
"mean_token_accuracy": 0.7651795536279679,
|
|
"num_tokens": 405228257.0,
|
|
"step": 4050
|
|
},
|
|
{
|
|
"entropy": 0.9013671875,
|
|
"epoch": 0.5549101346272125,
|
|
"grad_norm": 0.29942887284935915,
|
|
"learning_rate": 3.7251655629139076e-06,
|
|
"loss": 0.8913,
|
|
"mean_token_accuracy": 0.7703980922698974,
|
|
"num_tokens": 406262561.0,
|
|
"step": 4060
|
|
},
|
|
{
|
|
"entropy": 0.9458984375,
|
|
"epoch": 0.5562769083578214,
|
|
"grad_norm": 0.2635629362667431,
|
|
"learning_rate": 3.7216429477243907e-06,
|
|
"loss": 0.9317,
|
|
"mean_token_accuracy": 0.7611770480871201,
|
|
"num_tokens": 407259161.0,
|
|
"step": 4070
|
|
},
|
|
{
|
|
"entropy": 0.897265625,
|
|
"epoch": 0.5576436820884303,
|
|
"grad_norm": 0.26265451390789274,
|
|
"learning_rate": 3.7181203325348743e-06,
|
|
"loss": 0.8915,
|
|
"mean_token_accuracy": 0.7696784257888794,
|
|
"num_tokens": 408297673.0,
|
|
"step": 4080
|
|
},
|
|
{
|
|
"entropy": 0.938671875,
|
|
"epoch": 0.5590104558190392,
|
|
"grad_norm": 0.28259640762302224,
|
|
"learning_rate": 3.7145977173453574e-06,
|
|
"loss": 0.9282,
|
|
"mean_token_accuracy": 0.7610027700662613,
|
|
"num_tokens": 409266496.0,
|
|
"step": 4090
|
|
},
|
|
{
|
|
"entropy": 0.8912109375,
|
|
"epoch": 0.5603772295496481,
|
|
"grad_norm": 0.2864693958538041,
|
|
"learning_rate": 3.7110751021558405e-06,
|
|
"loss": 0.8918,
|
|
"mean_token_accuracy": 0.76878821849823,
|
|
"num_tokens": 410238401.0,
|
|
"step": 4100
|
|
},
|
|
{
|
|
"entropy": 0.8806640625,
|
|
"epoch": 0.561744003280257,
|
|
"grad_norm": 0.27079342987351834,
|
|
"learning_rate": 3.7075524869663245e-06,
|
|
"loss": 0.8752,
|
|
"mean_token_accuracy": 0.7720946639776229,
|
|
"num_tokens": 411280517.0,
|
|
"step": 4110
|
|
},
|
|
{
|
|
"entropy": 0.883984375,
|
|
"epoch": 0.5631107770108659,
|
|
"grad_norm": 0.279969981902978,
|
|
"learning_rate": 3.7040298717768076e-06,
|
|
"loss": 0.8777,
|
|
"mean_token_accuracy": 0.7743743747472763,
|
|
"num_tokens": 412309765.0,
|
|
"step": 4120
|
|
},
|
|
{
|
|
"entropy": 0.927734375,
|
|
"epoch": 0.5644775507414748,
|
|
"grad_norm": 0.2888917910192825,
|
|
"learning_rate": 3.7005072565872908e-06,
|
|
"loss": 0.9142,
|
|
"mean_token_accuracy": 0.76567742228508,
|
|
"num_tokens": 413340967.0,
|
|
"step": 4130
|
|
},
|
|
{
|
|
"entropy": 0.91640625,
|
|
"epoch": 0.5658443244720837,
|
|
"grad_norm": 0.2852087139517509,
|
|
"learning_rate": 3.696984641397774e-06,
|
|
"loss": 0.9196,
|
|
"mean_token_accuracy": 0.7639094412326812,
|
|
"num_tokens": 414340603.0,
|
|
"step": 4140
|
|
},
|
|
{
|
|
"entropy": 0.8740234375,
|
|
"epoch": 0.5672110982026926,
|
|
"grad_norm": 0.2959627163923763,
|
|
"learning_rate": 3.693462026208257e-06,
|
|
"loss": 0.8664,
|
|
"mean_token_accuracy": 0.774793666601181,
|
|
"num_tokens": 415339057.0,
|
|
"step": 4150
|
|
},
|
|
{
|
|
"entropy": 0.9109375,
|
|
"epoch": 0.5685778719333014,
|
|
"grad_norm": 0.3692617091265936,
|
|
"learning_rate": 3.689939411018741e-06,
|
|
"loss": 0.91,
|
|
"mean_token_accuracy": 0.7686929345130921,
|
|
"num_tokens": 416381610.0,
|
|
"step": 4160
|
|
},
|
|
{
|
|
"entropy": 0.84140625,
|
|
"epoch": 0.5699446456639103,
|
|
"grad_norm": 0.3216346745879844,
|
|
"learning_rate": 3.686416795829224e-06,
|
|
"loss": 0.8302,
|
|
"mean_token_accuracy": 0.7819727391004563,
|
|
"num_tokens": 417378752.0,
|
|
"step": 4170
|
|
},
|
|
{
|
|
"entropy": 0.8841796875,
|
|
"epoch": 0.5713114193945192,
|
|
"grad_norm": 0.29727537266058485,
|
|
"learning_rate": 3.6828941806397072e-06,
|
|
"loss": 0.886,
|
|
"mean_token_accuracy": 0.7720102846622467,
|
|
"num_tokens": 418347850.0,
|
|
"step": 4180
|
|
},
|
|
{
|
|
"entropy": 0.8958984375,
|
|
"epoch": 0.5726781931251281,
|
|
"grad_norm": 0.39910204497986207,
|
|
"learning_rate": 3.6793715654501903e-06,
|
|
"loss": 0.8877,
|
|
"mean_token_accuracy": 0.7696331143379211,
|
|
"num_tokens": 419363657.0,
|
|
"step": 4190
|
|
},
|
|
{
|
|
"entropy": 0.90390625,
|
|
"epoch": 0.574044966855737,
|
|
"grad_norm": 0.29413958236771154,
|
|
"learning_rate": 3.6758489502606735e-06,
|
|
"loss": 0.9011,
|
|
"mean_token_accuracy": 0.7686240643262863,
|
|
"num_tokens": 420407370.0,
|
|
"step": 4200
|
|
},
|
|
{
|
|
"entropy": 0.91328125,
|
|
"epoch": 0.5754117405863459,
|
|
"grad_norm": 0.3477957397091295,
|
|
"learning_rate": 3.672326335071157e-06,
|
|
"loss": 0.9029,
|
|
"mean_token_accuracy": 0.7681824207305908,
|
|
"num_tokens": 421382440.0,
|
|
"step": 4210
|
|
},
|
|
{
|
|
"entropy": 0.906640625,
|
|
"epoch": 0.5767785143169548,
|
|
"grad_norm": 0.2614502551675182,
|
|
"learning_rate": 3.6688037198816406e-06,
|
|
"loss": 0.9063,
|
|
"mean_token_accuracy": 0.7684712320566177,
|
|
"num_tokens": 422362775.0,
|
|
"step": 4220
|
|
},
|
|
{
|
|
"entropy": 0.88359375,
|
|
"epoch": 0.5781452880475637,
|
|
"grad_norm": 0.3071338589628598,
|
|
"learning_rate": 3.6652811046921237e-06,
|
|
"loss": 0.8849,
|
|
"mean_token_accuracy": 0.7722552746534348,
|
|
"num_tokens": 423331211.0,
|
|
"step": 4230
|
|
},
|
|
{
|
|
"entropy": 0.8986328125,
|
|
"epoch": 0.5795120617781726,
|
|
"grad_norm": 0.2667085868192701,
|
|
"learning_rate": 3.6617584895026072e-06,
|
|
"loss": 0.8968,
|
|
"mean_token_accuracy": 0.7682283788919448,
|
|
"num_tokens": 424291289.0,
|
|
"step": 4240
|
|
},
|
|
{
|
|
"entropy": 0.9244140625,
|
|
"epoch": 0.5808788355087815,
|
|
"grad_norm": 0.30841438935203924,
|
|
"learning_rate": 3.6582358743130904e-06,
|
|
"loss": 0.9189,
|
|
"mean_token_accuracy": 0.7605726987123489,
|
|
"num_tokens": 425304325.0,
|
|
"step": 4250
|
|
},
|
|
{
|
|
"entropy": 0.9072265625,
|
|
"epoch": 0.5822456092393904,
|
|
"grad_norm": 0.29460438039357223,
|
|
"learning_rate": 3.6547132591235735e-06,
|
|
"loss": 0.9015,
|
|
"mean_token_accuracy": 0.7671357303857803,
|
|
"num_tokens": 426270499.0,
|
|
"step": 4260
|
|
},
|
|
{
|
|
"entropy": 0.932421875,
|
|
"epoch": 0.5836123829699993,
|
|
"grad_norm": 0.3615508225448458,
|
|
"learning_rate": 3.651190643934057e-06,
|
|
"loss": 0.9337,
|
|
"mean_token_accuracy": 0.7631444364786149,
|
|
"num_tokens": 427187011.0,
|
|
"step": 4270
|
|
},
|
|
{
|
|
"entropy": 0.908984375,
|
|
"epoch": 0.5849791567006082,
|
|
"grad_norm": 0.28126709574922826,
|
|
"learning_rate": 3.6476680287445406e-06,
|
|
"loss": 0.9014,
|
|
"mean_token_accuracy": 0.7692148476839066,
|
|
"num_tokens": 428218892.0,
|
|
"step": 4280
|
|
},
|
|
{
|
|
"entropy": 0.8857421875,
|
|
"epoch": 0.5863459304312171,
|
|
"grad_norm": 0.2667841518895802,
|
|
"learning_rate": 3.6441454135550237e-06,
|
|
"loss": 0.8747,
|
|
"mean_token_accuracy": 0.7751999348402023,
|
|
"num_tokens": 429237497.0,
|
|
"step": 4290
|
|
},
|
|
{
|
|
"entropy": 0.89765625,
|
|
"epoch": 0.587712704161826,
|
|
"grad_norm": 0.3423865252808248,
|
|
"learning_rate": 3.640622798365507e-06,
|
|
"loss": 0.8964,
|
|
"mean_token_accuracy": 0.7692080080509186,
|
|
"num_tokens": 430237115.0,
|
|
"step": 4300
|
|
},
|
|
{
|
|
"entropy": 0.9326171875,
|
|
"epoch": 0.5890794778924349,
|
|
"grad_norm": 0.30646886938461976,
|
|
"learning_rate": 3.63710018317599e-06,
|
|
"loss": 0.9222,
|
|
"mean_token_accuracy": 0.7635771423578263,
|
|
"num_tokens": 431241122.0,
|
|
"step": 4310
|
|
},
|
|
{
|
|
"entropy": 0.951953125,
|
|
"epoch": 0.5904462516230438,
|
|
"grad_norm": 0.38682227558925103,
|
|
"learning_rate": 3.633577567986473e-06,
|
|
"loss": 0.9521,
|
|
"mean_token_accuracy": 0.7604047179222106,
|
|
"num_tokens": 432221995.0,
|
|
"step": 4320
|
|
},
|
|
{
|
|
"entropy": 0.9025390625,
|
|
"epoch": 0.5918130253536527,
|
|
"grad_norm": 0.2833755327787585,
|
|
"learning_rate": 3.630054952796957e-06,
|
|
"loss": 0.896,
|
|
"mean_token_accuracy": 0.7701098203659058,
|
|
"num_tokens": 433198674.0,
|
|
"step": 4330
|
|
},
|
|
{
|
|
"entropy": 0.9388671875,
|
|
"epoch": 0.5931797990842615,
|
|
"grad_norm": 0.3385504958628347,
|
|
"learning_rate": 3.62653233760744e-06,
|
|
"loss": 0.9301,
|
|
"mean_token_accuracy": 0.7613330453634262,
|
|
"num_tokens": 434197222.0,
|
|
"step": 4340
|
|
},
|
|
{
|
|
"entropy": 0.89765625,
|
|
"epoch": 0.5945465728148704,
|
|
"grad_norm": 0.30523611037446635,
|
|
"learning_rate": 3.6230097224179233e-06,
|
|
"loss": 0.8872,
|
|
"mean_token_accuracy": 0.7715286254882813,
|
|
"num_tokens": 435080429.0,
|
|
"step": 4350
|
|
},
|
|
{
|
|
"entropy": 0.9091796875,
|
|
"epoch": 0.5959133465454794,
|
|
"grad_norm": 0.32500079340332505,
|
|
"learning_rate": 3.6194871072284064e-06,
|
|
"loss": 0.9053,
|
|
"mean_token_accuracy": 0.7679129123687745,
|
|
"num_tokens": 436104149.0,
|
|
"step": 4360
|
|
},
|
|
{
|
|
"entropy": 0.9150390625,
|
|
"epoch": 0.5972801202760883,
|
|
"grad_norm": 0.31995470104490037,
|
|
"learning_rate": 3.61596449203889e-06,
|
|
"loss": 0.9043,
|
|
"mean_token_accuracy": 0.7682949662208557,
|
|
"num_tokens": 437087451.0,
|
|
"step": 4370
|
|
},
|
|
{
|
|
"entropy": 0.8619140625,
|
|
"epoch": 0.5986468940066972,
|
|
"grad_norm": 0.2703236700858406,
|
|
"learning_rate": 3.6124418768493735e-06,
|
|
"loss": 0.8521,
|
|
"mean_token_accuracy": 0.7781327456235886,
|
|
"num_tokens": 438073519.0,
|
|
"step": 4380
|
|
},
|
|
{
|
|
"entropy": 0.9083984375,
|
|
"epoch": 0.6000136677373061,
|
|
"grad_norm": 0.254975069242215,
|
|
"learning_rate": 3.6089192616598566e-06,
|
|
"loss": 0.91,
|
|
"mean_token_accuracy": 0.7687057435512543,
|
|
"num_tokens": 439150829.0,
|
|
"step": 4390
|
|
},
|
|
{
|
|
"entropy": 0.916796875,
|
|
"epoch": 0.601380441467915,
|
|
"grad_norm": 0.31017213848318087,
|
|
"learning_rate": 3.6053966464703398e-06,
|
|
"loss": 0.9215,
|
|
"mean_token_accuracy": 0.7649414807558059,
|
|
"num_tokens": 440124792.0,
|
|
"step": 4400
|
|
},
|
|
{
|
|
"entropy": 0.9408203125,
|
|
"epoch": 0.6027472151985239,
|
|
"grad_norm": 0.29831249633183327,
|
|
"learning_rate": 3.6018740312808233e-06,
|
|
"loss": 0.9414,
|
|
"mean_token_accuracy": 0.7585596114397049,
|
|
"num_tokens": 441178545.0,
|
|
"step": 4410
|
|
},
|
|
{
|
|
"entropy": 0.9005859375,
|
|
"epoch": 0.6041139889291328,
|
|
"grad_norm": 0.33651446027323456,
|
|
"learning_rate": 3.5983514160913064e-06,
|
|
"loss": 0.9012,
|
|
"mean_token_accuracy": 0.7654881864786148,
|
|
"num_tokens": 442210281.0,
|
|
"step": 4420
|
|
},
|
|
{
|
|
"entropy": 0.9056640625,
|
|
"epoch": 0.6054807626597417,
|
|
"grad_norm": 0.26208675240142676,
|
|
"learning_rate": 3.5948288009017895e-06,
|
|
"loss": 0.8964,
|
|
"mean_token_accuracy": 0.7683761119842529,
|
|
"num_tokens": 443225259.0,
|
|
"step": 4430
|
|
},
|
|
{
|
|
"entropy": 0.911328125,
|
|
"epoch": 0.6068475363903506,
|
|
"grad_norm": 0.27189537243143297,
|
|
"learning_rate": 3.591306185712273e-06,
|
|
"loss": 0.9015,
|
|
"mean_token_accuracy": 0.7685211420059204,
|
|
"num_tokens": 444220229.0,
|
|
"step": 4440
|
|
},
|
|
{
|
|
"entropy": 0.9224609375,
|
|
"epoch": 0.6082143101209595,
|
|
"grad_norm": 2.775916814448488,
|
|
"learning_rate": 3.5877835705227566e-06,
|
|
"loss": 0.9216,
|
|
"mean_token_accuracy": 0.7649797767400741,
|
|
"num_tokens": 445201473.0,
|
|
"step": 4450
|
|
},
|
|
{
|
|
"entropy": 0.87109375,
|
|
"epoch": 0.6095810838515684,
|
|
"grad_norm": 3.1802313655093677,
|
|
"learning_rate": 3.5842609553332398e-06,
|
|
"loss": 0.8707,
|
|
"mean_token_accuracy": 0.7746197015047074,
|
|
"num_tokens": 446200834.0,
|
|
"step": 4460
|
|
},
|
|
{
|
|
"entropy": 0.9068359375,
|
|
"epoch": 0.6109478575821773,
|
|
"grad_norm": 16.031071657674648,
|
|
"learning_rate": 3.580738340143723e-06,
|
|
"loss": 0.9096,
|
|
"mean_token_accuracy": 0.7673165440559387,
|
|
"num_tokens": 447203460.0,
|
|
"step": 4470
|
|
},
|
|
{
|
|
"entropy": 0.89765625,
|
|
"epoch": 0.6123146313127862,
|
|
"grad_norm": 0.866229238132588,
|
|
"learning_rate": 3.577215724954206e-06,
|
|
"loss": 0.8995,
|
|
"mean_token_accuracy": 0.7701589554548264,
|
|
"num_tokens": 448231820.0,
|
|
"step": 4480
|
|
},
|
|
{
|
|
"entropy": 0.9361328125,
|
|
"epoch": 0.6136814050433951,
|
|
"grad_norm": 0.30994226793830304,
|
|
"learning_rate": 3.57369310976469e-06,
|
|
"loss": 0.9358,
|
|
"mean_token_accuracy": 0.7607106119394302,
|
|
"num_tokens": 449181166.0,
|
|
"step": 4490
|
|
},
|
|
{
|
|
"entropy": 0.935546875,
|
|
"epoch": 0.615048178774004,
|
|
"grad_norm": 0.29869925611711307,
|
|
"learning_rate": 3.570170494575173e-06,
|
|
"loss": 0.9232,
|
|
"mean_token_accuracy": 0.7668235749006271,
|
|
"num_tokens": 450200977.0,
|
|
"step": 4500
|
|
},
|
|
{
|
|
"entropy": 0.9578125,
|
|
"epoch": 0.6164149525046129,
|
|
"grad_norm": 0.28939295482292404,
|
|
"learning_rate": 3.5666478793856562e-06,
|
|
"loss": 0.9687,
|
|
"mean_token_accuracy": 0.7549959182739258,
|
|
"num_tokens": 451176344.0,
|
|
"step": 4510
|
|
},
|
|
{
|
|
"entropy": 0.88515625,
|
|
"epoch": 0.6177817262352218,
|
|
"grad_norm": 0.3232409252527902,
|
|
"learning_rate": 3.5631252641961394e-06,
|
|
"loss": 0.8826,
|
|
"mean_token_accuracy": 0.7701918840408325,
|
|
"num_tokens": 452144687.0,
|
|
"step": 4520
|
|
},
|
|
{
|
|
"entropy": 0.8513671875,
|
|
"epoch": 0.6191484999658307,
|
|
"grad_norm": 0.28051862459568405,
|
|
"learning_rate": 3.5596026490066225e-06,
|
|
"loss": 0.852,
|
|
"mean_token_accuracy": 0.7782174438238144,
|
|
"num_tokens": 453172733.0,
|
|
"step": 4530
|
|
},
|
|
{
|
|
"entropy": 0.89453125,
|
|
"epoch": 0.6205152736964396,
|
|
"grad_norm": 0.2949264071309678,
|
|
"learning_rate": 3.556080033817106e-06,
|
|
"loss": 0.8986,
|
|
"mean_token_accuracy": 0.7691696524620056,
|
|
"num_tokens": 454139460.0,
|
|
"step": 4540
|
|
},
|
|
{
|
|
"entropy": 0.9453125,
|
|
"epoch": 0.6218820474270484,
|
|
"grad_norm": 0.3260401350016006,
|
|
"learning_rate": 3.5525574186275896e-06,
|
|
"loss": 0.9349,
|
|
"mean_token_accuracy": 0.7613766759634018,
|
|
"num_tokens": 455099585.0,
|
|
"step": 4550
|
|
},
|
|
{
|
|
"entropy": 0.940625,
|
|
"epoch": 0.6232488211576573,
|
|
"grad_norm": 0.31850130522685177,
|
|
"learning_rate": 3.5490348034380727e-06,
|
|
"loss": 0.9479,
|
|
"mean_token_accuracy": 0.7600121021270752,
|
|
"num_tokens": 456115164.0,
|
|
"step": 4560
|
|
},
|
|
{
|
|
"entropy": 0.9119140625,
|
|
"epoch": 0.6246155948882662,
|
|
"grad_norm": 0.2722759445746692,
|
|
"learning_rate": 3.545512188248556e-06,
|
|
"loss": 0.9096,
|
|
"mean_token_accuracy": 0.7654687970876694,
|
|
"num_tokens": 457102641.0,
|
|
"step": 4570
|
|
},
|
|
{
|
|
"entropy": 0.8783203125,
|
|
"epoch": 0.6259823686188751,
|
|
"grad_norm": 0.2794440923000885,
|
|
"learning_rate": 3.5419895730590394e-06,
|
|
"loss": 0.8797,
|
|
"mean_token_accuracy": 0.7714338034391404,
|
|
"num_tokens": 458063569.0,
|
|
"step": 4580
|
|
},
|
|
{
|
|
"entropy": 0.9384765625,
|
|
"epoch": 0.627349142349484,
|
|
"grad_norm": 0.2971331483338024,
|
|
"learning_rate": 3.5384669578695225e-06,
|
|
"loss": 0.9378,
|
|
"mean_token_accuracy": 0.7623136460781097,
|
|
"num_tokens": 459050393.0,
|
|
"step": 4590
|
|
},
|
|
{
|
|
"entropy": 0.9296875,
|
|
"epoch": 0.6287159160800929,
|
|
"grad_norm": 0.4270718868701037,
|
|
"learning_rate": 3.534944342680006e-06,
|
|
"loss": 0.9241,
|
|
"mean_token_accuracy": 0.7643370240926742,
|
|
"num_tokens": 460080370.0,
|
|
"step": 4600
|
|
},
|
|
{
|
|
"entropy": 0.889453125,
|
|
"epoch": 0.6300826898107018,
|
|
"grad_norm": 0.32710052863565214,
|
|
"learning_rate": 3.5314217274904896e-06,
|
|
"loss": 0.889,
|
|
"mean_token_accuracy": 0.7699246048927307,
|
|
"num_tokens": 461083660.0,
|
|
"step": 4610
|
|
},
|
|
{
|
|
"entropy": 0.9087890625,
|
|
"epoch": 0.6314494635413107,
|
|
"grad_norm": 0.2751196329027983,
|
|
"learning_rate": 3.5278991123009727e-06,
|
|
"loss": 0.9117,
|
|
"mean_token_accuracy": 0.7666348785161972,
|
|
"num_tokens": 462065871.0,
|
|
"step": 4620
|
|
},
|
|
{
|
|
"entropy": 0.9169921875,
|
|
"epoch": 0.6328162372719196,
|
|
"grad_norm": 0.33942406493120014,
|
|
"learning_rate": 3.524376497111456e-06,
|
|
"loss": 0.9099,
|
|
"mean_token_accuracy": 0.7672272711992264,
|
|
"num_tokens": 463080086.0,
|
|
"step": 4630
|
|
},
|
|
{
|
|
"entropy": 0.8982421875,
|
|
"epoch": 0.6341830110025285,
|
|
"grad_norm": 0.8599564300873088,
|
|
"learning_rate": 3.520853881921939e-06,
|
|
"loss": 0.8902,
|
|
"mean_token_accuracy": 0.7701105594635009,
|
|
"num_tokens": 464075969.0,
|
|
"step": 4640
|
|
},
|
|
{
|
|
"entropy": 0.8955078125,
|
|
"epoch": 0.6355497847331374,
|
|
"grad_norm": 0.2833167278921308,
|
|
"learning_rate": 3.517331266732422e-06,
|
|
"loss": 0.8925,
|
|
"mean_token_accuracy": 0.7710114240646362,
|
|
"num_tokens": 465113877.0,
|
|
"step": 4650
|
|
},
|
|
{
|
|
"entropy": 0.94453125,
|
|
"epoch": 0.6369165584637463,
|
|
"grad_norm": 0.3036089557019047,
|
|
"learning_rate": 3.513808651542906e-06,
|
|
"loss": 0.9467,
|
|
"mean_token_accuracy": 0.7605886608362198,
|
|
"num_tokens": 466103248.0,
|
|
"step": 4660
|
|
},
|
|
{
|
|
"entropy": 0.90546875,
|
|
"epoch": 0.6382833321943552,
|
|
"grad_norm": 0.31303460363540697,
|
|
"learning_rate": 3.510286036353389e-06,
|
|
"loss": 0.8932,
|
|
"mean_token_accuracy": 0.7687899887561798,
|
|
"num_tokens": 467036111.0,
|
|
"step": 4670
|
|
},
|
|
{
|
|
"entropy": 0.8990234375,
|
|
"epoch": 0.6396501059249641,
|
|
"grad_norm": 0.24503613290000897,
|
|
"learning_rate": 3.5067634211638723e-06,
|
|
"loss": 0.8967,
|
|
"mean_token_accuracy": 0.7681892305612564,
|
|
"num_tokens": 468077688.0,
|
|
"step": 4680
|
|
},
|
|
{
|
|
"entropy": 0.901953125,
|
|
"epoch": 0.641016879655573,
|
|
"grad_norm": 0.2787906302995673,
|
|
"learning_rate": 3.5032408059743554e-06,
|
|
"loss": 0.9058,
|
|
"mean_token_accuracy": 0.7673250615596772,
|
|
"num_tokens": 469081863.0,
|
|
"step": 4690
|
|
},
|
|
{
|
|
"entropy": 0.8951171875,
|
|
"epoch": 0.6423836533861819,
|
|
"grad_norm": 0.3248292720296991,
|
|
"learning_rate": 3.499718190784839e-06,
|
|
"loss": 0.8889,
|
|
"mean_token_accuracy": 0.7698028743267059,
|
|
"num_tokens": 470036745.0,
|
|
"step": 4700
|
|
},
|
|
{
|
|
"entropy": 0.9044921875,
|
|
"epoch": 0.6437504271167909,
|
|
"grad_norm": 0.2748949929644807,
|
|
"learning_rate": 3.4961955755953225e-06,
|
|
"loss": 0.8996,
|
|
"mean_token_accuracy": 0.7680906146764755,
|
|
"num_tokens": 471040901.0,
|
|
"step": 4710
|
|
},
|
|
{
|
|
"entropy": 0.91484375,
|
|
"epoch": 0.6451172008473998,
|
|
"grad_norm": 0.2751829942116833,
|
|
"learning_rate": 3.4926729604058056e-06,
|
|
"loss": 0.9139,
|
|
"mean_token_accuracy": 0.7662262111902237,
|
|
"num_tokens": 472034268.0,
|
|
"step": 4720
|
|
},
|
|
{
|
|
"entropy": 0.891796875,
|
|
"epoch": 0.6464839745780087,
|
|
"grad_norm": 0.33730104925751603,
|
|
"learning_rate": 3.4891503452162888e-06,
|
|
"loss": 0.8773,
|
|
"mean_token_accuracy": 0.7730095356702804,
|
|
"num_tokens": 472986643.0,
|
|
"step": 4730
|
|
},
|
|
{
|
|
"entropy": 0.882421875,
|
|
"epoch": 0.6478507483086176,
|
|
"grad_norm": 0.299039912364961,
|
|
"learning_rate": 3.4856277300267723e-06,
|
|
"loss": 0.8744,
|
|
"mean_token_accuracy": 0.7729396462440491,
|
|
"num_tokens": 473958882.0,
|
|
"step": 4740
|
|
},
|
|
{
|
|
"entropy": 0.9015625,
|
|
"epoch": 0.6492175220392264,
|
|
"grad_norm": 0.29291141718024755,
|
|
"learning_rate": 3.4821051148372554e-06,
|
|
"loss": 0.8997,
|
|
"mean_token_accuracy": 0.7674183160066604,
|
|
"num_tokens": 474928079.0,
|
|
"step": 4750
|
|
},
|
|
{
|
|
"entropy": 0.8435546875,
|
|
"epoch": 0.6505842957698353,
|
|
"grad_norm": 0.31833853248498367,
|
|
"learning_rate": 3.4785824996477386e-06,
|
|
"loss": 0.826,
|
|
"mean_token_accuracy": 0.7840016961097718,
|
|
"num_tokens": 475886095.0,
|
|
"step": 4760
|
|
},
|
|
{
|
|
"entropy": 0.833984375,
|
|
"epoch": 0.6519510695004442,
|
|
"grad_norm": 0.2995536988667641,
|
|
"learning_rate": 3.475059884458222e-06,
|
|
"loss": 0.8368,
|
|
"mean_token_accuracy": 0.7831269353628159,
|
|
"num_tokens": 476869328.0,
|
|
"step": 4770
|
|
},
|
|
{
|
|
"entropy": 0.878125,
|
|
"epoch": 0.6533178432310531,
|
|
"grad_norm": 0.3005905238969249,
|
|
"learning_rate": 3.4715372692687057e-06,
|
|
"loss": 0.8798,
|
|
"mean_token_accuracy": 0.7719609051942825,
|
|
"num_tokens": 477821773.0,
|
|
"step": 4780
|
|
},
|
|
{
|
|
"entropy": 0.922265625,
|
|
"epoch": 0.654684616961662,
|
|
"grad_norm": 0.3791750743328245,
|
|
"learning_rate": 3.4680146540791888e-06,
|
|
"loss": 0.9161,
|
|
"mean_token_accuracy": 0.7652972429990769,
|
|
"num_tokens": 478842083.0,
|
|
"step": 4790
|
|
},
|
|
{
|
|
"entropy": 0.91171875,
|
|
"epoch": 0.6560513906922709,
|
|
"grad_norm": 0.30694631975732867,
|
|
"learning_rate": 3.464492038889672e-06,
|
|
"loss": 0.9025,
|
|
"mean_token_accuracy": 0.7687332928180695,
|
|
"num_tokens": 479784211.0,
|
|
"step": 4800
|
|
},
|
|
{
|
|
"entropy": 0.90625,
|
|
"epoch": 0.6574181644228798,
|
|
"grad_norm": 0.2904844670723624,
|
|
"learning_rate": 3.460969423700155e-06,
|
|
"loss": 0.9146,
|
|
"mean_token_accuracy": 0.7642778187990189,
|
|
"num_tokens": 480833427.0,
|
|
"step": 4810
|
|
},
|
|
{
|
|
"entropy": 0.9076171875,
|
|
"epoch": 0.6587849381534887,
|
|
"grad_norm": 0.27626699593098836,
|
|
"learning_rate": 3.457446808510639e-06,
|
|
"loss": 0.8945,
|
|
"mean_token_accuracy": 0.7705834239721299,
|
|
"num_tokens": 481851310.0,
|
|
"step": 4820
|
|
},
|
|
{
|
|
"entropy": 0.8814453125,
|
|
"epoch": 0.6601517118840976,
|
|
"grad_norm": 0.27533018112900565,
|
|
"learning_rate": 3.453924193321122e-06,
|
|
"loss": 0.8846,
|
|
"mean_token_accuracy": 0.7698549866676331,
|
|
"num_tokens": 482891637.0,
|
|
"step": 4830
|
|
},
|
|
{
|
|
"entropy": 0.8763671875,
|
|
"epoch": 0.6615184856147065,
|
|
"grad_norm": 0.319439908941215,
|
|
"learning_rate": 3.4504015781316052e-06,
|
|
"loss": 0.8616,
|
|
"mean_token_accuracy": 0.7771168977022171,
|
|
"num_tokens": 483921621.0,
|
|
"step": 4840
|
|
},
|
|
{
|
|
"entropy": 0.9111328125,
|
|
"epoch": 0.6628852593453154,
|
|
"grad_norm": 0.2861501404232019,
|
|
"learning_rate": 3.4468789629420884e-06,
|
|
"loss": 0.9182,
|
|
"mean_token_accuracy": 0.7645470827817917,
|
|
"num_tokens": 484982954.0,
|
|
"step": 4850
|
|
},
|
|
{
|
|
"entropy": 0.8634765625,
|
|
"epoch": 0.6642520330759243,
|
|
"grad_norm": 0.2970394458074311,
|
|
"learning_rate": 3.4433563477525715e-06,
|
|
"loss": 0.8678,
|
|
"mean_token_accuracy": 0.7754582941532135,
|
|
"num_tokens": 485972247.0,
|
|
"step": 4860
|
|
},
|
|
{
|
|
"entropy": 0.9013671875,
|
|
"epoch": 0.6656188068065332,
|
|
"grad_norm": 0.31576572390017976,
|
|
"learning_rate": 3.439833732563055e-06,
|
|
"loss": 0.8919,
|
|
"mean_token_accuracy": 0.771066015958786,
|
|
"num_tokens": 486961829.0,
|
|
"step": 4870
|
|
},
|
|
{
|
|
"entropy": 0.8974609375,
|
|
"epoch": 0.6669855805371421,
|
|
"grad_norm": 0.3177460219627154,
|
|
"learning_rate": 3.4363111173735386e-06,
|
|
"loss": 0.8989,
|
|
"mean_token_accuracy": 0.7682592451572419,
|
|
"num_tokens": 487999991.0,
|
|
"step": 4880
|
|
},
|
|
{
|
|
"entropy": 0.9048828125,
|
|
"epoch": 0.668352354267751,
|
|
"grad_norm": 0.2787825779581879,
|
|
"learning_rate": 3.4327885021840217e-06,
|
|
"loss": 0.9042,
|
|
"mean_token_accuracy": 0.7682960718870163,
|
|
"num_tokens": 489015250.0,
|
|
"step": 4890
|
|
},
|
|
{
|
|
"entropy": 0.89921875,
|
|
"epoch": 0.6697191279983599,
|
|
"grad_norm": 0.2965702578292428,
|
|
"learning_rate": 3.429265886994505e-06,
|
|
"loss": 0.9003,
|
|
"mean_token_accuracy": 0.767874813079834,
|
|
"num_tokens": 490060131.0,
|
|
"step": 4900
|
|
},
|
|
{
|
|
"entropy": 0.8865234375,
|
|
"epoch": 0.6710859017289688,
|
|
"grad_norm": 0.28961278604573243,
|
|
"learning_rate": 3.4257432718049884e-06,
|
|
"loss": 0.8783,
|
|
"mean_token_accuracy": 0.7709373533725739,
|
|
"num_tokens": 491082366.0,
|
|
"step": 4910
|
|
},
|
|
{
|
|
"entropy": 0.8833984375,
|
|
"epoch": 0.6724526754595777,
|
|
"grad_norm": 0.28953415326788107,
|
|
"learning_rate": 3.4222206566154715e-06,
|
|
"loss": 0.8893,
|
|
"mean_token_accuracy": 0.7723812609910965,
|
|
"num_tokens": 492125674.0,
|
|
"step": 4920
|
|
},
|
|
{
|
|
"entropy": 0.926953125,
|
|
"epoch": 0.6738194491901865,
|
|
"grad_norm": 0.2916456840716115,
|
|
"learning_rate": 3.418698041425955e-06,
|
|
"loss": 0.9228,
|
|
"mean_token_accuracy": 0.7634715765714646,
|
|
"num_tokens": 493112088.0,
|
|
"step": 4930
|
|
},
|
|
{
|
|
"entropy": 0.8724609375,
|
|
"epoch": 0.6751862229207954,
|
|
"grad_norm": 0.27315131025866163,
|
|
"learning_rate": 3.415175426236438e-06,
|
|
"loss": 0.8694,
|
|
"mean_token_accuracy": 0.774756345152855,
|
|
"num_tokens": 494137107.0,
|
|
"step": 4940
|
|
},
|
|
{
|
|
"entropy": 0.8533203125,
|
|
"epoch": 0.6765529966514043,
|
|
"grad_norm": 0.31419422562342636,
|
|
"learning_rate": 3.4116528110469217e-06,
|
|
"loss": 0.8465,
|
|
"mean_token_accuracy": 0.7807996809482575,
|
|
"num_tokens": 495128504.0,
|
|
"step": 4950
|
|
},
|
|
{
|
|
"entropy": 0.8921875,
|
|
"epoch": 0.6779197703820132,
|
|
"grad_norm": 0.31727489482610804,
|
|
"learning_rate": 3.408130195857405e-06,
|
|
"loss": 0.8881,
|
|
"mean_token_accuracy": 0.7709563463926316,
|
|
"num_tokens": 496080043.0,
|
|
"step": 4960
|
|
},
|
|
{
|
|
"entropy": 0.9046875,
|
|
"epoch": 0.6792865441126221,
|
|
"grad_norm": 0.30985943823751066,
|
|
"learning_rate": 3.404607580667888e-06,
|
|
"loss": 0.8968,
|
|
"mean_token_accuracy": 0.7692039310932159,
|
|
"num_tokens": 497081232.0,
|
|
"step": 4970
|
|
},
|
|
{
|
|
"entropy": 0.8642578125,
|
|
"epoch": 0.680653317843231,
|
|
"grad_norm": 0.2872815005456844,
|
|
"learning_rate": 3.401084965478371e-06,
|
|
"loss": 0.8651,
|
|
"mean_token_accuracy": 0.7774709224700928,
|
|
"num_tokens": 498057583.0,
|
|
"step": 4980
|
|
},
|
|
{
|
|
"entropy": 0.89765625,
|
|
"epoch": 0.6820200915738399,
|
|
"grad_norm": 0.2770748894750208,
|
|
"learning_rate": 3.397562350288855e-06,
|
|
"loss": 0.8976,
|
|
"mean_token_accuracy": 0.7702643543481826,
|
|
"num_tokens": 499058823.0,
|
|
"step": 4990
|
|
},
|
|
{
|
|
"entropy": 0.8732421875,
|
|
"epoch": 0.6833868653044488,
|
|
"grad_norm": 0.3218350325191329,
|
|
"learning_rate": 3.394039735099338e-06,
|
|
"loss": 0.8707,
|
|
"mean_token_accuracy": 0.7749590545892715,
|
|
"num_tokens": 500054628.0,
|
|
"step": 5000
|
|
},
|
|
{
|
|
"entropy": 0.8583984375,
|
|
"epoch": 0.6847536390350577,
|
|
"grad_norm": 0.29734999861706474,
|
|
"learning_rate": 3.3905171199098213e-06,
|
|
"loss": 0.8581,
|
|
"mean_token_accuracy": 0.7765612661838531,
|
|
"num_tokens": 501048187.0,
|
|
"step": 5010
|
|
},
|
|
{
|
|
"entropy": 0.88671875,
|
|
"epoch": 0.6861204127656666,
|
|
"grad_norm": 0.28714407060523606,
|
|
"learning_rate": 3.3869945047203044e-06,
|
|
"loss": 0.8918,
|
|
"mean_token_accuracy": 0.7696502655744553,
|
|
"num_tokens": 502062268.0,
|
|
"step": 5020
|
|
},
|
|
{
|
|
"entropy": 0.8833984375,
|
|
"epoch": 0.6874871864962755,
|
|
"grad_norm": 0.3403137642953521,
|
|
"learning_rate": 3.3834718895307876e-06,
|
|
"loss": 0.87,
|
|
"mean_token_accuracy": 0.772771355509758,
|
|
"num_tokens": 503054152.0,
|
|
"step": 5030
|
|
},
|
|
{
|
|
"entropy": 0.871484375,
|
|
"epoch": 0.6888539602268844,
|
|
"grad_norm": 0.2566366445463334,
|
|
"learning_rate": 3.3799492743412715e-06,
|
|
"loss": 0.8735,
|
|
"mean_token_accuracy": 0.7745592027902604,
|
|
"num_tokens": 504061542.0,
|
|
"step": 5040
|
|
},
|
|
{
|
|
"entropy": 0.8900390625,
|
|
"epoch": 0.6902207339574933,
|
|
"grad_norm": 0.7077332176071757,
|
|
"learning_rate": 3.3764266591517547e-06,
|
|
"loss": 0.886,
|
|
"mean_token_accuracy": 0.7707038909196854,
|
|
"num_tokens": 505071467.0,
|
|
"step": 5050
|
|
},
|
|
{
|
|
"entropy": 0.88125,
|
|
"epoch": 0.6915875076881023,
|
|
"grad_norm": 0.4293674639986575,
|
|
"learning_rate": 3.3729040439622378e-06,
|
|
"loss": 0.8739,
|
|
"mean_token_accuracy": 0.7725519150495529,
|
|
"num_tokens": 506090809.0,
|
|
"step": 5060
|
|
},
|
|
{
|
|
"entropy": 0.90390625,
|
|
"epoch": 0.6929542814187112,
|
|
"grad_norm": 0.3161928902658119,
|
|
"learning_rate": 3.3693814287727213e-06,
|
|
"loss": 0.8903,
|
|
"mean_token_accuracy": 0.7690071821212768,
|
|
"num_tokens": 507069713.0,
|
|
"step": 5070
|
|
},
|
|
{
|
|
"entropy": 0.928125,
|
|
"epoch": 0.6943210551493201,
|
|
"grad_norm": 0.30178741873498244,
|
|
"learning_rate": 3.3658588135832044e-06,
|
|
"loss": 0.9238,
|
|
"mean_token_accuracy": 0.7646104663610458,
|
|
"num_tokens": 508131315.0,
|
|
"step": 5080
|
|
},
|
|
{
|
|
"entropy": 0.8775390625,
|
|
"epoch": 0.695687828879929,
|
|
"grad_norm": 0.2897122635569694,
|
|
"learning_rate": 3.3623361983936876e-06,
|
|
"loss": 0.8741,
|
|
"mean_token_accuracy": 0.7746311366558075,
|
|
"num_tokens": 509194218.0,
|
|
"step": 5090
|
|
},
|
|
{
|
|
"entropy": 0.8873046875,
|
|
"epoch": 0.6970546026105379,
|
|
"grad_norm": 0.26408480641221255,
|
|
"learning_rate": 3.358813583204171e-06,
|
|
"loss": 0.8888,
|
|
"mean_token_accuracy": 0.7705213755369187,
|
|
"num_tokens": 510256059.0,
|
|
"step": 5100
|
|
},
|
|
{
|
|
"entropy": 0.934375,
|
|
"epoch": 0.6984213763411468,
|
|
"grad_norm": 0.32604663231348857,
|
|
"learning_rate": 3.3552909680146547e-06,
|
|
"loss": 0.9238,
|
|
"mean_token_accuracy": 0.7632334351539611,
|
|
"num_tokens": 511260466.0,
|
|
"step": 5110
|
|
},
|
|
{
|
|
"entropy": 0.8671875,
|
|
"epoch": 0.6997881500717557,
|
|
"grad_norm": 0.30926435730578455,
|
|
"learning_rate": 3.3517683528251378e-06,
|
|
"loss": 0.8651,
|
|
"mean_token_accuracy": 0.7751440852880478,
|
|
"num_tokens": 512259871.0,
|
|
"step": 5120
|
|
},
|
|
{
|
|
"entropy": 0.871484375,
|
|
"epoch": 0.7011549238023645,
|
|
"grad_norm": 0.3114384579081125,
|
|
"learning_rate": 3.348245737635621e-06,
|
|
"loss": 0.8648,
|
|
"mean_token_accuracy": 0.775934636592865,
|
|
"num_tokens": 513284332.0,
|
|
"step": 5130
|
|
},
|
|
{
|
|
"entropy": 0.89609375,
|
|
"epoch": 0.7025216975329734,
|
|
"grad_norm": 0.26525095790383235,
|
|
"learning_rate": 3.344723122446104e-06,
|
|
"loss": 0.8876,
|
|
"mean_token_accuracy": 0.7694653332233429,
|
|
"num_tokens": 514253926.0,
|
|
"step": 5140
|
|
},
|
|
{
|
|
"entropy": 0.9177734375,
|
|
"epoch": 0.7038884712635823,
|
|
"grad_norm": 0.27539918682120945,
|
|
"learning_rate": 3.341200507256587e-06,
|
|
"loss": 0.9099,
|
|
"mean_token_accuracy": 0.7672122329473495,
|
|
"num_tokens": 515239960.0,
|
|
"step": 5150
|
|
},
|
|
{
|
|
"entropy": 0.9232421875,
|
|
"epoch": 0.7052552449941912,
|
|
"grad_norm": 0.3019526498341698,
|
|
"learning_rate": 3.337677892067071e-06,
|
|
"loss": 0.9181,
|
|
"mean_token_accuracy": 0.7673907667398453,
|
|
"num_tokens": 516225008.0,
|
|
"step": 5160
|
|
},
|
|
{
|
|
"entropy": 0.9087890625,
|
|
"epoch": 0.7066220187248001,
|
|
"grad_norm": 0.3581462306155906,
|
|
"learning_rate": 3.3341552768775543e-06,
|
|
"loss": 0.9046,
|
|
"mean_token_accuracy": 0.7653254628181457,
|
|
"num_tokens": 517241190.0,
|
|
"step": 5170
|
|
},
|
|
{
|
|
"entropy": 0.9,
|
|
"epoch": 0.707988792455409,
|
|
"grad_norm": 0.31505244515765934,
|
|
"learning_rate": 3.3306326616880374e-06,
|
|
"loss": 0.8996,
|
|
"mean_token_accuracy": 0.7681323975324631,
|
|
"num_tokens": 518251490.0,
|
|
"step": 5180
|
|
},
|
|
{
|
|
"entropy": 0.9021484375,
|
|
"epoch": 0.7093555661860179,
|
|
"grad_norm": 0.2663319337986455,
|
|
"learning_rate": 3.3271100464985205e-06,
|
|
"loss": 0.8912,
|
|
"mean_token_accuracy": 0.7713909447193146,
|
|
"num_tokens": 519272357.0,
|
|
"step": 5190
|
|
},
|
|
{
|
|
"entropy": 0.892578125,
|
|
"epoch": 0.7107223399166268,
|
|
"grad_norm": 0.27226265511062225,
|
|
"learning_rate": 3.323587431309004e-06,
|
|
"loss": 0.8872,
|
|
"mean_token_accuracy": 0.7722367852926254,
|
|
"num_tokens": 520319517.0,
|
|
"step": 5200
|
|
},
|
|
{
|
|
"entropy": 0.910546875,
|
|
"epoch": 0.7120891136472357,
|
|
"grad_norm": 0.3311729278659517,
|
|
"learning_rate": 3.3200648161194876e-06,
|
|
"loss": 0.8927,
|
|
"mean_token_accuracy": 0.7671993255615235,
|
|
"num_tokens": 521333969.0,
|
|
"step": 5210
|
|
},
|
|
{
|
|
"entropy": 0.88984375,
|
|
"epoch": 0.7134558873778446,
|
|
"grad_norm": 0.3196379198600141,
|
|
"learning_rate": 3.3165422009299707e-06,
|
|
"loss": 0.8869,
|
|
"mean_token_accuracy": 0.7705925077199935,
|
|
"num_tokens": 522342219.0,
|
|
"step": 5220
|
|
},
|
|
{
|
|
"entropy": 0.9119140625,
|
|
"epoch": 0.7148226611084535,
|
|
"grad_norm": 0.29110492554004064,
|
|
"learning_rate": 3.313019585740454e-06,
|
|
"loss": 0.9066,
|
|
"mean_token_accuracy": 0.7670833200216294,
|
|
"num_tokens": 523334861.0,
|
|
"step": 5230
|
|
},
|
|
{
|
|
"entropy": 0.912890625,
|
|
"epoch": 0.7161894348390624,
|
|
"grad_norm": 0.30267904538231966,
|
|
"learning_rate": 3.3094969705509374e-06,
|
|
"loss": 0.9128,
|
|
"mean_token_accuracy": 0.7657604962587357,
|
|
"num_tokens": 524347920.0,
|
|
"step": 5240
|
|
},
|
|
{
|
|
"entropy": 0.8888671875,
|
|
"epoch": 0.7175562085696713,
|
|
"grad_norm": 0.2760127655114215,
|
|
"learning_rate": 3.3059743553614205e-06,
|
|
"loss": 0.892,
|
|
"mean_token_accuracy": 0.7710931539535523,
|
|
"num_tokens": 525329939.0,
|
|
"step": 5250
|
|
},
|
|
{
|
|
"entropy": 0.887890625,
|
|
"epoch": 0.7189229823002802,
|
|
"grad_norm": 0.2815853482927409,
|
|
"learning_rate": 3.3024517401719036e-06,
|
|
"loss": 0.8784,
|
|
"mean_token_accuracy": 0.7707245528697968,
|
|
"num_tokens": 526310927.0,
|
|
"step": 5260
|
|
},
|
|
{
|
|
"entropy": 0.909375,
|
|
"epoch": 0.7202897560308891,
|
|
"grad_norm": 0.3322432383703372,
|
|
"learning_rate": 3.298929124982387e-06,
|
|
"loss": 0.9156,
|
|
"mean_token_accuracy": 0.7701619863510132,
|
|
"num_tokens": 527328309.0,
|
|
"step": 5270
|
|
},
|
|
{
|
|
"entropy": 0.9109375,
|
|
"epoch": 0.721656529761498,
|
|
"grad_norm": 0.25406167705154714,
|
|
"learning_rate": 3.2954065097928707e-06,
|
|
"loss": 0.9138,
|
|
"mean_token_accuracy": 0.7667540937662125,
|
|
"num_tokens": 528355417.0,
|
|
"step": 5280
|
|
},
|
|
{
|
|
"entropy": 0.897265625,
|
|
"epoch": 0.7230233034921069,
|
|
"grad_norm": 0.2807643528162766,
|
|
"learning_rate": 3.291883894603354e-06,
|
|
"loss": 0.895,
|
|
"mean_token_accuracy": 0.7695012569427491,
|
|
"num_tokens": 529367701.0,
|
|
"step": 5290
|
|
},
|
|
{
|
|
"entropy": 0.86796875,
|
|
"epoch": 0.7243900772227158,
|
|
"grad_norm": 0.30783971348748573,
|
|
"learning_rate": 3.288361279413837e-06,
|
|
"loss": 0.8622,
|
|
"mean_token_accuracy": 0.7744955480098724,
|
|
"num_tokens": 530358442.0,
|
|
"step": 5300
|
|
},
|
|
{
|
|
"entropy": 0.869140625,
|
|
"epoch": 0.7257568509533247,
|
|
"grad_norm": 0.2775555194258539,
|
|
"learning_rate": 3.28483866422432e-06,
|
|
"loss": 0.8648,
|
|
"mean_token_accuracy": 0.7765936881303788,
|
|
"num_tokens": 531346660.0,
|
|
"step": 5310
|
|
},
|
|
{
|
|
"entropy": 0.9138671875,
|
|
"epoch": 0.7271236246839335,
|
|
"grad_norm": 0.2683814004271425,
|
|
"learning_rate": 3.281316049034804e-06,
|
|
"loss": 0.9167,
|
|
"mean_token_accuracy": 0.7629865437746048,
|
|
"num_tokens": 532403220.0,
|
|
"step": 5320
|
|
},
|
|
{
|
|
"entropy": 0.869921875,
|
|
"epoch": 0.7284903984145424,
|
|
"grad_norm": 0.26786261361687197,
|
|
"learning_rate": 3.277793433845287e-06,
|
|
"loss": 0.8573,
|
|
"mean_token_accuracy": 0.7740849852561951,
|
|
"num_tokens": 533433256.0,
|
|
"step": 5330
|
|
},
|
|
{
|
|
"entropy": 0.8876953125,
|
|
"epoch": 0.7298571721451513,
|
|
"grad_norm": 0.2971326958671774,
|
|
"learning_rate": 3.2742708186557703e-06,
|
|
"loss": 0.8861,
|
|
"mean_token_accuracy": 0.7738748639822006,
|
|
"num_tokens": 534455546.0,
|
|
"step": 5340
|
|
},
|
|
{
|
|
"entropy": 0.89921875,
|
|
"epoch": 0.7312239458757602,
|
|
"grad_norm": 0.3633689183801399,
|
|
"learning_rate": 3.2707482034662534e-06,
|
|
"loss": 0.8901,
|
|
"mean_token_accuracy": 0.7687522798776627,
|
|
"num_tokens": 535465383.0,
|
|
"step": 5350
|
|
},
|
|
{
|
|
"entropy": 0.9037109375,
|
|
"epoch": 0.7325907196063691,
|
|
"grad_norm": 0.29383479806836266,
|
|
"learning_rate": 3.2672255882767366e-06,
|
|
"loss": 0.898,
|
|
"mean_token_accuracy": 0.7710727483034134,
|
|
"num_tokens": 536463324.0,
|
|
"step": 5360
|
|
},
|
|
{
|
|
"entropy": 0.87265625,
|
|
"epoch": 0.733957493336978,
|
|
"grad_norm": 0.24497873097109432,
|
|
"learning_rate": 3.26370297308722e-06,
|
|
"loss": 0.868,
|
|
"mean_token_accuracy": 0.7753186762332916,
|
|
"num_tokens": 537502644.0,
|
|
"step": 5370
|
|
},
|
|
{
|
|
"entropy": 0.878515625,
|
|
"epoch": 0.7353242670675869,
|
|
"grad_norm": 0.31016975382328865,
|
|
"learning_rate": 3.2601803578977037e-06,
|
|
"loss": 0.8782,
|
|
"mean_token_accuracy": 0.7717351198196412,
|
|
"num_tokens": 538501841.0,
|
|
"step": 5380
|
|
},
|
|
{
|
|
"entropy": 0.8701171875,
|
|
"epoch": 0.7366910407981958,
|
|
"grad_norm": 0.27955051543992593,
|
|
"learning_rate": 3.2566577427081868e-06,
|
|
"loss": 0.8688,
|
|
"mean_token_accuracy": 0.7754341244697571,
|
|
"num_tokens": 539529171.0,
|
|
"step": 5390
|
|
},
|
|
{
|
|
"entropy": 0.893359375,
|
|
"epoch": 0.7380578145288047,
|
|
"grad_norm": 0.2950432320125493,
|
|
"learning_rate": 3.25313512751867e-06,
|
|
"loss": 0.8958,
|
|
"mean_token_accuracy": 0.7699394792318344,
|
|
"num_tokens": 540506909.0,
|
|
"step": 5400
|
|
},
|
|
{
|
|
"entropy": 0.9197265625,
|
|
"epoch": 0.7394245882594137,
|
|
"grad_norm": 0.2965903454057777,
|
|
"learning_rate": 3.2496125123291535e-06,
|
|
"loss": 0.9204,
|
|
"mean_token_accuracy": 0.7633383959531784,
|
|
"num_tokens": 541508626.0,
|
|
"step": 5410
|
|
},
|
|
{
|
|
"entropy": 0.876171875,
|
|
"epoch": 0.7407913619900226,
|
|
"grad_norm": 0.2714020422724535,
|
|
"learning_rate": 3.2460898971396366e-06,
|
|
"loss": 0.8688,
|
|
"mean_token_accuracy": 0.7768491715192795,
|
|
"num_tokens": 542509987.0,
|
|
"step": 5420
|
|
},
|
|
{
|
|
"entropy": 0.8970703125,
|
|
"epoch": 0.7421581357206315,
|
|
"grad_norm": 0.2907723778099968,
|
|
"learning_rate": 3.24256728195012e-06,
|
|
"loss": 0.8921,
|
|
"mean_token_accuracy": 0.7691859573125839,
|
|
"num_tokens": 543492581.0,
|
|
"step": 5430
|
|
},
|
|
{
|
|
"entropy": 0.8888671875,
|
|
"epoch": 0.7435249094512404,
|
|
"grad_norm": 0.26310262879345303,
|
|
"learning_rate": 3.2390446667606037e-06,
|
|
"loss": 0.8919,
|
|
"mean_token_accuracy": 0.7701226651668549,
|
|
"num_tokens": 544522249.0,
|
|
"step": 5440
|
|
},
|
|
{
|
|
"entropy": 0.8955078125,
|
|
"epoch": 0.7448916831818493,
|
|
"grad_norm": 0.3476537300017594,
|
|
"learning_rate": 3.235522051571087e-06,
|
|
"loss": 0.8902,
|
|
"mean_token_accuracy": 0.7710148870944977,
|
|
"num_tokens": 545521894.0,
|
|
"step": 5450
|
|
},
|
|
{
|
|
"entropy": 0.9078125,
|
|
"epoch": 0.7462584569124582,
|
|
"grad_norm": 0.2626627725730156,
|
|
"learning_rate": 3.23199943638157e-06,
|
|
"loss": 0.9028,
|
|
"mean_token_accuracy": 0.7684774816036224,
|
|
"num_tokens": 546492940.0,
|
|
"step": 5460
|
|
},
|
|
{
|
|
"entropy": 0.9080078125,
|
|
"epoch": 0.7476252306430671,
|
|
"grad_norm": 0.3362579883840761,
|
|
"learning_rate": 3.228476821192053e-06,
|
|
"loss": 0.9231,
|
|
"mean_token_accuracy": 0.763124018907547,
|
|
"num_tokens": 547508015.0,
|
|
"step": 5470
|
|
},
|
|
{
|
|
"entropy": 0.9205078125,
|
|
"epoch": 0.748992004373676,
|
|
"grad_norm": 0.3120676741429271,
|
|
"learning_rate": 3.224954206002536e-06,
|
|
"loss": 0.9261,
|
|
"mean_token_accuracy": 0.7636157691478729,
|
|
"num_tokens": 548533834.0,
|
|
"step": 5480
|
|
},
|
|
{
|
|
"entropy": 0.865625,
|
|
"epoch": 0.7503587781042849,
|
|
"grad_norm": 0.35358362888236927,
|
|
"learning_rate": 3.22143159081302e-06,
|
|
"loss": 0.8541,
|
|
"mean_token_accuracy": 0.7780008107423783,
|
|
"num_tokens": 549548230.0,
|
|
"step": 5490
|
|
},
|
|
{
|
|
"entropy": 0.907421875,
|
|
"epoch": 0.7517255518348938,
|
|
"grad_norm": 0.3061013591166059,
|
|
"learning_rate": 3.2179089756235033e-06,
|
|
"loss": 0.8917,
|
|
"mean_token_accuracy": 0.7688974410295486,
|
|
"num_tokens": 550550533.0,
|
|
"step": 5500
|
|
},
|
|
{
|
|
"entropy": 0.907421875,
|
|
"epoch": 0.7530923255655027,
|
|
"grad_norm": 0.2678067803036011,
|
|
"learning_rate": 3.2143863604339864e-06,
|
|
"loss": 0.9017,
|
|
"mean_token_accuracy": 0.7672990500926972,
|
|
"num_tokens": 551582423.0,
|
|
"step": 5510
|
|
},
|
|
{
|
|
"entropy": 0.9076171875,
|
|
"epoch": 0.7544590992961115,
|
|
"grad_norm": 0.3634285007578262,
|
|
"learning_rate": 3.2108637452444695e-06,
|
|
"loss": 0.9106,
|
|
"mean_token_accuracy": 0.7672886610031128,
|
|
"num_tokens": 552575638.0,
|
|
"step": 5520
|
|
},
|
|
{
|
|
"entropy": 0.933203125,
|
|
"epoch": 0.7558258730267204,
|
|
"grad_norm": 0.32989346964245464,
|
|
"learning_rate": 3.2073411300549526e-06,
|
|
"loss": 0.9268,
|
|
"mean_token_accuracy": 0.7639785975217819,
|
|
"num_tokens": 553546977.0,
|
|
"step": 5530
|
|
},
|
|
{
|
|
"entropy": 0.8451171875,
|
|
"epoch": 0.7571926467573293,
|
|
"grad_norm": 0.2693544574814418,
|
|
"learning_rate": 3.2038185148654366e-06,
|
|
"loss": 0.8359,
|
|
"mean_token_accuracy": 0.7815655440092086,
|
|
"num_tokens": 554537362.0,
|
|
"step": 5540
|
|
},
|
|
{
|
|
"entropy": 0.9142578125,
|
|
"epoch": 0.7585594204879382,
|
|
"grad_norm": 0.3351571707362659,
|
|
"learning_rate": 3.2002958996759197e-06,
|
|
"loss": 0.917,
|
|
"mean_token_accuracy": 0.7660951018333435,
|
|
"num_tokens": 555506336.0,
|
|
"step": 5550
|
|
},
|
|
{
|
|
"entropy": 0.934765625,
|
|
"epoch": 0.7599261942185471,
|
|
"grad_norm": 0.29262148896505474,
|
|
"learning_rate": 3.196773284486403e-06,
|
|
"loss": 0.9344,
|
|
"mean_token_accuracy": 0.761007422208786,
|
|
"num_tokens": 556496732.0,
|
|
"step": 5560
|
|
},
|
|
{
|
|
"entropy": 0.9181640625,
|
|
"epoch": 0.761292967949156,
|
|
"grad_norm": 0.34914530393017396,
|
|
"learning_rate": 3.1932506692968864e-06,
|
|
"loss": 0.9143,
|
|
"mean_token_accuracy": 0.765267837047577,
|
|
"num_tokens": 557522731.0,
|
|
"step": 5570
|
|
},
|
|
{
|
|
"entropy": 0.872265625,
|
|
"epoch": 0.7626597416797649,
|
|
"grad_norm": 0.27659033246844356,
|
|
"learning_rate": 3.1897280541073695e-06,
|
|
"loss": 0.8705,
|
|
"mean_token_accuracy": 0.7735489726066589,
|
|
"num_tokens": 558552808.0,
|
|
"step": 5580
|
|
},
|
|
{
|
|
"entropy": 0.890625,
|
|
"epoch": 0.7640265154103738,
|
|
"grad_norm": 0.25641592258076734,
|
|
"learning_rate": 3.1862054389178526e-06,
|
|
"loss": 0.8877,
|
|
"mean_token_accuracy": 0.7711788356304169,
|
|
"num_tokens": 559613029.0,
|
|
"step": 5590
|
|
},
|
|
{
|
|
"entropy": 0.8861328125,
|
|
"epoch": 0.7653932891409827,
|
|
"grad_norm": 0.28097394045631,
|
|
"learning_rate": 3.182682823728336e-06,
|
|
"loss": 0.8793,
|
|
"mean_token_accuracy": 0.7701630532741547,
|
|
"num_tokens": 560628142.0,
|
|
"step": 5600
|
|
},
|
|
{
|
|
"entropy": 0.86484375,
|
|
"epoch": 0.7667600628715916,
|
|
"grad_norm": 0.27223638196836547,
|
|
"learning_rate": 3.1791602085388197e-06,
|
|
"loss": 0.8647,
|
|
"mean_token_accuracy": 0.7757947117090225,
|
|
"num_tokens": 561653303.0,
|
|
"step": 5610
|
|
},
|
|
{
|
|
"entropy": 0.9384765625,
|
|
"epoch": 0.7681268366022005,
|
|
"grad_norm": 0.275395488043837,
|
|
"learning_rate": 3.175637593349303e-06,
|
|
"loss": 0.9412,
|
|
"mean_token_accuracy": 0.761640727519989,
|
|
"num_tokens": 562622003.0,
|
|
"step": 5620
|
|
},
|
|
{
|
|
"entropy": 0.9216796875,
|
|
"epoch": 0.7694936103328094,
|
|
"grad_norm": 0.34044140671708795,
|
|
"learning_rate": 3.172114978159786e-06,
|
|
"loss": 0.9195,
|
|
"mean_token_accuracy": 0.7659975230693817,
|
|
"num_tokens": 563647661.0,
|
|
"step": 5630
|
|
},
|
|
{
|
|
"entropy": 0.9021484375,
|
|
"epoch": 0.7708603840634183,
|
|
"grad_norm": 0.26416303123456836,
|
|
"learning_rate": 3.168592362970269e-06,
|
|
"loss": 0.9078,
|
|
"mean_token_accuracy": 0.7690748184919357,
|
|
"num_tokens": 564603750.0,
|
|
"step": 5640
|
|
},
|
|
{
|
|
"entropy": 0.8673828125,
|
|
"epoch": 0.7722271577940272,
|
|
"grad_norm": 0.27296727901648327,
|
|
"learning_rate": 3.165069747780753e-06,
|
|
"loss": 0.8677,
|
|
"mean_token_accuracy": 0.7747874736785889,
|
|
"num_tokens": 565577122.0,
|
|
"step": 5650
|
|
},
|
|
{
|
|
"entropy": 0.9013671875,
|
|
"epoch": 0.7735939315246361,
|
|
"grad_norm": 0.2746938018914984,
|
|
"learning_rate": 3.161547132591236e-06,
|
|
"loss": 0.8981,
|
|
"mean_token_accuracy": 0.7690708667039872,
|
|
"num_tokens": 566604600.0,
|
|
"step": 5660
|
|
},
|
|
{
|
|
"entropy": 0.9125,
|
|
"epoch": 0.774960705255245,
|
|
"grad_norm": 0.2664290844007511,
|
|
"learning_rate": 3.1580245174017193e-06,
|
|
"loss": 0.9207,
|
|
"mean_token_accuracy": 0.764921435713768,
|
|
"num_tokens": 567617845.0,
|
|
"step": 5670
|
|
},
|
|
{
|
|
"entropy": 0.8994140625,
|
|
"epoch": 0.7763274789858539,
|
|
"grad_norm": 0.32292279229224485,
|
|
"learning_rate": 3.1545019022122025e-06,
|
|
"loss": 0.8903,
|
|
"mean_token_accuracy": 0.7699287116527558,
|
|
"num_tokens": 568612057.0,
|
|
"step": 5680
|
|
},
|
|
{
|
|
"entropy": 0.9275390625,
|
|
"epoch": 0.7776942527164628,
|
|
"grad_norm": 0.29936240756471966,
|
|
"learning_rate": 3.1509792870226856e-06,
|
|
"loss": 0.9303,
|
|
"mean_token_accuracy": 0.7620971828699112,
|
|
"num_tokens": 569613099.0,
|
|
"step": 5690
|
|
},
|
|
{
|
|
"entropy": 0.890625,
|
|
"epoch": 0.7790610264470716,
|
|
"grad_norm": 0.3018128806533985,
|
|
"learning_rate": 3.147456671833169e-06,
|
|
"loss": 0.89,
|
|
"mean_token_accuracy": 0.7703081846237183,
|
|
"num_tokens": 570607026.0,
|
|
"step": 5700
|
|
},
|
|
{
|
|
"entropy": 0.8939453125,
|
|
"epoch": 0.7804278001776805,
|
|
"grad_norm": 0.30870490766706604,
|
|
"learning_rate": 3.1439340566436527e-06,
|
|
"loss": 0.884,
|
|
"mean_token_accuracy": 0.7718601077795029,
|
|
"num_tokens": 571616119.0,
|
|
"step": 5710
|
|
},
|
|
{
|
|
"entropy": 0.9046875,
|
|
"epoch": 0.7817945739082894,
|
|
"grad_norm": 0.3042790491923807,
|
|
"learning_rate": 3.140411441454136e-06,
|
|
"loss": 0.8998,
|
|
"mean_token_accuracy": 0.768331789970398,
|
|
"num_tokens": 572598084.0,
|
|
"step": 5720
|
|
},
|
|
{
|
|
"entropy": 0.90625,
|
|
"epoch": 0.7831613476388983,
|
|
"grad_norm": 0.281497368556221,
|
|
"learning_rate": 3.136888826264619e-06,
|
|
"loss": 0.909,
|
|
"mean_token_accuracy": 0.7662631750106812,
|
|
"num_tokens": 573590736.0,
|
|
"step": 5730
|
|
},
|
|
{
|
|
"entropy": 0.8728515625,
|
|
"epoch": 0.7845281213695072,
|
|
"grad_norm": 0.2973502498507691,
|
|
"learning_rate": 3.1333662110751025e-06,
|
|
"loss": 0.866,
|
|
"mean_token_accuracy": 0.7761127412319183,
|
|
"num_tokens": 574574376.0,
|
|
"step": 5740
|
|
},
|
|
{
|
|
"entropy": 0.8556640625,
|
|
"epoch": 0.7858948951001162,
|
|
"grad_norm": 0.30895054634760766,
|
|
"learning_rate": 3.1298435958855856e-06,
|
|
"loss": 0.8512,
|
|
"mean_token_accuracy": 0.7793437242507935,
|
|
"num_tokens": 575523219.0,
|
|
"step": 5750
|
|
},
|
|
{
|
|
"entropy": 0.894921875,
|
|
"epoch": 0.7872616688307251,
|
|
"grad_norm": 0.29472574145118235,
|
|
"learning_rate": 3.126320980696069e-06,
|
|
"loss": 0.8906,
|
|
"mean_token_accuracy": 0.7693179190158844,
|
|
"num_tokens": 576500257.0,
|
|
"step": 5760
|
|
},
|
|
{
|
|
"entropy": 0.9138671875,
|
|
"epoch": 0.788628442561334,
|
|
"grad_norm": 0.2827775332600573,
|
|
"learning_rate": 3.1227983655065523e-06,
|
|
"loss": 0.9071,
|
|
"mean_token_accuracy": 0.7665531575679779,
|
|
"num_tokens": 577473156.0,
|
|
"step": 5770
|
|
},
|
|
{
|
|
"entropy": 0.8666015625,
|
|
"epoch": 0.7899952162919429,
|
|
"grad_norm": 0.27632840989946184,
|
|
"learning_rate": 3.119275750317036e-06,
|
|
"loss": 0.8602,
|
|
"mean_token_accuracy": 0.7761566579341889,
|
|
"num_tokens": 578458348.0,
|
|
"step": 5780
|
|
},
|
|
{
|
|
"entropy": 0.9275390625,
|
|
"epoch": 0.7913619900225518,
|
|
"grad_norm": 0.30803241988782665,
|
|
"learning_rate": 3.115753135127519e-06,
|
|
"loss": 0.9242,
|
|
"mean_token_accuracy": 0.7619533866643906,
|
|
"num_tokens": 579469890.0,
|
|
"step": 5790
|
|
},
|
|
{
|
|
"entropy": 0.892578125,
|
|
"epoch": 0.7927287637531607,
|
|
"grad_norm": 0.300096470313119,
|
|
"learning_rate": 3.112230519938002e-06,
|
|
"loss": 0.8851,
|
|
"mean_token_accuracy": 0.7716841757297516,
|
|
"num_tokens": 580508736.0,
|
|
"step": 5800
|
|
},
|
|
{
|
|
"entropy": 0.8669921875,
|
|
"epoch": 0.7940955374837696,
|
|
"grad_norm": 0.2947895636621664,
|
|
"learning_rate": 3.108707904748485e-06,
|
|
"loss": 0.8629,
|
|
"mean_token_accuracy": 0.7762851566076279,
|
|
"num_tokens": 581445920.0,
|
|
"step": 5810
|
|
},
|
|
{
|
|
"entropy": 0.9,
|
|
"epoch": 0.7954623112143785,
|
|
"grad_norm": 0.2701714566578323,
|
|
"learning_rate": 3.105185289558969e-06,
|
|
"loss": 0.9043,
|
|
"mean_token_accuracy": 0.7689761877059936,
|
|
"num_tokens": 582487593.0,
|
|
"step": 5820
|
|
},
|
|
{
|
|
"entropy": 0.87265625,
|
|
"epoch": 0.7968290849449874,
|
|
"grad_norm": 0.2654638939905909,
|
|
"learning_rate": 3.1016626743694523e-06,
|
|
"loss": 0.8677,
|
|
"mean_token_accuracy": 0.7744100511074066,
|
|
"num_tokens": 583488287.0,
|
|
"step": 5830
|
|
},
|
|
{
|
|
"entropy": 0.8548828125,
|
|
"epoch": 0.7981958586755963,
|
|
"grad_norm": 0.27001932483122504,
|
|
"learning_rate": 3.0981400591799354e-06,
|
|
"loss": 0.8485,
|
|
"mean_token_accuracy": 0.7783486515283584,
|
|
"num_tokens": 584551415.0,
|
|
"step": 5840
|
|
},
|
|
{
|
|
"entropy": 0.8689453125,
|
|
"epoch": 0.7995626324062052,
|
|
"grad_norm": 0.25373940155758223,
|
|
"learning_rate": 3.0946174439904185e-06,
|
|
"loss": 0.8641,
|
|
"mean_token_accuracy": 0.7766380131244659,
|
|
"num_tokens": 585569034.0,
|
|
"step": 5850
|
|
},
|
|
{
|
|
"entropy": 0.891015625,
|
|
"epoch": 0.8009294061368141,
|
|
"grad_norm": 0.38413817412220114,
|
|
"learning_rate": 3.0910948288009016e-06,
|
|
"loss": 0.8764,
|
|
"mean_token_accuracy": 0.7734408110380173,
|
|
"num_tokens": 586536939.0,
|
|
"step": 5860
|
|
},
|
|
{
|
|
"entropy": 0.8833984375,
|
|
"epoch": 0.802296179867423,
|
|
"grad_norm": 0.3025608748724618,
|
|
"learning_rate": 3.0875722136113856e-06,
|
|
"loss": 0.8806,
|
|
"mean_token_accuracy": 0.7711967915296555,
|
|
"num_tokens": 587530274.0,
|
|
"step": 5870
|
|
},
|
|
{
|
|
"entropy": 0.9041015625,
|
|
"epoch": 0.8036629535980319,
|
|
"grad_norm": 0.274585661400754,
|
|
"learning_rate": 3.0840495984218687e-06,
|
|
"loss": 0.8996,
|
|
"mean_token_accuracy": 0.7676082521677017,
|
|
"num_tokens": 588558033.0,
|
|
"step": 5880
|
|
},
|
|
{
|
|
"entropy": 0.90859375,
|
|
"epoch": 0.8050297273286408,
|
|
"grad_norm": 0.28203823349508544,
|
|
"learning_rate": 3.080526983232352e-06,
|
|
"loss": 0.9074,
|
|
"mean_token_accuracy": 0.7661036610603332,
|
|
"num_tokens": 589542926.0,
|
|
"step": 5890
|
|
},
|
|
{
|
|
"entropy": 0.8984375,
|
|
"epoch": 0.8063965010592496,
|
|
"grad_norm": 0.3038363502775667,
|
|
"learning_rate": 3.0770043680428354e-06,
|
|
"loss": 0.8913,
|
|
"mean_token_accuracy": 0.7687708616256714,
|
|
"num_tokens": 590551239.0,
|
|
"step": 5900
|
|
},
|
|
{
|
|
"entropy": 0.8578125,
|
|
"epoch": 0.8077632747898585,
|
|
"grad_norm": 0.2831527603847802,
|
|
"learning_rate": 3.0734817528533185e-06,
|
|
"loss": 0.8555,
|
|
"mean_token_accuracy": 0.778011578321457,
|
|
"num_tokens": 591541550.0,
|
|
"step": 5910
|
|
},
|
|
{
|
|
"entropy": 0.927734375,
|
|
"epoch": 0.8091300485204674,
|
|
"grad_norm": 0.29009949629434406,
|
|
"learning_rate": 3.0699591376638017e-06,
|
|
"loss": 0.9229,
|
|
"mean_token_accuracy": 0.7636651486158371,
|
|
"num_tokens": 592526054.0,
|
|
"step": 5920
|
|
},
|
|
{
|
|
"entropy": 0.8830078125,
|
|
"epoch": 0.8104968222510763,
|
|
"grad_norm": 0.2851961307526368,
|
|
"learning_rate": 3.066436522474285e-06,
|
|
"loss": 0.8736,
|
|
"mean_token_accuracy": 0.7756423354148865,
|
|
"num_tokens": 593517164.0,
|
|
"step": 5930
|
|
},
|
|
{
|
|
"entropy": 0.8416015625,
|
|
"epoch": 0.8118635959816852,
|
|
"grad_norm": 0.24808028440984853,
|
|
"learning_rate": 3.0629139072847688e-06,
|
|
"loss": 0.8431,
|
|
"mean_token_accuracy": 0.7804061889648437,
|
|
"num_tokens": 594586885.0,
|
|
"step": 5940
|
|
},
|
|
{
|
|
"entropy": 0.916015625,
|
|
"epoch": 0.8132303697122941,
|
|
"grad_norm": 0.3702362083262213,
|
|
"learning_rate": 3.059391292095252e-06,
|
|
"loss": 0.9095,
|
|
"mean_token_accuracy": 0.7654503285884857,
|
|
"num_tokens": 595562782.0,
|
|
"step": 5950
|
|
},
|
|
{
|
|
"entropy": 0.8943359375,
|
|
"epoch": 0.814597143442903,
|
|
"grad_norm": 0.28545378494084495,
|
|
"learning_rate": 3.055868676905735e-06,
|
|
"loss": 0.8932,
|
|
"mean_token_accuracy": 0.7706179589033126,
|
|
"num_tokens": 596620797.0,
|
|
"step": 5960
|
|
},
|
|
{
|
|
"entropy": 0.91328125,
|
|
"epoch": 0.8159639171735119,
|
|
"grad_norm": 0.3610587324568096,
|
|
"learning_rate": 3.052346061716218e-06,
|
|
"loss": 0.9191,
|
|
"mean_token_accuracy": 0.7660729050636291,
|
|
"num_tokens": 597544265.0,
|
|
"step": 5970
|
|
},
|
|
{
|
|
"entropy": 0.891015625,
|
|
"epoch": 0.8173306909041208,
|
|
"grad_norm": 0.2815080569752477,
|
|
"learning_rate": 3.048823446526702e-06,
|
|
"loss": 0.8868,
|
|
"mean_token_accuracy": 0.771172821521759,
|
|
"num_tokens": 598556947.0,
|
|
"step": 5980
|
|
},
|
|
{
|
|
"entropy": 0.9146484375,
|
|
"epoch": 0.8186974646347297,
|
|
"grad_norm": 0.33177204432434243,
|
|
"learning_rate": 3.0453008313371852e-06,
|
|
"loss": 0.9111,
|
|
"mean_token_accuracy": 0.766373211145401,
|
|
"num_tokens": 599603049.0,
|
|
"step": 5990
|
|
},
|
|
{
|
|
"entropy": 0.8828125,
|
|
"epoch": 0.8200642383653386,
|
|
"grad_norm": 0.3195176397700544,
|
|
"learning_rate": 3.0417782161476683e-06,
|
|
"loss": 0.8883,
|
|
"mean_token_accuracy": 0.76996029317379,
|
|
"num_tokens": 600590091.0,
|
|
"step": 6000
|
|
},
|
|
{
|
|
"entropy": 0.9376953125,
|
|
"epoch": 0.8214310120959475,
|
|
"grad_norm": 0.3671883373151563,
|
|
"learning_rate": 3.0382556009581515e-06,
|
|
"loss": 0.9437,
|
|
"mean_token_accuracy": 0.7601265281438827,
|
|
"num_tokens": 601575708.0,
|
|
"step": 6010
|
|
},
|
|
{
|
|
"entropy": 0.8822265625,
|
|
"epoch": 0.8227977858265564,
|
|
"grad_norm": 0.2738898238283687,
|
|
"learning_rate": 3.0347329857686346e-06,
|
|
"loss": 0.866,
|
|
"mean_token_accuracy": 0.7740317285060883,
|
|
"num_tokens": 602543161.0,
|
|
"step": 6020
|
|
},
|
|
{
|
|
"entropy": 0.8748046875,
|
|
"epoch": 0.8241645595571653,
|
|
"grad_norm": 0.29100339432489547,
|
|
"learning_rate": 3.031210370579118e-06,
|
|
"loss": 0.8767,
|
|
"mean_token_accuracy": 0.7753896892070771,
|
|
"num_tokens": 603550842.0,
|
|
"step": 6030
|
|
},
|
|
{
|
|
"entropy": 0.8708984375,
|
|
"epoch": 0.8255313332877742,
|
|
"grad_norm": 0.2968304893586347,
|
|
"learning_rate": 3.0276877553896017e-06,
|
|
"loss": 0.8606,
|
|
"mean_token_accuracy": 0.7753192543983459,
|
|
"num_tokens": 604559325.0,
|
|
"step": 6040
|
|
},
|
|
{
|
|
"entropy": 0.9009765625,
|
|
"epoch": 0.8268981070183831,
|
|
"grad_norm": 0.32764080011625735,
|
|
"learning_rate": 3.024165140200085e-06,
|
|
"loss": 0.895,
|
|
"mean_token_accuracy": 0.770570182800293,
|
|
"num_tokens": 605552607.0,
|
|
"step": 6050
|
|
},
|
|
{
|
|
"entropy": 0.897265625,
|
|
"epoch": 0.828264880748992,
|
|
"grad_norm": 0.2938554435291315,
|
|
"learning_rate": 3.020642525010568e-06,
|
|
"loss": 0.9003,
|
|
"mean_token_accuracy": 0.7688303917646409,
|
|
"num_tokens": 606538979.0,
|
|
"step": 6060
|
|
},
|
|
{
|
|
"entropy": 0.8794921875,
|
|
"epoch": 0.8296316544796009,
|
|
"grad_norm": 0.282877223433461,
|
|
"learning_rate": 3.0171199098210515e-06,
|
|
"loss": 0.8703,
|
|
"mean_token_accuracy": 0.774592998623848,
|
|
"num_tokens": 607508204.0,
|
|
"step": 6070
|
|
},
|
|
{
|
|
"entropy": 0.8912109375,
|
|
"epoch": 0.8309984282102097,
|
|
"grad_norm": 0.29047987434470307,
|
|
"learning_rate": 3.0135972946315346e-06,
|
|
"loss": 0.8862,
|
|
"mean_token_accuracy": 0.7719777435064316,
|
|
"num_tokens": 608524577.0,
|
|
"step": 6080
|
|
},
|
|
{
|
|
"entropy": 0.907421875,
|
|
"epoch": 0.8323652019408186,
|
|
"grad_norm": 0.27581305182431626,
|
|
"learning_rate": 3.010074679442018e-06,
|
|
"loss": 0.9048,
|
|
"mean_token_accuracy": 0.767045396566391,
|
|
"num_tokens": 609491404.0,
|
|
"step": 6090
|
|
},
|
|
{
|
|
"entropy": 0.91484375,
|
|
"epoch": 0.8337319756714276,
|
|
"grad_norm": 0.40412802190109115,
|
|
"learning_rate": 3.0065520642525013e-06,
|
|
"loss": 0.916,
|
|
"mean_token_accuracy": 0.7664613515138626,
|
|
"num_tokens": 610456894.0,
|
|
"step": 6100
|
|
},
|
|
{
|
|
"entropy": 0.91171875,
|
|
"epoch": 0.8350987494020365,
|
|
"grad_norm": 0.28689840579528164,
|
|
"learning_rate": 3.003029449062985e-06,
|
|
"loss": 0.9003,
|
|
"mean_token_accuracy": 0.7690335929393768,
|
|
"num_tokens": 611476065.0,
|
|
"step": 6110
|
|
},
|
|
{
|
|
"entropy": 0.896875,
|
|
"epoch": 0.8364655231326454,
|
|
"grad_norm": 0.31004564144693375,
|
|
"learning_rate": 2.999506833873468e-06,
|
|
"loss": 0.893,
|
|
"mean_token_accuracy": 0.7694867730140686,
|
|
"num_tokens": 612425310.0,
|
|
"step": 6120
|
|
},
|
|
{
|
|
"entropy": 0.9193359375,
|
|
"epoch": 0.8378322968632543,
|
|
"grad_norm": 0.37486648449273013,
|
|
"learning_rate": 2.995984218683951e-06,
|
|
"loss": 0.9046,
|
|
"mean_token_accuracy": 0.7655063241720199,
|
|
"num_tokens": 613425793.0,
|
|
"step": 6130
|
|
},
|
|
{
|
|
"entropy": 0.8857421875,
|
|
"epoch": 0.8391990705938632,
|
|
"grad_norm": 0.41844726609013555,
|
|
"learning_rate": 2.992461603494434e-06,
|
|
"loss": 0.8754,
|
|
"mean_token_accuracy": 0.7719518572092057,
|
|
"num_tokens": 614450318.0,
|
|
"step": 6140
|
|
},
|
|
{
|
|
"entropy": 0.89140625,
|
|
"epoch": 0.8405658443244721,
|
|
"grad_norm": 0.3645615287594191,
|
|
"learning_rate": 2.988938988304918e-06,
|
|
"loss": 0.8855,
|
|
"mean_token_accuracy": 0.771328791975975,
|
|
"num_tokens": 615468699.0,
|
|
"step": 6150
|
|
},
|
|
{
|
|
"entropy": 0.876953125,
|
|
"epoch": 0.841932618055081,
|
|
"grad_norm": 0.3124806337992799,
|
|
"learning_rate": 2.9854163731154013e-06,
|
|
"loss": 0.8709,
|
|
"mean_token_accuracy": 0.7755225718021392,
|
|
"num_tokens": 616457438.0,
|
|
"step": 6160
|
|
},
|
|
{
|
|
"entropy": 0.9056640625,
|
|
"epoch": 0.8432993917856899,
|
|
"grad_norm": 0.25854129104716156,
|
|
"learning_rate": 2.9818937579258844e-06,
|
|
"loss": 0.9068,
|
|
"mean_token_accuracy": 0.7659855157136917,
|
|
"num_tokens": 617408605.0,
|
|
"step": 6170
|
|
},
|
|
{
|
|
"entropy": 0.86953125,
|
|
"epoch": 0.8446661655162988,
|
|
"grad_norm": 0.30574876140082013,
|
|
"learning_rate": 2.9783711427363675e-06,
|
|
"loss": 0.8681,
|
|
"mean_token_accuracy": 0.7745053678750992,
|
|
"num_tokens": 618412780.0,
|
|
"step": 6180
|
|
},
|
|
{
|
|
"entropy": 0.9017578125,
|
|
"epoch": 0.8460329392469077,
|
|
"grad_norm": 0.3465700156954281,
|
|
"learning_rate": 2.9748485275468507e-06,
|
|
"loss": 0.8969,
|
|
"mean_token_accuracy": 0.7703600198030471,
|
|
"num_tokens": 619400950.0,
|
|
"step": 6190
|
|
},
|
|
{
|
|
"entropy": 0.8798828125,
|
|
"epoch": 0.8473997129775166,
|
|
"grad_norm": 0.2823660356870364,
|
|
"learning_rate": 2.9713259123573346e-06,
|
|
"loss": 0.8784,
|
|
"mean_token_accuracy": 0.7737836420536042,
|
|
"num_tokens": 620361772.0,
|
|
"step": 6200
|
|
},
|
|
{
|
|
"entropy": 0.8822265625,
|
|
"epoch": 0.8487664867081255,
|
|
"grad_norm": 0.27322114315962576,
|
|
"learning_rate": 2.9678032971678177e-06,
|
|
"loss": 0.8836,
|
|
"mean_token_accuracy": 0.7705852806568145,
|
|
"num_tokens": 621309759.0,
|
|
"step": 6210
|
|
},
|
|
{
|
|
"entropy": 0.8814453125,
|
|
"epoch": 0.8501332604387344,
|
|
"grad_norm": 0.2837729258750379,
|
|
"learning_rate": 2.964280681978301e-06,
|
|
"loss": 0.882,
|
|
"mean_token_accuracy": 0.77015500664711,
|
|
"num_tokens": 622344321.0,
|
|
"step": 6220
|
|
},
|
|
{
|
|
"entropy": 0.862890625,
|
|
"epoch": 0.8515000341693433,
|
|
"grad_norm": 0.33320821725012845,
|
|
"learning_rate": 2.960758066788784e-06,
|
|
"loss": 0.8524,
|
|
"mean_token_accuracy": 0.7772625178098679,
|
|
"num_tokens": 623307147.0,
|
|
"step": 6230
|
|
},
|
|
{
|
|
"entropy": 0.921875,
|
|
"epoch": 0.8528668078999522,
|
|
"grad_norm": 0.2734156114194939,
|
|
"learning_rate": 2.9572354515992675e-06,
|
|
"loss": 0.9093,
|
|
"mean_token_accuracy": 0.7629626095294952,
|
|
"num_tokens": 624284064.0,
|
|
"step": 6240
|
|
},
|
|
{
|
|
"entropy": 0.835546875,
|
|
"epoch": 0.8542335816305611,
|
|
"grad_norm": 0.2899292779003155,
|
|
"learning_rate": 2.9537128364097507e-06,
|
|
"loss": 0.8257,
|
|
"mean_token_accuracy": 0.7855453580617905,
|
|
"num_tokens": 625283185.0,
|
|
"step": 6250
|
|
},
|
|
{
|
|
"entropy": 0.87890625,
|
|
"epoch": 0.85560035536117,
|
|
"grad_norm": 0.29386007613203097,
|
|
"learning_rate": 2.9501902212202342e-06,
|
|
"loss": 0.8731,
|
|
"mean_token_accuracy": 0.7744217306375504,
|
|
"num_tokens": 626319054.0,
|
|
"step": 6260
|
|
},
|
|
{
|
|
"entropy": 0.8904296875,
|
|
"epoch": 0.8569671290917789,
|
|
"grad_norm": 0.26853020394553734,
|
|
"learning_rate": 2.9466676060307178e-06,
|
|
"loss": 0.8909,
|
|
"mean_token_accuracy": 0.7687609195709229,
|
|
"num_tokens": 627327798.0,
|
|
"step": 6270
|
|
},
|
|
{
|
|
"entropy": 0.907421875,
|
|
"epoch": 0.8583339028223878,
|
|
"grad_norm": 0.28652457759446265,
|
|
"learning_rate": 2.943144990841201e-06,
|
|
"loss": 0.9038,
|
|
"mean_token_accuracy": 0.7694362670183181,
|
|
"num_tokens": 628345763.0,
|
|
"step": 6280
|
|
},
|
|
{
|
|
"entropy": 0.8841796875,
|
|
"epoch": 0.8597006765529966,
|
|
"grad_norm": 0.27092711649821094,
|
|
"learning_rate": 2.939622375651684e-06,
|
|
"loss": 0.8708,
|
|
"mean_token_accuracy": 0.7756720900535583,
|
|
"num_tokens": 629349687.0,
|
|
"step": 6290
|
|
},
|
|
{
|
|
"entropy": 0.905078125,
|
|
"epoch": 0.8610674502836055,
|
|
"grad_norm": 0.31907258701129465,
|
|
"learning_rate": 2.936099760462167e-06,
|
|
"loss": 0.9037,
|
|
"mean_token_accuracy": 0.7685008376836777,
|
|
"num_tokens": 630396591.0,
|
|
"step": 6300
|
|
},
|
|
{
|
|
"entropy": 0.87421875,
|
|
"epoch": 0.8624342240142144,
|
|
"grad_norm": 0.28363102841992893,
|
|
"learning_rate": 2.932577145272651e-06,
|
|
"loss": 0.8683,
|
|
"mean_token_accuracy": 0.7747282147407532,
|
|
"num_tokens": 631372808.0,
|
|
"step": 6310
|
|
},
|
|
{
|
|
"entropy": 0.9078125,
|
|
"epoch": 0.8638009977448233,
|
|
"grad_norm": 0.31883856919936154,
|
|
"learning_rate": 2.9290545300831342e-06,
|
|
"loss": 0.9054,
|
|
"mean_token_accuracy": 0.7676671206951141,
|
|
"num_tokens": 632389956.0,
|
|
"step": 6320
|
|
},
|
|
{
|
|
"entropy": 0.9197265625,
|
|
"epoch": 0.8651677714754322,
|
|
"grad_norm": 0.30177849498562304,
|
|
"learning_rate": 2.9255319148936174e-06,
|
|
"loss": 0.9114,
|
|
"mean_token_accuracy": 0.7681419640779495,
|
|
"num_tokens": 633382741.0,
|
|
"step": 6330
|
|
},
|
|
{
|
|
"entropy": 0.85234375,
|
|
"epoch": 0.8665345452060411,
|
|
"grad_norm": 0.2911428737793762,
|
|
"learning_rate": 2.9220092997041005e-06,
|
|
"loss": 0.8474,
|
|
"mean_token_accuracy": 0.7829389095306396,
|
|
"num_tokens": 634392872.0,
|
|
"step": 6340
|
|
},
|
|
{
|
|
"entropy": 0.89375,
|
|
"epoch": 0.86790131893665,
|
|
"grad_norm": 0.25962561719047317,
|
|
"learning_rate": 2.9184866845145836e-06,
|
|
"loss": 0.8805,
|
|
"mean_token_accuracy": 0.7721905618906021,
|
|
"num_tokens": 635397192.0,
|
|
"step": 6350
|
|
},
|
|
{
|
|
"entropy": 0.8849609375,
|
|
"epoch": 0.8692680926672589,
|
|
"grad_norm": 0.29472648804987384,
|
|
"learning_rate": 2.9149640693250667e-06,
|
|
"loss": 0.8729,
|
|
"mean_token_accuracy": 0.773974609375,
|
|
"num_tokens": 636429865.0,
|
|
"step": 6360
|
|
},
|
|
{
|
|
"entropy": 0.88984375,
|
|
"epoch": 0.8706348663978678,
|
|
"grad_norm": 0.2976678036320609,
|
|
"learning_rate": 2.9114414541355507e-06,
|
|
"loss": 0.8876,
|
|
"mean_token_accuracy": 0.7709997862577438,
|
|
"num_tokens": 637412707.0,
|
|
"step": 6370
|
|
},
|
|
{
|
|
"entropy": 0.912109375,
|
|
"epoch": 0.8720016401284767,
|
|
"grad_norm": 0.26123672663722414,
|
|
"learning_rate": 2.907918838946034e-06,
|
|
"loss": 0.9043,
|
|
"mean_token_accuracy": 0.7685527354478836,
|
|
"num_tokens": 638419040.0,
|
|
"step": 6380
|
|
},
|
|
{
|
|
"entropy": 0.891796875,
|
|
"epoch": 0.8733684138590856,
|
|
"grad_norm": 0.29333362308197286,
|
|
"learning_rate": 2.904396223756517e-06,
|
|
"loss": 0.8813,
|
|
"mean_token_accuracy": 0.7728693336248398,
|
|
"num_tokens": 639409575.0,
|
|
"step": 6390
|
|
},
|
|
{
|
|
"entropy": 0.8919921875,
|
|
"epoch": 0.8747351875896945,
|
|
"grad_norm": 0.28632471074654403,
|
|
"learning_rate": 2.9008736085670005e-06,
|
|
"loss": 0.8806,
|
|
"mean_token_accuracy": 0.7722423315048218,
|
|
"num_tokens": 640403980.0,
|
|
"step": 6400
|
|
},
|
|
{
|
|
"entropy": 0.8537109375,
|
|
"epoch": 0.8761019613203034,
|
|
"grad_norm": 0.30524315685302217,
|
|
"learning_rate": 2.8973509933774836e-06,
|
|
"loss": 0.8411,
|
|
"mean_token_accuracy": 0.7812278360128403,
|
|
"num_tokens": 641380100.0,
|
|
"step": 6410
|
|
},
|
|
{
|
|
"entropy": 0.9107421875,
|
|
"epoch": 0.8774687350509123,
|
|
"grad_norm": 0.3733701507128681,
|
|
"learning_rate": 2.893828378187967e-06,
|
|
"loss": 0.9144,
|
|
"mean_token_accuracy": 0.7660555303096771,
|
|
"num_tokens": 642396526.0,
|
|
"step": 6420
|
|
},
|
|
{
|
|
"entropy": 0.865234375,
|
|
"epoch": 0.8788355087815212,
|
|
"grad_norm": 0.29484673447409643,
|
|
"learning_rate": 2.8903057629984503e-06,
|
|
"loss": 0.8586,
|
|
"mean_token_accuracy": 0.7770076036453247,
|
|
"num_tokens": 643365779.0,
|
|
"step": 6430
|
|
},
|
|
{
|
|
"entropy": 0.882421875,
|
|
"epoch": 0.8802022825121301,
|
|
"grad_norm": 0.30895641724381384,
|
|
"learning_rate": 2.886783147808934e-06,
|
|
"loss": 0.8762,
|
|
"mean_token_accuracy": 0.7743527173995972,
|
|
"num_tokens": 644374233.0,
|
|
"step": 6440
|
|
},
|
|
{
|
|
"entropy": 0.931640625,
|
|
"epoch": 0.8815690562427391,
|
|
"grad_norm": 0.2998520257046201,
|
|
"learning_rate": 2.883260532619417e-06,
|
|
"loss": 0.9242,
|
|
"mean_token_accuracy": 0.7635059386491776,
|
|
"num_tokens": 645333738.0,
|
|
"step": 6450
|
|
},
|
|
{
|
|
"entropy": 0.8869140625,
|
|
"epoch": 0.882935829973348,
|
|
"grad_norm": 0.3210165816758822,
|
|
"learning_rate": 2.8797379174299e-06,
|
|
"loss": 0.8748,
|
|
"mean_token_accuracy": 0.7737338364124298,
|
|
"num_tokens": 646280138.0,
|
|
"step": 6460
|
|
},
|
|
{
|
|
"entropy": 0.838671875,
|
|
"epoch": 0.8843026037039569,
|
|
"grad_norm": 0.30676855826282945,
|
|
"learning_rate": 2.876215302240383e-06,
|
|
"loss": 0.8348,
|
|
"mean_token_accuracy": 0.7849259167909622,
|
|
"num_tokens": 647278647.0,
|
|
"step": 6470
|
|
},
|
|
{
|
|
"entropy": 0.9193359375,
|
|
"epoch": 0.8856693774345658,
|
|
"grad_norm": 0.30594542224690174,
|
|
"learning_rate": 2.872692687050867e-06,
|
|
"loss": 0.905,
|
|
"mean_token_accuracy": 0.769131389260292,
|
|
"num_tokens": 648304920.0,
|
|
"step": 6480
|
|
},
|
|
{
|
|
"entropy": 0.8896484375,
|
|
"epoch": 0.8870361511651746,
|
|
"grad_norm": 0.27650512671463223,
|
|
"learning_rate": 2.8691700718613503e-06,
|
|
"loss": 0.8847,
|
|
"mean_token_accuracy": 0.7703093409538269,
|
|
"num_tokens": 649296959.0,
|
|
"step": 6490
|
|
},
|
|
{
|
|
"entropy": 0.878515625,
|
|
"epoch": 0.8884029248957835,
|
|
"grad_norm": 0.2820211382113378,
|
|
"learning_rate": 2.8656474566718334e-06,
|
|
"loss": 0.8743,
|
|
"mean_token_accuracy": 0.7723534286022187,
|
|
"num_tokens": 650329433.0,
|
|
"step": 6500
|
|
},
|
|
{
|
|
"entropy": 0.8626953125,
|
|
"epoch": 0.8897696986263924,
|
|
"grad_norm": 0.2676014278347393,
|
|
"learning_rate": 2.8621248414823165e-06,
|
|
"loss": 0.8569,
|
|
"mean_token_accuracy": 0.7757056713104248,
|
|
"num_tokens": 651325368.0,
|
|
"step": 6510
|
|
},
|
|
{
|
|
"entropy": 0.9001953125,
|
|
"epoch": 0.8911364723570013,
|
|
"grad_norm": 0.302589715644279,
|
|
"learning_rate": 2.8586022262927997e-06,
|
|
"loss": 0.8865,
|
|
"mean_token_accuracy": 0.7705445438623428,
|
|
"num_tokens": 652253282.0,
|
|
"step": 6520
|
|
},
|
|
{
|
|
"entropy": 0.9228515625,
|
|
"epoch": 0.8925032460876102,
|
|
"grad_norm": 0.2863552202135847,
|
|
"learning_rate": 2.8550796111032836e-06,
|
|
"loss": 0.9176,
|
|
"mean_token_accuracy": 0.7643186658620834,
|
|
"num_tokens": 653261800.0,
|
|
"step": 6530
|
|
},
|
|
{
|
|
"entropy": 0.886328125,
|
|
"epoch": 0.8938700198182191,
|
|
"grad_norm": 0.2863705197951552,
|
|
"learning_rate": 2.8515569959137668e-06,
|
|
"loss": 0.8922,
|
|
"mean_token_accuracy": 0.770050299167633,
|
|
"num_tokens": 654258824.0,
|
|
"step": 6540
|
|
},
|
|
{
|
|
"entropy": 0.8787109375,
|
|
"epoch": 0.895236793548828,
|
|
"grad_norm": 0.28281080033698736,
|
|
"learning_rate": 2.84803438072425e-06,
|
|
"loss": 0.8764,
|
|
"mean_token_accuracy": 0.775055068731308,
|
|
"num_tokens": 655251617.0,
|
|
"step": 6550
|
|
},
|
|
{
|
|
"entropy": 0.8810546875,
|
|
"epoch": 0.8966035672794369,
|
|
"grad_norm": 0.2830167037145398,
|
|
"learning_rate": 2.844511765534733e-06,
|
|
"loss": 0.8783,
|
|
"mean_token_accuracy": 0.7740787178277969,
|
|
"num_tokens": 656244842.0,
|
|
"step": 6560
|
|
},
|
|
{
|
|
"entropy": 0.8505859375,
|
|
"epoch": 0.8979703410100458,
|
|
"grad_norm": 0.2560273556870744,
|
|
"learning_rate": 2.8409891503452166e-06,
|
|
"loss": 0.8397,
|
|
"mean_token_accuracy": 0.7804037094116211,
|
|
"num_tokens": 657263161.0,
|
|
"step": 6570
|
|
},
|
|
{
|
|
"entropy": 0.88828125,
|
|
"epoch": 0.8993371147406547,
|
|
"grad_norm": 0.30804080000645184,
|
|
"learning_rate": 2.8374665351556997e-06,
|
|
"loss": 0.8873,
|
|
"mean_token_accuracy": 0.7701047152280808,
|
|
"num_tokens": 658229030.0,
|
|
"step": 6580
|
|
},
|
|
{
|
|
"entropy": 0.898828125,
|
|
"epoch": 0.9007038884712636,
|
|
"grad_norm": 0.28306072528043635,
|
|
"learning_rate": 2.8339439199661832e-06,
|
|
"loss": 0.8923,
|
|
"mean_token_accuracy": 0.7718476355075836,
|
|
"num_tokens": 659270232.0,
|
|
"step": 6590
|
|
},
|
|
{
|
|
"entropy": 0.904296875,
|
|
"epoch": 0.9020706622018725,
|
|
"grad_norm": 0.3056476181703863,
|
|
"learning_rate": 2.8304213047766663e-06,
|
|
"loss": 0.9067,
|
|
"mean_token_accuracy": 0.7684412777423859,
|
|
"num_tokens": 660230684.0,
|
|
"step": 6600
|
|
},
|
|
{
|
|
"entropy": 0.8580078125,
|
|
"epoch": 0.9034374359324814,
|
|
"grad_norm": 0.27203694147290014,
|
|
"learning_rate": 2.82689868958715e-06,
|
|
"loss": 0.8573,
|
|
"mean_token_accuracy": 0.7790825486183166,
|
|
"num_tokens": 661227468.0,
|
|
"step": 6610
|
|
},
|
|
{
|
|
"entropy": 0.8751953125,
|
|
"epoch": 0.9048042096630903,
|
|
"grad_norm": 0.30854344301436437,
|
|
"learning_rate": 2.823376074397633e-06,
|
|
"loss": 0.8658,
|
|
"mean_token_accuracy": 0.7741186052560807,
|
|
"num_tokens": 662247616.0,
|
|
"step": 6620
|
|
},
|
|
{
|
|
"entropy": 0.873828125,
|
|
"epoch": 0.9061709833936992,
|
|
"grad_norm": 0.27355425006623807,
|
|
"learning_rate": 2.819853459208116e-06,
|
|
"loss": 0.8617,
|
|
"mean_token_accuracy": 0.7762023210525513,
|
|
"num_tokens": 663208530.0,
|
|
"step": 6630
|
|
},
|
|
{
|
|
"entropy": 0.9302734375,
|
|
"epoch": 0.9075377571243081,
|
|
"grad_norm": 0.281809787740782,
|
|
"learning_rate": 2.8163308440186e-06,
|
|
"loss": 0.93,
|
|
"mean_token_accuracy": 0.7628299981355667,
|
|
"num_tokens": 664207117.0,
|
|
"step": 6640
|
|
},
|
|
{
|
|
"entropy": 0.8904296875,
|
|
"epoch": 0.908904530854917,
|
|
"grad_norm": 0.29534489010424364,
|
|
"learning_rate": 2.8128082288290832e-06,
|
|
"loss": 0.886,
|
|
"mean_token_accuracy": 0.7707959443330765,
|
|
"num_tokens": 665189465.0,
|
|
"step": 6650
|
|
},
|
|
{
|
|
"entropy": 0.903125,
|
|
"epoch": 0.9102713045855259,
|
|
"grad_norm": 0.29792257595004895,
|
|
"learning_rate": 2.8092856136395664e-06,
|
|
"loss": 0.8962,
|
|
"mean_token_accuracy": 0.7689910411834717,
|
|
"num_tokens": 666199452.0,
|
|
"step": 6660
|
|
},
|
|
{
|
|
"entropy": 0.8958984375,
|
|
"epoch": 0.9116380783161347,
|
|
"grad_norm": 0.29168940921223036,
|
|
"learning_rate": 2.8057629984500495e-06,
|
|
"loss": 0.891,
|
|
"mean_token_accuracy": 0.7687996655702591,
|
|
"num_tokens": 667240091.0,
|
|
"step": 6670
|
|
},
|
|
{
|
|
"entropy": 0.876953125,
|
|
"epoch": 0.9130048520467436,
|
|
"grad_norm": 0.26545718627894493,
|
|
"learning_rate": 2.8022403832605326e-06,
|
|
"loss": 0.8645,
|
|
"mean_token_accuracy": 0.7762909710407258,
|
|
"num_tokens": 668253541.0,
|
|
"step": 6680
|
|
},
|
|
{
|
|
"entropy": 0.8939453125,
|
|
"epoch": 0.9143716257773525,
|
|
"grad_norm": 0.2595287114976205,
|
|
"learning_rate": 2.7987177680710157e-06,
|
|
"loss": 0.8929,
|
|
"mean_token_accuracy": 0.7713271856307984,
|
|
"num_tokens": 669273366.0,
|
|
"step": 6690
|
|
},
|
|
{
|
|
"entropy": 0.8962890625,
|
|
"epoch": 0.9157383995079614,
|
|
"grad_norm": 0.32906407143089933,
|
|
"learning_rate": 2.7951951528814997e-06,
|
|
"loss": 0.8998,
|
|
"mean_token_accuracy": 0.7679939687252044,
|
|
"num_tokens": 670296894.0,
|
|
"step": 6700
|
|
},
|
|
{
|
|
"entropy": 0.9021484375,
|
|
"epoch": 0.9171051732385703,
|
|
"grad_norm": 0.25862394967914376,
|
|
"learning_rate": 2.791672537691983e-06,
|
|
"loss": 0.9103,
|
|
"mean_token_accuracy": 0.7646872192621231,
|
|
"num_tokens": 671344584.0,
|
|
"step": 6710
|
|
},
|
|
{
|
|
"entropy": 0.8955078125,
|
|
"epoch": 0.9184719469691792,
|
|
"grad_norm": 0.2911314699017373,
|
|
"learning_rate": 2.788149922502466e-06,
|
|
"loss": 0.8853,
|
|
"mean_token_accuracy": 0.7729169666767121,
|
|
"num_tokens": 672327599.0,
|
|
"step": 6720
|
|
},
|
|
{
|
|
"entropy": 0.88359375,
|
|
"epoch": 0.9198387206997881,
|
|
"grad_norm": 0.32281476381936064,
|
|
"learning_rate": 2.784627307312949e-06,
|
|
"loss": 0.8829,
|
|
"mean_token_accuracy": 0.7730214953422546,
|
|
"num_tokens": 673345908.0,
|
|
"step": 6730
|
|
},
|
|
{
|
|
"entropy": 0.8609375,
|
|
"epoch": 0.921205494430397,
|
|
"grad_norm": 0.29675241685103926,
|
|
"learning_rate": 2.7811046921234326e-06,
|
|
"loss": 0.8489,
|
|
"mean_token_accuracy": 0.776990532875061,
|
|
"num_tokens": 674352632.0,
|
|
"step": 6740
|
|
},
|
|
{
|
|
"entropy": 0.9177734375,
|
|
"epoch": 0.9225722681610059,
|
|
"grad_norm": 0.3026945719259425,
|
|
"learning_rate": 2.777582076933916e-06,
|
|
"loss": 0.9108,
|
|
"mean_token_accuracy": 0.7681530892848969,
|
|
"num_tokens": 675334845.0,
|
|
"step": 6750
|
|
},
|
|
{
|
|
"entropy": 0.878515625,
|
|
"epoch": 0.9239390418916148,
|
|
"grad_norm": 0.24011723447479955,
|
|
"learning_rate": 2.7740594617443993e-06,
|
|
"loss": 0.872,
|
|
"mean_token_accuracy": 0.7727409988641739,
|
|
"num_tokens": 676355454.0,
|
|
"step": 6760
|
|
},
|
|
{
|
|
"entropy": 0.8708984375,
|
|
"epoch": 0.9253058156222237,
|
|
"grad_norm": 0.4323593699681083,
|
|
"learning_rate": 2.770536846554883e-06,
|
|
"loss": 0.8631,
|
|
"mean_token_accuracy": 0.7772344887256623,
|
|
"num_tokens": 677398895.0,
|
|
"step": 6770
|
|
},
|
|
{
|
|
"entropy": 0.9107421875,
|
|
"epoch": 0.9266725893528326,
|
|
"grad_norm": 0.2650944830497955,
|
|
"learning_rate": 2.767014231365366e-06,
|
|
"loss": 0.9104,
|
|
"mean_token_accuracy": 0.767673271894455,
|
|
"num_tokens": 678386628.0,
|
|
"step": 6780
|
|
},
|
|
{
|
|
"entropy": 0.89765625,
|
|
"epoch": 0.9280393630834415,
|
|
"grad_norm": 0.33385965495388525,
|
|
"learning_rate": 2.763491616175849e-06,
|
|
"loss": 0.8866,
|
|
"mean_token_accuracy": 0.7709019035100937,
|
|
"num_tokens": 679438210.0,
|
|
"step": 6790
|
|
},
|
|
{
|
|
"entropy": 0.880859375,
|
|
"epoch": 0.9294061368140505,
|
|
"grad_norm": 0.31328697636479763,
|
|
"learning_rate": 2.759969000986332e-06,
|
|
"loss": 0.8765,
|
|
"mean_token_accuracy": 0.772159281373024,
|
|
"num_tokens": 680454998.0,
|
|
"step": 6800
|
|
},
|
|
{
|
|
"entropy": 0.846484375,
|
|
"epoch": 0.9307729105446594,
|
|
"grad_norm": 0.3950960802944359,
|
|
"learning_rate": 2.756446385796816e-06,
|
|
"loss": 0.8403,
|
|
"mean_token_accuracy": 0.7801008284091949,
|
|
"num_tokens": 681444975.0,
|
|
"step": 6810
|
|
},
|
|
{
|
|
"entropy": 0.876953125,
|
|
"epoch": 0.9321396842752683,
|
|
"grad_norm": 0.27568930319618873,
|
|
"learning_rate": 2.7529237706072993e-06,
|
|
"loss": 0.8648,
|
|
"mean_token_accuracy": 0.775461059808731,
|
|
"num_tokens": 682457981.0,
|
|
"step": 6820
|
|
},
|
|
{
|
|
"entropy": 0.883203125,
|
|
"epoch": 0.9335064580058772,
|
|
"grad_norm": 0.2776775550861942,
|
|
"learning_rate": 2.7494011554177824e-06,
|
|
"loss": 0.8743,
|
|
"mean_token_accuracy": 0.7739621192216873,
|
|
"num_tokens": 683453311.0,
|
|
"step": 6830
|
|
},
|
|
{
|
|
"entropy": 0.9166015625,
|
|
"epoch": 0.9348732317364861,
|
|
"grad_norm": 0.31067619263828444,
|
|
"learning_rate": 2.7458785402282656e-06,
|
|
"loss": 0.9087,
|
|
"mean_token_accuracy": 0.7666392415761948,
|
|
"num_tokens": 684437155.0,
|
|
"step": 6840
|
|
},
|
|
{
|
|
"entropy": 0.9150390625,
|
|
"epoch": 0.936240005467095,
|
|
"grad_norm": 0.2984813661795795,
|
|
"learning_rate": 2.7423559250387487e-06,
|
|
"loss": 0.9046,
|
|
"mean_token_accuracy": 0.7674717724323272,
|
|
"num_tokens": 685428193.0,
|
|
"step": 6850
|
|
},
|
|
{
|
|
"entropy": 0.8890625,
|
|
"epoch": 0.9376067791977039,
|
|
"grad_norm": 0.2800601896553872,
|
|
"learning_rate": 2.7388333098492326e-06,
|
|
"loss": 0.8711,
|
|
"mean_token_accuracy": 0.7747564524412155,
|
|
"num_tokens": 686363719.0,
|
|
"step": 6860
|
|
},
|
|
{
|
|
"entropy": 0.896875,
|
|
"epoch": 0.9389735529283127,
|
|
"grad_norm": 0.2930705400044768,
|
|
"learning_rate": 2.7353106946597158e-06,
|
|
"loss": 0.8924,
|
|
"mean_token_accuracy": 0.7689283937215805,
|
|
"num_tokens": 687423240.0,
|
|
"step": 6870
|
|
},
|
|
{
|
|
"entropy": 0.9044921875,
|
|
"epoch": 0.9403403266589216,
|
|
"grad_norm": 0.33408921493823035,
|
|
"learning_rate": 2.731788079470199e-06,
|
|
"loss": 0.9086,
|
|
"mean_token_accuracy": 0.7646285116672515,
|
|
"num_tokens": 688380864.0,
|
|
"step": 6880
|
|
},
|
|
{
|
|
"entropy": 0.8951171875,
|
|
"epoch": 0.9417071003895305,
|
|
"grad_norm": 0.2819055689432927,
|
|
"learning_rate": 2.728265464280682e-06,
|
|
"loss": 0.8954,
|
|
"mean_token_accuracy": 0.7688406348228455,
|
|
"num_tokens": 689366091.0,
|
|
"step": 6890
|
|
},
|
|
{
|
|
"entropy": 0.89296875,
|
|
"epoch": 0.9430738741201394,
|
|
"grad_norm": 0.2842402563834125,
|
|
"learning_rate": 2.7247428490911656e-06,
|
|
"loss": 0.896,
|
|
"mean_token_accuracy": 0.7701351910829544,
|
|
"num_tokens": 690330964.0,
|
|
"step": 6900
|
|
},
|
|
{
|
|
"entropy": 0.8990234375,
|
|
"epoch": 0.9444406478507483,
|
|
"grad_norm": 0.3124791948393444,
|
|
"learning_rate": 2.7212202339016487e-06,
|
|
"loss": 0.8968,
|
|
"mean_token_accuracy": 0.7703346192836762,
|
|
"num_tokens": 691337981.0,
|
|
"step": 6910
|
|
},
|
|
{
|
|
"entropy": 0.8796875,
|
|
"epoch": 0.9458074215813572,
|
|
"grad_norm": 0.25061472204847435,
|
|
"learning_rate": 2.7176976187121322e-06,
|
|
"loss": 0.8804,
|
|
"mean_token_accuracy": 0.7713771134614944,
|
|
"num_tokens": 692402156.0,
|
|
"step": 6920
|
|
},
|
|
{
|
|
"entropy": 0.905078125,
|
|
"epoch": 0.9471741953119661,
|
|
"grad_norm": 0.33919845064541126,
|
|
"learning_rate": 2.7141750035226154e-06,
|
|
"loss": 0.8959,
|
|
"mean_token_accuracy": 0.7690528512001038,
|
|
"num_tokens": 693447012.0,
|
|
"step": 6930
|
|
},
|
|
{
|
|
"entropy": 0.8939453125,
|
|
"epoch": 0.948540969042575,
|
|
"grad_norm": 0.29541549831035635,
|
|
"learning_rate": 2.710652388333099e-06,
|
|
"loss": 0.8894,
|
|
"mean_token_accuracy": 0.7711980164051055,
|
|
"num_tokens": 694464422.0,
|
|
"step": 6940
|
|
},
|
|
{
|
|
"entropy": 0.9013671875,
|
|
"epoch": 0.9499077427731839,
|
|
"grad_norm": 0.2764537617117485,
|
|
"learning_rate": 2.707129773143582e-06,
|
|
"loss": 0.9065,
|
|
"mean_token_accuracy": 0.7658989787101745,
|
|
"num_tokens": 695496716.0,
|
|
"step": 6950
|
|
},
|
|
{
|
|
"entropy": 0.8603515625,
|
|
"epoch": 0.9512745165037928,
|
|
"grad_norm": 0.2972816813684828,
|
|
"learning_rate": 2.703607157954065e-06,
|
|
"loss": 0.8541,
|
|
"mean_token_accuracy": 0.7776772379875183,
|
|
"num_tokens": 696522504.0,
|
|
"step": 6960
|
|
},
|
|
{
|
|
"entropy": 0.924609375,
|
|
"epoch": 0.9526412902344017,
|
|
"grad_norm": 0.2920282532897111,
|
|
"learning_rate": 2.7000845427645483e-06,
|
|
"loss": 0.9204,
|
|
"mean_token_accuracy": 0.7649168133735657,
|
|
"num_tokens": 697477258.0,
|
|
"step": 6970
|
|
},
|
|
{
|
|
"entropy": 0.8677734375,
|
|
"epoch": 0.9540080639650106,
|
|
"grad_norm": 0.2898836167537822,
|
|
"learning_rate": 2.6965619275750322e-06,
|
|
"loss": 0.8698,
|
|
"mean_token_accuracy": 0.7733672797679901,
|
|
"num_tokens": 698459724.0,
|
|
"step": 6980
|
|
},
|
|
{
|
|
"entropy": 0.9255859375,
|
|
"epoch": 0.9553748376956195,
|
|
"grad_norm": 0.3069966605335132,
|
|
"learning_rate": 2.6930393123855154e-06,
|
|
"loss": 0.9328,
|
|
"mean_token_accuracy": 0.7634543091058731,
|
|
"num_tokens": 699484062.0,
|
|
"step": 6990
|
|
},
|
|
{
|
|
"entropy": 0.9328125,
|
|
"epoch": 0.9567416114262284,
|
|
"grad_norm": 0.28253706197286943,
|
|
"learning_rate": 2.6895166971959985e-06,
|
|
"loss": 0.9315,
|
|
"mean_token_accuracy": 0.7622436374425888,
|
|
"num_tokens": 700454967.0,
|
|
"step": 7000
|
|
},
|
|
{
|
|
"entropy": 0.8595703125,
|
|
"epoch": 0.9581083851568373,
|
|
"grad_norm": 0.2805137464364089,
|
|
"learning_rate": 2.6859940820064816e-06,
|
|
"loss": 0.855,
|
|
"mean_token_accuracy": 0.7790907323360443,
|
|
"num_tokens": 701442196.0,
|
|
"step": 7010
|
|
},
|
|
{
|
|
"entropy": 0.919140625,
|
|
"epoch": 0.9594751588874462,
|
|
"grad_norm": 0.3725668653872117,
|
|
"learning_rate": 2.6824714668169647e-06,
|
|
"loss": 0.9137,
|
|
"mean_token_accuracy": 0.7659894198179245,
|
|
"num_tokens": 702419631.0,
|
|
"step": 7020
|
|
},
|
|
{
|
|
"entropy": 0.8984375,
|
|
"epoch": 0.9608419326180551,
|
|
"grad_norm": 0.28286599947344093,
|
|
"learning_rate": 2.6789488516274487e-06,
|
|
"loss": 0.8954,
|
|
"mean_token_accuracy": 0.7688846826553345,
|
|
"num_tokens": 703488927.0,
|
|
"step": 7030
|
|
},
|
|
{
|
|
"entropy": 0.8892578125,
|
|
"epoch": 0.962208706348664,
|
|
"grad_norm": 0.3102358955179633,
|
|
"learning_rate": 2.675426236437932e-06,
|
|
"loss": 0.8902,
|
|
"mean_token_accuracy": 0.7722076952457428,
|
|
"num_tokens": 704531284.0,
|
|
"step": 7040
|
|
},
|
|
{
|
|
"entropy": 0.8716796875,
|
|
"epoch": 0.9635754800792729,
|
|
"grad_norm": 0.32226608872818374,
|
|
"learning_rate": 2.671903621248415e-06,
|
|
"loss": 0.8746,
|
|
"mean_token_accuracy": 0.7722139030694961,
|
|
"num_tokens": 705501913.0,
|
|
"step": 7050
|
|
},
|
|
{
|
|
"entropy": 0.88671875,
|
|
"epoch": 0.9649422538098817,
|
|
"grad_norm": 0.3023300869324572,
|
|
"learning_rate": 2.668381006058898e-06,
|
|
"loss": 0.8819,
|
|
"mean_token_accuracy": 0.7708177655935288,
|
|
"num_tokens": 706485055.0,
|
|
"step": 7060
|
|
},
|
|
{
|
|
"entropy": 0.9001953125,
|
|
"epoch": 0.9663090275404906,
|
|
"grad_norm": 0.2854433473703563,
|
|
"learning_rate": 2.6648583908693816e-06,
|
|
"loss": 0.8882,
|
|
"mean_token_accuracy": 0.7698571592569351,
|
|
"num_tokens": 707478154.0,
|
|
"step": 7070
|
|
},
|
|
{
|
|
"entropy": 0.931640625,
|
|
"epoch": 0.9676758012710995,
|
|
"grad_norm": 0.31574456483464547,
|
|
"learning_rate": 2.6613357756798648e-06,
|
|
"loss": 0.9347,
|
|
"mean_token_accuracy": 0.7620939552783966,
|
|
"num_tokens": 708467999.0,
|
|
"step": 7080
|
|
},
|
|
{
|
|
"entropy": 0.8857421875,
|
|
"epoch": 0.9690425750017084,
|
|
"grad_norm": 0.28330940130453375,
|
|
"learning_rate": 2.6578131604903483e-06,
|
|
"loss": 0.8831,
|
|
"mean_token_accuracy": 0.7723428666591644,
|
|
"num_tokens": 709429624.0,
|
|
"step": 7090
|
|
},
|
|
{
|
|
"entropy": 0.8810546875,
|
|
"epoch": 0.9704093487323173,
|
|
"grad_norm": 0.30291449000066883,
|
|
"learning_rate": 2.6542905453008314e-06,
|
|
"loss": 0.8704,
|
|
"mean_token_accuracy": 0.7757569700479507,
|
|
"num_tokens": 710434764.0,
|
|
"step": 7100
|
|
},
|
|
{
|
|
"entropy": 0.8939453125,
|
|
"epoch": 0.9717761224629262,
|
|
"grad_norm": 0.26023146104271777,
|
|
"learning_rate": 2.650767930111315e-06,
|
|
"loss": 0.8835,
|
|
"mean_token_accuracy": 0.7726508110761643,
|
|
"num_tokens": 711448154.0,
|
|
"step": 7110
|
|
},
|
|
{
|
|
"entropy": 0.8978515625,
|
|
"epoch": 0.9731428961935351,
|
|
"grad_norm": 0.3189640759620763,
|
|
"learning_rate": 2.647245314921798e-06,
|
|
"loss": 0.8948,
|
|
"mean_token_accuracy": 0.7695181220769882,
|
|
"num_tokens": 712485461.0,
|
|
"step": 7120
|
|
},
|
|
{
|
|
"entropy": 0.8783203125,
|
|
"epoch": 0.974509669924144,
|
|
"grad_norm": 0.26519879336361546,
|
|
"learning_rate": 2.6437226997322812e-06,
|
|
"loss": 0.8662,
|
|
"mean_token_accuracy": 0.7753661274909973,
|
|
"num_tokens": 713537375.0,
|
|
"step": 7130
|
|
},
|
|
{
|
|
"entropy": 0.86953125,
|
|
"epoch": 0.9758764436547529,
|
|
"grad_norm": 0.3177118896243432,
|
|
"learning_rate": 2.640200084542765e-06,
|
|
"loss": 0.8574,
|
|
"mean_token_accuracy": 0.7752474278211594,
|
|
"num_tokens": 714549530.0,
|
|
"step": 7140
|
|
},
|
|
{
|
|
"entropy": 0.881640625,
|
|
"epoch": 0.9772432173853619,
|
|
"grad_norm": 0.3607479135018395,
|
|
"learning_rate": 2.6366774693532483e-06,
|
|
"loss": 0.8858,
|
|
"mean_token_accuracy": 0.7718848347663879,
|
|
"num_tokens": 715478350.0,
|
|
"step": 7150
|
|
},
|
|
{
|
|
"entropy": 0.90625,
|
|
"epoch": 0.9786099911159708,
|
|
"grad_norm": 0.2969104456771781,
|
|
"learning_rate": 2.6331548541637314e-06,
|
|
"loss": 0.9121,
|
|
"mean_token_accuracy": 0.7657370775938034,
|
|
"num_tokens": 716476986.0,
|
|
"step": 7160
|
|
},
|
|
{
|
|
"entropy": 0.871484375,
|
|
"epoch": 0.9799767648465797,
|
|
"grad_norm": 0.2518831659390098,
|
|
"learning_rate": 2.6296322389742146e-06,
|
|
"loss": 0.8668,
|
|
"mean_token_accuracy": 0.7762292563915253,
|
|
"num_tokens": 717495137.0,
|
|
"step": 7170
|
|
},
|
|
{
|
|
"entropy": 0.8595703125,
|
|
"epoch": 0.9813435385771886,
|
|
"grad_norm": 0.27349409070276265,
|
|
"learning_rate": 2.6261096237846977e-06,
|
|
"loss": 0.8514,
|
|
"mean_token_accuracy": 0.7770450711250305,
|
|
"num_tokens": 718485758.0,
|
|
"step": 7180
|
|
},
|
|
{
|
|
"entropy": 0.8736328125,
|
|
"epoch": 0.9827103123077975,
|
|
"grad_norm": 0.2835558871395632,
|
|
"learning_rate": 2.622587008595181e-06,
|
|
"loss": 0.8708,
|
|
"mean_token_accuracy": 0.7751644968986511,
|
|
"num_tokens": 719460644.0,
|
|
"step": 7190
|
|
},
|
|
{
|
|
"entropy": 0.900390625,
|
|
"epoch": 0.9840770860384064,
|
|
"grad_norm": 0.3120513785607619,
|
|
"learning_rate": 2.6190643934056648e-06,
|
|
"loss": 0.8884,
|
|
"mean_token_accuracy": 0.7687642008066178,
|
|
"num_tokens": 720475792.0,
|
|
"step": 7200
|
|
},
|
|
{
|
|
"entropy": 0.9283203125,
|
|
"epoch": 0.9854438597690153,
|
|
"grad_norm": 0.3107641216687444,
|
|
"learning_rate": 2.615541778216148e-06,
|
|
"loss": 0.9275,
|
|
"mean_token_accuracy": 0.7628724306821824,
|
|
"num_tokens": 721453125.0,
|
|
"step": 7210
|
|
},
|
|
{
|
|
"entropy": 0.8818359375,
|
|
"epoch": 0.9868106334996242,
|
|
"grad_norm": 0.2838970406310771,
|
|
"learning_rate": 2.612019163026631e-06,
|
|
"loss": 0.8764,
|
|
"mean_token_accuracy": 0.7744882971048355,
|
|
"num_tokens": 722432480.0,
|
|
"step": 7220
|
|
},
|
|
{
|
|
"entropy": 0.93203125,
|
|
"epoch": 0.9881774072302331,
|
|
"grad_norm": 0.3277479932357939,
|
|
"learning_rate": 2.6084965478371146e-06,
|
|
"loss": 0.9385,
|
|
"mean_token_accuracy": 0.7612004101276397,
|
|
"num_tokens": 723432525.0,
|
|
"step": 7230
|
|
},
|
|
{
|
|
"entropy": 0.863671875,
|
|
"epoch": 0.989544180960842,
|
|
"grad_norm": 0.26438411925029226,
|
|
"learning_rate": 2.6049739326475977e-06,
|
|
"loss": 0.8675,
|
|
"mean_token_accuracy": 0.7770102679729461,
|
|
"num_tokens": 724435158.0,
|
|
"step": 7240
|
|
},
|
|
{
|
|
"entropy": 0.8921875,
|
|
"epoch": 0.9909109546914509,
|
|
"grad_norm": 0.3197352740221935,
|
|
"learning_rate": 2.6014513174580812e-06,
|
|
"loss": 0.8858,
|
|
"mean_token_accuracy": 0.7701819777488709,
|
|
"num_tokens": 725410072.0,
|
|
"step": 7250
|
|
},
|
|
{
|
|
"entropy": 0.88671875,
|
|
"epoch": 0.9922777284220597,
|
|
"grad_norm": 0.2799990214007194,
|
|
"learning_rate": 2.5979287022685644e-06,
|
|
"loss": 0.891,
|
|
"mean_token_accuracy": 0.7723621696233749,
|
|
"num_tokens": 726377155.0,
|
|
"step": 7260
|
|
},
|
|
{
|
|
"entropy": 0.85625,
|
|
"epoch": 0.9936445021526686,
|
|
"grad_norm": 0.28282598030909306,
|
|
"learning_rate": 2.594406087079048e-06,
|
|
"loss": 0.8483,
|
|
"mean_token_accuracy": 0.7782673209905624,
|
|
"num_tokens": 727361527.0,
|
|
"step": 7270
|
|
},
|
|
{
|
|
"entropy": 0.9099609375,
|
|
"epoch": 0.9950112758832775,
|
|
"grad_norm": 0.2724657441785446,
|
|
"learning_rate": 2.590883471889531e-06,
|
|
"loss": 0.9073,
|
|
"mean_token_accuracy": 0.7670032143592834,
|
|
"num_tokens": 728344461.0,
|
|
"step": 7280
|
|
},
|
|
{
|
|
"entropy": 0.88046875,
|
|
"epoch": 0.9963780496138864,
|
|
"grad_norm": 0.29247448799361486,
|
|
"learning_rate": 2.587360856700014e-06,
|
|
"loss": 0.8758,
|
|
"mean_token_accuracy": 0.7732854515314103,
|
|
"num_tokens": 729330460.0,
|
|
"step": 7290
|
|
},
|
|
{
|
|
"entropy": 0.8783203125,
|
|
"epoch": 0.9977448233444953,
|
|
"grad_norm": 0.27730582819903415,
|
|
"learning_rate": 2.5838382415104973e-06,
|
|
"loss": 0.8738,
|
|
"mean_token_accuracy": 0.7739197134971618,
|
|
"num_tokens": 730339383.0,
|
|
"step": 7300
|
|
},
|
|
{
|
|
"entropy": 0.9197265625,
|
|
"epoch": 0.9991115970751042,
|
|
"grad_norm": 0.274409741899104,
|
|
"learning_rate": 2.5803156263209813e-06,
|
|
"loss": 0.9221,
|
|
"mean_token_accuracy": 0.7632205307483673,
|
|
"num_tokens": 731309956.0,
|
|
"step": 7310
|
|
},
|
|
{
|
|
"entropy": 0.919202302631579,
|
|
"epoch": 1.0004100321191827,
|
|
"grad_norm": 0.2929611990886528,
|
|
"learning_rate": 2.5767930111314644e-06,
|
|
"loss": 0.9136,
|
|
"mean_token_accuracy": 0.7637978415740164,
|
|
"num_tokens": 732312529.0,
|
|
"step": 7320
|
|
},
|
|
{
|
|
"entropy": 0.8876953125,
|
|
"epoch": 1.0017768058497916,
|
|
"grad_norm": 0.27387594600290566,
|
|
"learning_rate": 2.5732703959419475e-06,
|
|
"loss": 0.8825,
|
|
"mean_token_accuracy": 0.7729101330041885,
|
|
"num_tokens": 733269565.0,
|
|
"step": 7330
|
|
},
|
|
{
|
|
"entropy": 0.8615234375,
|
|
"epoch": 1.0031435795804005,
|
|
"grad_norm": 0.2694210494482208,
|
|
"learning_rate": 2.5697477807524306e-06,
|
|
"loss": 0.8594,
|
|
"mean_token_accuracy": 0.7753072261810303,
|
|
"num_tokens": 734347586.0,
|
|
"step": 7340
|
|
},
|
|
{
|
|
"entropy": 0.843359375,
|
|
"epoch": 1.0045103533110094,
|
|
"grad_norm": 0.30413836717336756,
|
|
"learning_rate": 2.5662251655629138e-06,
|
|
"loss": 0.8308,
|
|
"mean_token_accuracy": 0.7828973293304443,
|
|
"num_tokens": 735348584.0,
|
|
"step": 7350
|
|
},
|
|
{
|
|
"entropy": 0.901171875,
|
|
"epoch": 1.0058771270416182,
|
|
"grad_norm": 0.31849698680683114,
|
|
"learning_rate": 2.5627025503733977e-06,
|
|
"loss": 0.8989,
|
|
"mean_token_accuracy": 0.7665347427129745,
|
|
"num_tokens": 736363999.0,
|
|
"step": 7360
|
|
},
|
|
{
|
|
"entropy": 0.9013671875,
|
|
"epoch": 1.0072439007722271,
|
|
"grad_norm": 0.3351369249417876,
|
|
"learning_rate": 2.559179935183881e-06,
|
|
"loss": 0.8878,
|
|
"mean_token_accuracy": 0.7716638177633286,
|
|
"num_tokens": 737347845.0,
|
|
"step": 7370
|
|
},
|
|
{
|
|
"entropy": 0.903125,
|
|
"epoch": 1.008610674502836,
|
|
"grad_norm": 0.3047294852807805,
|
|
"learning_rate": 2.555657319994364e-06,
|
|
"loss": 0.9025,
|
|
"mean_token_accuracy": 0.7685324668884277,
|
|
"num_tokens": 738338641.0,
|
|
"step": 7380
|
|
},
|
|
{
|
|
"entropy": 0.8751953125,
|
|
"epoch": 1.009977448233445,
|
|
"grad_norm": 0.2771813409908932,
|
|
"learning_rate": 2.552134704804847e-06,
|
|
"loss": 0.859,
|
|
"mean_token_accuracy": 0.7756396114826203,
|
|
"num_tokens": 739384077.0,
|
|
"step": 7390
|
|
},
|
|
{
|
|
"entropy": 0.9103515625,
|
|
"epoch": 1.0113442219640538,
|
|
"grad_norm": 0.3023259013981775,
|
|
"learning_rate": 2.5486120896153306e-06,
|
|
"loss": 0.9193,
|
|
"mean_token_accuracy": 0.7634977221488952,
|
|
"num_tokens": 740364332.0,
|
|
"step": 7400
|
|
},
|
|
{
|
|
"entropy": 0.8787109375,
|
|
"epoch": 1.0127109956946627,
|
|
"grad_norm": 0.2950185640985522,
|
|
"learning_rate": 2.5450894744258138e-06,
|
|
"loss": 0.8747,
|
|
"mean_token_accuracy": 0.7760294824838638,
|
|
"num_tokens": 741361827.0,
|
|
"step": 7410
|
|
},
|
|
{
|
|
"entropy": 0.843359375,
|
|
"epoch": 1.0140777694252716,
|
|
"grad_norm": 0.288177564673061,
|
|
"learning_rate": 2.5415668592362973e-06,
|
|
"loss": 0.8366,
|
|
"mean_token_accuracy": 0.7819704860448837,
|
|
"num_tokens": 742316425.0,
|
|
"step": 7420
|
|
},
|
|
{
|
|
"entropy": 0.8654296875,
|
|
"epoch": 1.0154445431558805,
|
|
"grad_norm": 0.27373534423810025,
|
|
"learning_rate": 2.5380442440467804e-06,
|
|
"loss": 0.8673,
|
|
"mean_token_accuracy": 0.7784995615482331,
|
|
"num_tokens": 743335553.0,
|
|
"step": 7430
|
|
},
|
|
{
|
|
"entropy": 0.9287109375,
|
|
"epoch": 1.0168113168864894,
|
|
"grad_norm": 0.32343014331382086,
|
|
"learning_rate": 2.534521628857264e-06,
|
|
"loss": 0.925,
|
|
"mean_token_accuracy": 0.761812487244606,
|
|
"num_tokens": 744322171.0,
|
|
"step": 7440
|
|
},
|
|
{
|
|
"entropy": 0.8853515625,
|
|
"epoch": 1.0181780906170983,
|
|
"grad_norm": 0.253983532133421,
|
|
"learning_rate": 2.530999013667747e-06,
|
|
"loss": 0.8813,
|
|
"mean_token_accuracy": 0.7729214519262314,
|
|
"num_tokens": 745302213.0,
|
|
"step": 7450
|
|
},
|
|
{
|
|
"entropy": 0.958203125,
|
|
"epoch": 1.0195448643477072,
|
|
"grad_norm": 0.31086219624478534,
|
|
"learning_rate": 2.5274763984782302e-06,
|
|
"loss": 0.9472,
|
|
"mean_token_accuracy": 0.761288532614708,
|
|
"num_tokens": 746297728.0,
|
|
"step": 7460
|
|
},
|
|
{
|
|
"entropy": 0.8447265625,
|
|
"epoch": 1.020911638078316,
|
|
"grad_norm": 0.30487603952023434,
|
|
"learning_rate": 2.5239537832887138e-06,
|
|
"loss": 0.8448,
|
|
"mean_token_accuracy": 0.7810387730598449,
|
|
"num_tokens": 747340423.0,
|
|
"step": 7470
|
|
},
|
|
{
|
|
"entropy": 0.8900390625,
|
|
"epoch": 1.022278411808925,
|
|
"grad_norm": 0.2952000496603266,
|
|
"learning_rate": 2.5204311680991973e-06,
|
|
"loss": 0.8873,
|
|
"mean_token_accuracy": 0.7692166984081268,
|
|
"num_tokens": 748347782.0,
|
|
"step": 7480
|
|
},
|
|
{
|
|
"entropy": 0.85078125,
|
|
"epoch": 1.0236451855395339,
|
|
"grad_norm": 0.27432494408310615,
|
|
"learning_rate": 2.5169085529096804e-06,
|
|
"loss": 0.8497,
|
|
"mean_token_accuracy": 0.7793044775724411,
|
|
"num_tokens": 749280994.0,
|
|
"step": 7490
|
|
},
|
|
{
|
|
"entropy": 0.8982421875,
|
|
"epoch": 1.0250119592701428,
|
|
"grad_norm": 0.2948619911065857,
|
|
"learning_rate": 2.5133859377201636e-06,
|
|
"loss": 0.8952,
|
|
"mean_token_accuracy": 0.7678454935550689,
|
|
"num_tokens": 750267400.0,
|
|
"step": 7500
|
|
},
|
|
{
|
|
"entropy": 0.9224609375,
|
|
"epoch": 1.0263787330007517,
|
|
"grad_norm": 0.27792921112108776,
|
|
"learning_rate": 2.5098633225306467e-06,
|
|
"loss": 0.9186,
|
|
"mean_token_accuracy": 0.763878458738327,
|
|
"num_tokens": 751278605.0,
|
|
"step": 7510
|
|
},
|
|
{
|
|
"entropy": 0.873828125,
|
|
"epoch": 1.0277455067313606,
|
|
"grad_norm": 0.27086595679554715,
|
|
"learning_rate": 2.50634070734113e-06,
|
|
"loss": 0.8717,
|
|
"mean_token_accuracy": 0.7737829416990281,
|
|
"num_tokens": 752284553.0,
|
|
"step": 7520
|
|
},
|
|
{
|
|
"entropy": 0.8767578125,
|
|
"epoch": 1.0291122804619695,
|
|
"grad_norm": 0.2560251833805921,
|
|
"learning_rate": 2.502818092151614e-06,
|
|
"loss": 0.87,
|
|
"mean_token_accuracy": 0.7759152323007583,
|
|
"num_tokens": 753259545.0,
|
|
"step": 7530
|
|
},
|
|
{
|
|
"entropy": 0.863671875,
|
|
"epoch": 1.0304790541925783,
|
|
"grad_norm": 0.29660167124928244,
|
|
"learning_rate": 2.499295476962097e-06,
|
|
"loss": 0.85,
|
|
"mean_token_accuracy": 0.7777520507574082,
|
|
"num_tokens": 754265484.0,
|
|
"step": 7540
|
|
},
|
|
{
|
|
"entropy": 0.884375,
|
|
"epoch": 1.0318458279231872,
|
|
"grad_norm": 0.2838901429629701,
|
|
"learning_rate": 2.49577286177258e-06,
|
|
"loss": 0.8858,
|
|
"mean_token_accuracy": 0.7741414994001389,
|
|
"num_tokens": 755298143.0,
|
|
"step": 7550
|
|
},
|
|
{
|
|
"entropy": 0.8849609375,
|
|
"epoch": 1.0332126016537961,
|
|
"grad_norm": 0.30114534589439595,
|
|
"learning_rate": 2.492250246583063e-06,
|
|
"loss": 0.8833,
|
|
"mean_token_accuracy": 0.7717240184545517,
|
|
"num_tokens": 756282943.0,
|
|
"step": 7560
|
|
},
|
|
{
|
|
"entropy": 0.8724609375,
|
|
"epoch": 1.034579375384405,
|
|
"grad_norm": 0.28887965328444737,
|
|
"learning_rate": 2.4887276313935467e-06,
|
|
"loss": 0.8642,
|
|
"mean_token_accuracy": 0.7746973007917404,
|
|
"num_tokens": 757270119.0,
|
|
"step": 7570
|
|
},
|
|
{
|
|
"entropy": 0.8517578125,
|
|
"epoch": 1.035946149115014,
|
|
"grad_norm": 0.259632183598031,
|
|
"learning_rate": 2.48520501620403e-06,
|
|
"loss": 0.8467,
|
|
"mean_token_accuracy": 0.7784045845270157,
|
|
"num_tokens": 758304475.0,
|
|
"step": 7580
|
|
},
|
|
{
|
|
"entropy": 0.855078125,
|
|
"epoch": 1.0373129228456228,
|
|
"grad_norm": 0.2896330392470846,
|
|
"learning_rate": 2.4816824010145134e-06,
|
|
"loss": 0.8572,
|
|
"mean_token_accuracy": 0.7758593797683716,
|
|
"num_tokens": 759280363.0,
|
|
"step": 7590
|
|
},
|
|
{
|
|
"entropy": 0.8865234375,
|
|
"epoch": 1.0386796965762317,
|
|
"grad_norm": 0.3195956535627538,
|
|
"learning_rate": 2.478159785824997e-06,
|
|
"loss": 0.8803,
|
|
"mean_token_accuracy": 0.7723922610282898,
|
|
"num_tokens": 760341882.0,
|
|
"step": 7600
|
|
},
|
|
{
|
|
"entropy": 0.8943359375,
|
|
"epoch": 1.0400464703068406,
|
|
"grad_norm": 0.30119033033073406,
|
|
"learning_rate": 2.47463717063548e-06,
|
|
"loss": 0.901,
|
|
"mean_token_accuracy": 0.7697214096784591,
|
|
"num_tokens": 761352000.0,
|
|
"step": 7610
|
|
},
|
|
{
|
|
"entropy": 0.908984375,
|
|
"epoch": 1.0414132440374495,
|
|
"grad_norm": 0.2875719343315632,
|
|
"learning_rate": 2.4711145554459636e-06,
|
|
"loss": 0.9008,
|
|
"mean_token_accuracy": 0.7686017096042633,
|
|
"num_tokens": 762378949.0,
|
|
"step": 7620
|
|
},
|
|
{
|
|
"entropy": 0.9005859375,
|
|
"epoch": 1.0427800177680584,
|
|
"grad_norm": 0.2616347485177473,
|
|
"learning_rate": 2.4675919402564467e-06,
|
|
"loss": 0.892,
|
|
"mean_token_accuracy": 0.7696782857179642,
|
|
"num_tokens": 763382263.0,
|
|
"step": 7630
|
|
},
|
|
{
|
|
"entropy": 0.893359375,
|
|
"epoch": 1.0441467914986673,
|
|
"grad_norm": 0.2770489297935016,
|
|
"learning_rate": 2.46406932506693e-06,
|
|
"loss": 0.8896,
|
|
"mean_token_accuracy": 0.7705306947231293,
|
|
"num_tokens": 764417973.0,
|
|
"step": 7640
|
|
},
|
|
{
|
|
"entropy": 0.8583984375,
|
|
"epoch": 1.0455135652292762,
|
|
"grad_norm": 0.28843646637004644,
|
|
"learning_rate": 2.4605467098774134e-06,
|
|
"loss": 0.8525,
|
|
"mean_token_accuracy": 0.7797170639038086,
|
|
"num_tokens": 765463300.0,
|
|
"step": 7650
|
|
},
|
|
{
|
|
"entropy": 0.8662109375,
|
|
"epoch": 1.0468803389598853,
|
|
"grad_norm": 0.2668297735634067,
|
|
"learning_rate": 2.4570240946878965e-06,
|
|
"loss": 0.8648,
|
|
"mean_token_accuracy": 0.7754337161779403,
|
|
"num_tokens": 766453588.0,
|
|
"step": 7660
|
|
},
|
|
{
|
|
"entropy": 0.89375,
|
|
"epoch": 1.0482471126904942,
|
|
"grad_norm": 0.2755043193865477,
|
|
"learning_rate": 2.4535014794983796e-06,
|
|
"loss": 0.8959,
|
|
"mean_token_accuracy": 0.7700162678956985,
|
|
"num_tokens": 767445217.0,
|
|
"step": 7670
|
|
},
|
|
{
|
|
"entropy": 0.857421875,
|
|
"epoch": 1.049613886421103,
|
|
"grad_norm": 0.322136430875566,
|
|
"learning_rate": 2.449978864308863e-06,
|
|
"loss": 0.8573,
|
|
"mean_token_accuracy": 0.7772758603096008,
|
|
"num_tokens": 768410930.0,
|
|
"step": 7680
|
|
},
|
|
{
|
|
"entropy": 0.88125,
|
|
"epoch": 1.050980660151712,
|
|
"grad_norm": 0.2947669192846649,
|
|
"learning_rate": 2.4464562491193463e-06,
|
|
"loss": 0.8732,
|
|
"mean_token_accuracy": 0.7733403265476226,
|
|
"num_tokens": 769384267.0,
|
|
"step": 7690
|
|
},
|
|
{
|
|
"entropy": 0.9025390625,
|
|
"epoch": 1.0523474338823209,
|
|
"grad_norm": 0.2972805252465549,
|
|
"learning_rate": 2.4429336339298294e-06,
|
|
"loss": 0.9149,
|
|
"mean_token_accuracy": 0.7671808689832688,
|
|
"num_tokens": 770407371.0,
|
|
"step": 7700
|
|
},
|
|
{
|
|
"entropy": 0.837109375,
|
|
"epoch": 1.0537142076129298,
|
|
"grad_norm": 0.2983682490742246,
|
|
"learning_rate": 2.439411018740313e-06,
|
|
"loss": 0.8316,
|
|
"mean_token_accuracy": 0.7804022431373596,
|
|
"num_tokens": 771368434.0,
|
|
"step": 7710
|
|
},
|
|
{
|
|
"entropy": 0.84296875,
|
|
"epoch": 1.0550809813435387,
|
|
"grad_norm": 0.2884981471701354,
|
|
"learning_rate": 2.435888403550796e-06,
|
|
"loss": 0.8311,
|
|
"mean_token_accuracy": 0.7811567425727844,
|
|
"num_tokens": 772343303.0,
|
|
"step": 7720
|
|
},
|
|
{
|
|
"entropy": 0.893359375,
|
|
"epoch": 1.0564477550741476,
|
|
"grad_norm": 0.5228241720561636,
|
|
"learning_rate": 2.4323657883612797e-06,
|
|
"loss": 0.8985,
|
|
"mean_token_accuracy": 0.7698929309844971,
|
|
"num_tokens": 773356381.0,
|
|
"step": 7730
|
|
},
|
|
{
|
|
"entropy": 0.89765625,
|
|
"epoch": 1.0578145288047565,
|
|
"grad_norm": 0.2851377352883215,
|
|
"learning_rate": 2.4288431731717628e-06,
|
|
"loss": 0.898,
|
|
"mean_token_accuracy": 0.7694161891937256,
|
|
"num_tokens": 774321962.0,
|
|
"step": 7740
|
|
},
|
|
{
|
|
"entropy": 0.8796875,
|
|
"epoch": 1.0591813025353654,
|
|
"grad_norm": 0.3112341211398025,
|
|
"learning_rate": 2.4253205579822463e-06,
|
|
"loss": 0.8761,
|
|
"mean_token_accuracy": 0.7717499494552612,
|
|
"num_tokens": 775292597.0,
|
|
"step": 7750
|
|
},
|
|
{
|
|
"entropy": 0.8740234375,
|
|
"epoch": 1.0605480762659742,
|
|
"grad_norm": 0.2906889205282676,
|
|
"learning_rate": 2.4217979427927294e-06,
|
|
"loss": 0.861,
|
|
"mean_token_accuracy": 0.7779184758663178,
|
|
"num_tokens": 776256196.0,
|
|
"step": 7760
|
|
},
|
|
{
|
|
"entropy": 0.849609375,
|
|
"epoch": 1.0619148499965831,
|
|
"grad_norm": 0.2586480847538262,
|
|
"learning_rate": 2.418275327603213e-06,
|
|
"loss": 0.8493,
|
|
"mean_token_accuracy": 0.7787239968776702,
|
|
"num_tokens": 777241987.0,
|
|
"step": 7770
|
|
},
|
|
{
|
|
"entropy": 0.8716796875,
|
|
"epoch": 1.063281623727192,
|
|
"grad_norm": 0.3207389587570615,
|
|
"learning_rate": 2.414752712413696e-06,
|
|
"loss": 0.8618,
|
|
"mean_token_accuracy": 0.7744259595870971,
|
|
"num_tokens": 778237815.0,
|
|
"step": 7780
|
|
},
|
|
{
|
|
"entropy": 0.8365234375,
|
|
"epoch": 1.064648397457801,
|
|
"grad_norm": 0.35717220889485013,
|
|
"learning_rate": 2.4112300972241797e-06,
|
|
"loss": 0.8331,
|
|
"mean_token_accuracy": 0.7843637436628341,
|
|
"num_tokens": 779227550.0,
|
|
"step": 7790
|
|
},
|
|
{
|
|
"entropy": 0.8810546875,
|
|
"epoch": 1.0660151711884098,
|
|
"grad_norm": 0.2722603777507139,
|
|
"learning_rate": 2.4077074820346628e-06,
|
|
"loss": 0.8786,
|
|
"mean_token_accuracy": 0.7733519166707993,
|
|
"num_tokens": 780238352.0,
|
|
"step": 7800
|
|
},
|
|
{
|
|
"entropy": 0.911328125,
|
|
"epoch": 1.0673819449190187,
|
|
"grad_norm": 0.308054238575354,
|
|
"learning_rate": 2.404184866845146e-06,
|
|
"loss": 0.9107,
|
|
"mean_token_accuracy": 0.7651260524988175,
|
|
"num_tokens": 781266598.0,
|
|
"step": 7810
|
|
},
|
|
{
|
|
"entropy": 0.892578125,
|
|
"epoch": 1.0687487186496276,
|
|
"grad_norm": 0.28119976069059144,
|
|
"learning_rate": 2.4006622516556295e-06,
|
|
"loss": 0.8851,
|
|
"mean_token_accuracy": 0.771707820892334,
|
|
"num_tokens": 782271940.0,
|
|
"step": 7820
|
|
},
|
|
{
|
|
"entropy": 0.9009765625,
|
|
"epoch": 1.0701154923802365,
|
|
"grad_norm": 0.30694646686012544,
|
|
"learning_rate": 2.3971396364661126e-06,
|
|
"loss": 0.8988,
|
|
"mean_token_accuracy": 0.7684226632118225,
|
|
"num_tokens": 783261830.0,
|
|
"step": 7830
|
|
},
|
|
{
|
|
"entropy": 0.9056640625,
|
|
"epoch": 1.0714822661108454,
|
|
"grad_norm": 0.2919828267431373,
|
|
"learning_rate": 2.393617021276596e-06,
|
|
"loss": 0.9028,
|
|
"mean_token_accuracy": 0.7687452644109726,
|
|
"num_tokens": 784254480.0,
|
|
"step": 7840
|
|
},
|
|
{
|
|
"entropy": 0.8873046875,
|
|
"epoch": 1.0728490398414543,
|
|
"grad_norm": 0.2703349693930874,
|
|
"learning_rate": 2.3900944060870793e-06,
|
|
"loss": 0.8909,
|
|
"mean_token_accuracy": 0.7710470885038376,
|
|
"num_tokens": 785218899.0,
|
|
"step": 7850
|
|
},
|
|
{
|
|
"entropy": 0.891015625,
|
|
"epoch": 1.0742158135720632,
|
|
"grad_norm": 0.3002278848077637,
|
|
"learning_rate": 2.3865717908975624e-06,
|
|
"loss": 0.8819,
|
|
"mean_token_accuracy": 0.7716854929924011,
|
|
"num_tokens": 786230816.0,
|
|
"step": 7860
|
|
},
|
|
{
|
|
"entropy": 0.8919921875,
|
|
"epoch": 1.075582587302672,
|
|
"grad_norm": 0.3041983737553101,
|
|
"learning_rate": 2.383049175708046e-06,
|
|
"loss": 0.8806,
|
|
"mean_token_accuracy": 0.7712246835231781,
|
|
"num_tokens": 787230555.0,
|
|
"step": 7870
|
|
},
|
|
{
|
|
"entropy": 0.8556640625,
|
|
"epoch": 1.076949361033281,
|
|
"grad_norm": 0.32375386825901603,
|
|
"learning_rate": 2.379526560518529e-06,
|
|
"loss": 0.854,
|
|
"mean_token_accuracy": 0.7774552226066589,
|
|
"num_tokens": 788216131.0,
|
|
"step": 7880
|
|
},
|
|
{
|
|
"entropy": 0.9169921875,
|
|
"epoch": 1.0783161347638899,
|
|
"grad_norm": 0.29044161553125775,
|
|
"learning_rate": 2.376003945329012e-06,
|
|
"loss": 0.9084,
|
|
"mean_token_accuracy": 0.7649928748607635,
|
|
"num_tokens": 789230407.0,
|
|
"step": 7890
|
|
},
|
|
{
|
|
"entropy": 0.8341796875,
|
|
"epoch": 1.0796829084944988,
|
|
"grad_norm": 0.34201854596654063,
|
|
"learning_rate": 2.3724813301394957e-06,
|
|
"loss": 0.8281,
|
|
"mean_token_accuracy": 0.7836710900068283,
|
|
"num_tokens": 790202084.0,
|
|
"step": 7900
|
|
},
|
|
{
|
|
"entropy": 0.8529296875,
|
|
"epoch": 1.0810496822251077,
|
|
"grad_norm": 0.2740726699799023,
|
|
"learning_rate": 2.368958714949979e-06,
|
|
"loss": 0.8415,
|
|
"mean_token_accuracy": 0.7816561102867127,
|
|
"num_tokens": 791159122.0,
|
|
"step": 7910
|
|
},
|
|
{
|
|
"entropy": 0.8775390625,
|
|
"epoch": 1.0824164559557166,
|
|
"grad_norm": 0.28152409944060025,
|
|
"learning_rate": 2.3654360997604624e-06,
|
|
"loss": 0.8579,
|
|
"mean_token_accuracy": 0.7772873550653457,
|
|
"num_tokens": 792135223.0,
|
|
"step": 7920
|
|
},
|
|
{
|
|
"entropy": 0.88984375,
|
|
"epoch": 1.0837832296863255,
|
|
"grad_norm": 0.283710835487499,
|
|
"learning_rate": 2.3619134845709455e-06,
|
|
"loss": 0.8919,
|
|
"mean_token_accuracy": 0.7725451946258545,
|
|
"num_tokens": 793161555.0,
|
|
"step": 7930
|
|
},
|
|
{
|
|
"entropy": 0.89140625,
|
|
"epoch": 1.0851500034169344,
|
|
"grad_norm": 0.265775937767258,
|
|
"learning_rate": 2.358390869381429e-06,
|
|
"loss": 0.8806,
|
|
"mean_token_accuracy": 0.7731119066476821,
|
|
"num_tokens": 794150499.0,
|
|
"step": 7940
|
|
},
|
|
{
|
|
"entropy": 0.9056640625,
|
|
"epoch": 1.0865167771475432,
|
|
"grad_norm": 0.28693837058699784,
|
|
"learning_rate": 2.354868254191912e-06,
|
|
"loss": 0.9113,
|
|
"mean_token_accuracy": 0.767306563258171,
|
|
"num_tokens": 795165922.0,
|
|
"step": 7950
|
|
},
|
|
{
|
|
"entropy": 0.88203125,
|
|
"epoch": 1.0878835508781521,
|
|
"grad_norm": 0.29947088578564524,
|
|
"learning_rate": 2.3513456390023957e-06,
|
|
"loss": 0.8786,
|
|
"mean_token_accuracy": 0.773172116279602,
|
|
"num_tokens": 796180992.0,
|
|
"step": 7960
|
|
},
|
|
{
|
|
"entropy": 0.9029296875,
|
|
"epoch": 1.089250324608761,
|
|
"grad_norm": 0.29850405112863504,
|
|
"learning_rate": 2.347823023812879e-06,
|
|
"loss": 0.9061,
|
|
"mean_token_accuracy": 0.7679647415876388,
|
|
"num_tokens": 797138287.0,
|
|
"step": 7970
|
|
},
|
|
{
|
|
"entropy": 0.91015625,
|
|
"epoch": 1.09061709833937,
|
|
"grad_norm": 0.30795035551531064,
|
|
"learning_rate": 2.3443004086233624e-06,
|
|
"loss": 0.9015,
|
|
"mean_token_accuracy": 0.7690918207168579,
|
|
"num_tokens": 798120086.0,
|
|
"step": 7980
|
|
},
|
|
{
|
|
"entropy": 0.868359375,
|
|
"epoch": 1.0919838720699788,
|
|
"grad_norm": 0.3179767132622585,
|
|
"learning_rate": 2.3407777934338455e-06,
|
|
"loss": 0.8711,
|
|
"mean_token_accuracy": 0.7754184484481812,
|
|
"num_tokens": 799136096.0,
|
|
"step": 7990
|
|
},
|
|
{
|
|
"entropy": 0.8705078125,
|
|
"epoch": 1.0933506458005877,
|
|
"grad_norm": 0.3202541586969867,
|
|
"learning_rate": 2.3372551782443286e-06,
|
|
"loss": 0.8615,
|
|
"mean_token_accuracy": 0.7766802579164505,
|
|
"num_tokens": 800127250.0,
|
|
"step": 8000
|
|
},
|
|
{
|
|
"entropy": 0.883984375,
|
|
"epoch": 1.0947174195311966,
|
|
"grad_norm": 0.2738459992787939,
|
|
"learning_rate": 2.333732563054812e-06,
|
|
"loss": 0.8832,
|
|
"mean_token_accuracy": 0.7726639717817306,
|
|
"num_tokens": 801158551.0,
|
|
"step": 8010
|
|
},
|
|
{
|
|
"entropy": 0.891015625,
|
|
"epoch": 1.0960841932618055,
|
|
"grad_norm": 0.3279895401369924,
|
|
"learning_rate": 2.3302099478652953e-06,
|
|
"loss": 0.8868,
|
|
"mean_token_accuracy": 0.7728369474411011,
|
|
"num_tokens": 802172508.0,
|
|
"step": 8020
|
|
},
|
|
{
|
|
"entropy": 0.88515625,
|
|
"epoch": 1.0974509669924144,
|
|
"grad_norm": 0.3187174587488924,
|
|
"learning_rate": 2.3266873326757784e-06,
|
|
"loss": 0.8904,
|
|
"mean_token_accuracy": 0.7710654973983765,
|
|
"num_tokens": 803132547.0,
|
|
"step": 8030
|
|
},
|
|
{
|
|
"entropy": 0.942578125,
|
|
"epoch": 1.0988177407230233,
|
|
"grad_norm": 0.30790766542540043,
|
|
"learning_rate": 2.323164717486262e-06,
|
|
"loss": 0.9432,
|
|
"mean_token_accuracy": 0.7573939681053161,
|
|
"num_tokens": 804164208.0,
|
|
"step": 8040
|
|
},
|
|
{
|
|
"entropy": 0.8318359375,
|
|
"epoch": 1.1001845144536322,
|
|
"grad_norm": 0.2917158392440747,
|
|
"learning_rate": 2.319642102296745e-06,
|
|
"loss": 0.8296,
|
|
"mean_token_accuracy": 0.7845388770103454,
|
|
"num_tokens": 805170699.0,
|
|
"step": 8050
|
|
},
|
|
{
|
|
"entropy": 0.8783203125,
|
|
"epoch": 1.101551288184241,
|
|
"grad_norm": 0.2764726080246425,
|
|
"learning_rate": 2.3161194871072287e-06,
|
|
"loss": 0.8707,
|
|
"mean_token_accuracy": 0.7736043572425843,
|
|
"num_tokens": 806152913.0,
|
|
"step": 8060
|
|
},
|
|
{
|
|
"entropy": 0.8615234375,
|
|
"epoch": 1.10291806191485,
|
|
"grad_norm": 0.2713226995073508,
|
|
"learning_rate": 2.3125968719177118e-06,
|
|
"loss": 0.8547,
|
|
"mean_token_accuracy": 0.7778900384902954,
|
|
"num_tokens": 807131660.0,
|
|
"step": 8070
|
|
},
|
|
{
|
|
"entropy": 0.894921875,
|
|
"epoch": 1.1042848356454589,
|
|
"grad_norm": 0.3155371993665204,
|
|
"learning_rate": 2.3090742567281953e-06,
|
|
"loss": 0.8856,
|
|
"mean_token_accuracy": 0.7725750893354416,
|
|
"num_tokens": 808172360.0,
|
|
"step": 8080
|
|
},
|
|
{
|
|
"entropy": 0.87421875,
|
|
"epoch": 1.1056516093760678,
|
|
"grad_norm": 0.33602343572224463,
|
|
"learning_rate": 2.3055516415386785e-06,
|
|
"loss": 0.8706,
|
|
"mean_token_accuracy": 0.7763155251741409,
|
|
"num_tokens": 809156426.0,
|
|
"step": 8090
|
|
},
|
|
{
|
|
"entropy": 0.88359375,
|
|
"epoch": 1.1070183831066767,
|
|
"grad_norm": 0.35923672187360944,
|
|
"learning_rate": 2.302029026349162e-06,
|
|
"loss": 0.8856,
|
|
"mean_token_accuracy": 0.7718951761722564,
|
|
"num_tokens": 810136869.0,
|
|
"step": 8100
|
|
},
|
|
{
|
|
"entropy": 0.896875,
|
|
"epoch": 1.1083851568372856,
|
|
"grad_norm": 0.33780389993440346,
|
|
"learning_rate": 2.298506411159645e-06,
|
|
"loss": 0.8942,
|
|
"mean_token_accuracy": 0.7693793773651123,
|
|
"num_tokens": 811163591.0,
|
|
"step": 8110
|
|
},
|
|
{
|
|
"entropy": 0.8822265625,
|
|
"epoch": 1.1097519305678945,
|
|
"grad_norm": 0.26669180664964776,
|
|
"learning_rate": 2.2949837959701287e-06,
|
|
"loss": 0.8754,
|
|
"mean_token_accuracy": 0.7732379794120788,
|
|
"num_tokens": 812236907.0,
|
|
"step": 8120
|
|
},
|
|
{
|
|
"entropy": 0.9029296875,
|
|
"epoch": 1.1111187042985033,
|
|
"grad_norm": 0.293866925295392,
|
|
"learning_rate": 2.291461180780612e-06,
|
|
"loss": 0.8953,
|
|
"mean_token_accuracy": 0.7700059890747071,
|
|
"num_tokens": 813270666.0,
|
|
"step": 8130
|
|
},
|
|
{
|
|
"entropy": 0.880078125,
|
|
"epoch": 1.1124854780291122,
|
|
"grad_norm": 0.2848651915525826,
|
|
"learning_rate": 2.287938565591095e-06,
|
|
"loss": 0.877,
|
|
"mean_token_accuracy": 0.774398684501648,
|
|
"num_tokens": 814273100.0,
|
|
"step": 8140
|
|
},
|
|
{
|
|
"entropy": 0.84453125,
|
|
"epoch": 1.1138522517597211,
|
|
"grad_norm": 0.2784138414862953,
|
|
"learning_rate": 2.2844159504015785e-06,
|
|
"loss": 0.8408,
|
|
"mean_token_accuracy": 0.7797045677900314,
|
|
"num_tokens": 815298510.0,
|
|
"step": 8150
|
|
},
|
|
{
|
|
"entropy": 0.8865234375,
|
|
"epoch": 1.11521902549033,
|
|
"grad_norm": 0.3243779637964536,
|
|
"learning_rate": 2.2808933352120616e-06,
|
|
"loss": 0.8736,
|
|
"mean_token_accuracy": 0.7740034967660904,
|
|
"num_tokens": 816246236.0,
|
|
"step": 8160
|
|
},
|
|
{
|
|
"entropy": 0.88125,
|
|
"epoch": 1.116585799220939,
|
|
"grad_norm": 0.2889153397414198,
|
|
"learning_rate": 2.277370720022545e-06,
|
|
"loss": 0.8757,
|
|
"mean_token_accuracy": 0.7712274640798569,
|
|
"num_tokens": 817240153.0,
|
|
"step": 8170
|
|
},
|
|
{
|
|
"entropy": 0.8833984375,
|
|
"epoch": 1.1179525729515478,
|
|
"grad_norm": 0.2992735689253027,
|
|
"learning_rate": 2.2738481048330283e-06,
|
|
"loss": 0.8804,
|
|
"mean_token_accuracy": 0.7709307193756103,
|
|
"num_tokens": 818254280.0,
|
|
"step": 8180
|
|
},
|
|
{
|
|
"entropy": 0.86875,
|
|
"epoch": 1.1193193466821567,
|
|
"grad_norm": 0.28925240418341724,
|
|
"learning_rate": 2.2703254896435114e-06,
|
|
"loss": 0.8697,
|
|
"mean_token_accuracy": 0.7739754796028138,
|
|
"num_tokens": 819294564.0,
|
|
"step": 8190
|
|
},
|
|
{
|
|
"entropy": 0.905078125,
|
|
"epoch": 1.1206861204127656,
|
|
"grad_norm": 0.2681283206413371,
|
|
"learning_rate": 2.266802874453995e-06,
|
|
"loss": 0.9121,
|
|
"mean_token_accuracy": 0.7649559646844863,
|
|
"num_tokens": 820306648.0,
|
|
"step": 8200
|
|
},
|
|
{
|
|
"entropy": 0.8896484375,
|
|
"epoch": 1.1220528941433745,
|
|
"grad_norm": 0.3393732317970729,
|
|
"learning_rate": 2.263280259264478e-06,
|
|
"loss": 0.88,
|
|
"mean_token_accuracy": 0.7726121306419372,
|
|
"num_tokens": 821282694.0,
|
|
"step": 8210
|
|
},
|
|
{
|
|
"entropy": 0.8490234375,
|
|
"epoch": 1.1234196678739834,
|
|
"grad_norm": 0.2750429056684657,
|
|
"learning_rate": 2.259757644074961e-06,
|
|
"loss": 0.834,
|
|
"mean_token_accuracy": 0.7836462050676346,
|
|
"num_tokens": 822266150.0,
|
|
"step": 8220
|
|
},
|
|
{
|
|
"entropy": 0.8912109375,
|
|
"epoch": 1.1247864416045923,
|
|
"grad_norm": 0.2923855996695632,
|
|
"learning_rate": 2.2562350288854447e-06,
|
|
"loss": 0.8788,
|
|
"mean_token_accuracy": 0.7722830355167389,
|
|
"num_tokens": 823268751.0,
|
|
"step": 8230
|
|
},
|
|
{
|
|
"entropy": 0.842578125,
|
|
"epoch": 1.1261532153352012,
|
|
"grad_norm": 0.29505650251845655,
|
|
"learning_rate": 2.252712413695928e-06,
|
|
"loss": 0.8344,
|
|
"mean_token_accuracy": 0.7821342349052429,
|
|
"num_tokens": 824232508.0,
|
|
"step": 8240
|
|
},
|
|
{
|
|
"entropy": 0.8916015625,
|
|
"epoch": 1.12751998906581,
|
|
"grad_norm": 0.2808646995810378,
|
|
"learning_rate": 2.2491897985064114e-06,
|
|
"loss": 0.8902,
|
|
"mean_token_accuracy": 0.7694691866636276,
|
|
"num_tokens": 825239659.0,
|
|
"step": 8250
|
|
},
|
|
{
|
|
"entropy": 0.8658203125,
|
|
"epoch": 1.128886762796419,
|
|
"grad_norm": 0.30900985939281217,
|
|
"learning_rate": 2.2456671833168945e-06,
|
|
"loss": 0.8657,
|
|
"mean_token_accuracy": 0.7754149675369263,
|
|
"num_tokens": 826204127.0,
|
|
"step": 8260
|
|
},
|
|
{
|
|
"entropy": 0.8826171875,
|
|
"epoch": 1.1302535365270279,
|
|
"grad_norm": 0.2909265114564738,
|
|
"learning_rate": 2.242144568127378e-06,
|
|
"loss": 0.8763,
|
|
"mean_token_accuracy": 0.7735009461641311,
|
|
"num_tokens": 827176933.0,
|
|
"step": 8270
|
|
},
|
|
{
|
|
"entropy": 0.9193359375,
|
|
"epoch": 1.1316203102576368,
|
|
"grad_norm": 0.26352549300593264,
|
|
"learning_rate": 2.238621952937861e-06,
|
|
"loss": 0.9224,
|
|
"mean_token_accuracy": 0.7645892828702927,
|
|
"num_tokens": 828203921.0,
|
|
"step": 8280
|
|
},
|
|
{
|
|
"entropy": 0.912109375,
|
|
"epoch": 1.1329870839882457,
|
|
"grad_norm": 0.3292369443718077,
|
|
"learning_rate": 2.2350993377483447e-06,
|
|
"loss": 0.9048,
|
|
"mean_token_accuracy": 0.7686350703239441,
|
|
"num_tokens": 829191366.0,
|
|
"step": 8290
|
|
},
|
|
{
|
|
"entropy": 0.871875,
|
|
"epoch": 1.1343538577188546,
|
|
"grad_norm": 0.27546705207032435,
|
|
"learning_rate": 2.231576722558828e-06,
|
|
"loss": 0.8687,
|
|
"mean_token_accuracy": 0.774681881070137,
|
|
"num_tokens": 830174005.0,
|
|
"step": 8300
|
|
},
|
|
{
|
|
"entropy": 0.8591796875,
|
|
"epoch": 1.1357206314494634,
|
|
"grad_norm": 0.31013154041371493,
|
|
"learning_rate": 2.2280541073693114e-06,
|
|
"loss": 0.8533,
|
|
"mean_token_accuracy": 0.778465959429741,
|
|
"num_tokens": 831187447.0,
|
|
"step": 8310
|
|
},
|
|
{
|
|
"entropy": 0.9166015625,
|
|
"epoch": 1.1370874051800723,
|
|
"grad_norm": 0.30999698673182763,
|
|
"learning_rate": 2.2245314921797945e-06,
|
|
"loss": 0.9153,
|
|
"mean_token_accuracy": 0.7663945019245147,
|
|
"num_tokens": 832195212.0,
|
|
"step": 8320
|
|
},
|
|
{
|
|
"entropy": 0.89375,
|
|
"epoch": 1.1384541789106812,
|
|
"grad_norm": 0.2840177530679021,
|
|
"learning_rate": 2.2210088769902777e-06,
|
|
"loss": 0.8909,
|
|
"mean_token_accuracy": 0.7699512153863907,
|
|
"num_tokens": 833214334.0,
|
|
"step": 8330
|
|
},
|
|
{
|
|
"entropy": 0.812109375,
|
|
"epoch": 1.1398209526412901,
|
|
"grad_norm": 0.3061361455364483,
|
|
"learning_rate": 2.217486261800761e-06,
|
|
"loss": 0.8133,
|
|
"mean_token_accuracy": 0.7844528049230576,
|
|
"num_tokens": 834152536.0,
|
|
"step": 8340
|
|
},
|
|
{
|
|
"entropy": 0.88203125,
|
|
"epoch": 1.141187726371899,
|
|
"grad_norm": 0.31562688449759896,
|
|
"learning_rate": 2.2139636466112443e-06,
|
|
"loss": 0.8777,
|
|
"mean_token_accuracy": 0.7745613992214203,
|
|
"num_tokens": 835183047.0,
|
|
"step": 8350
|
|
},
|
|
{
|
|
"entropy": 0.85703125,
|
|
"epoch": 1.142554500102508,
|
|
"grad_norm": 0.27042326295923175,
|
|
"learning_rate": 2.2104410314217275e-06,
|
|
"loss": 0.8545,
|
|
"mean_token_accuracy": 0.7787758588790894,
|
|
"num_tokens": 836174945.0,
|
|
"step": 8360
|
|
},
|
|
{
|
|
"entropy": 0.899609375,
|
|
"epoch": 1.1439212738331168,
|
|
"grad_norm": 0.3209482276357742,
|
|
"learning_rate": 2.206918416232211e-06,
|
|
"loss": 0.8971,
|
|
"mean_token_accuracy": 0.7691022336483002,
|
|
"num_tokens": 837204894.0,
|
|
"step": 8370
|
|
},
|
|
{
|
|
"entropy": 0.9091796875,
|
|
"epoch": 1.1452880475637257,
|
|
"grad_norm": 0.2720624367783337,
|
|
"learning_rate": 2.203395801042694e-06,
|
|
"loss": 0.9018,
|
|
"mean_token_accuracy": 0.7668502718210221,
|
|
"num_tokens": 838221943.0,
|
|
"step": 8380
|
|
},
|
|
{
|
|
"entropy": 0.88046875,
|
|
"epoch": 1.1466548212943346,
|
|
"grad_norm": 0.36967418289429205,
|
|
"learning_rate": 2.1998731858531777e-06,
|
|
"loss": 0.8826,
|
|
"mean_token_accuracy": 0.7727677702903748,
|
|
"num_tokens": 839270110.0,
|
|
"step": 8390
|
|
},
|
|
{
|
|
"entropy": 0.894140625,
|
|
"epoch": 1.1480215950249435,
|
|
"grad_norm": 0.30111225201428243,
|
|
"learning_rate": 2.196350570663661e-06,
|
|
"loss": 0.8947,
|
|
"mean_token_accuracy": 0.771626016497612,
|
|
"num_tokens": 840277305.0,
|
|
"step": 8400
|
|
},
|
|
{
|
|
"entropy": 0.87734375,
|
|
"epoch": 1.1493883687555526,
|
|
"grad_norm": 0.26893776715498025,
|
|
"learning_rate": 2.192827955474144e-06,
|
|
"loss": 0.8638,
|
|
"mean_token_accuracy": 0.7754395067691803,
|
|
"num_tokens": 841258279.0,
|
|
"step": 8410
|
|
},
|
|
{
|
|
"entropy": 0.85,
|
|
"epoch": 1.1507551424861615,
|
|
"grad_norm": 0.26418881468616484,
|
|
"learning_rate": 2.1893053402846275e-06,
|
|
"loss": 0.8458,
|
|
"mean_token_accuracy": 0.7803683370351792,
|
|
"num_tokens": 842266370.0,
|
|
"step": 8420
|
|
},
|
|
{
|
|
"entropy": 0.9205078125,
|
|
"epoch": 1.1521219162167704,
|
|
"grad_norm": 0.27998466744081835,
|
|
"learning_rate": 2.1857827250951106e-06,
|
|
"loss": 0.9131,
|
|
"mean_token_accuracy": 0.769468155503273,
|
|
"num_tokens": 843290392.0,
|
|
"step": 8430
|
|
},
|
|
{
|
|
"entropy": 0.8912109375,
|
|
"epoch": 1.1534886899473793,
|
|
"grad_norm": 0.28369615049860086,
|
|
"learning_rate": 2.182260109905594e-06,
|
|
"loss": 0.8906,
|
|
"mean_token_accuracy": 0.7686574339866639,
|
|
"num_tokens": 844293297.0,
|
|
"step": 8440
|
|
},
|
|
{
|
|
"entropy": 0.848046875,
|
|
"epoch": 1.1548554636779882,
|
|
"grad_norm": 0.29001444549323685,
|
|
"learning_rate": 2.1787374947160777e-06,
|
|
"loss": 0.8432,
|
|
"mean_token_accuracy": 0.7801264584064483,
|
|
"num_tokens": 845265572.0,
|
|
"step": 8450
|
|
},
|
|
{
|
|
"entropy": 0.85546875,
|
|
"epoch": 1.156222237408597,
|
|
"grad_norm": 0.26316554004026294,
|
|
"learning_rate": 2.175214879526561e-06,
|
|
"loss": 0.8529,
|
|
"mean_token_accuracy": 0.7776524603366852,
|
|
"num_tokens": 846309389.0,
|
|
"step": 8460
|
|
},
|
|
{
|
|
"entropy": 0.9064453125,
|
|
"epoch": 1.157589011139206,
|
|
"grad_norm": 0.26031983215716764,
|
|
"learning_rate": 2.171692264337044e-06,
|
|
"loss": 0.8902,
|
|
"mean_token_accuracy": 0.7699259132146835,
|
|
"num_tokens": 847290772.0,
|
|
"step": 8470
|
|
},
|
|
{
|
|
"entropy": 0.8736328125,
|
|
"epoch": 1.1589557848698149,
|
|
"grad_norm": 0.3026672610223644,
|
|
"learning_rate": 2.1681696491475275e-06,
|
|
"loss": 0.8612,
|
|
"mean_token_accuracy": 0.7761404126882553,
|
|
"num_tokens": 848259241.0,
|
|
"step": 8480
|
|
},
|
|
{
|
|
"entropy": 0.873046875,
|
|
"epoch": 1.1603225586004238,
|
|
"grad_norm": 0.30300453431281643,
|
|
"learning_rate": 2.1646470339580106e-06,
|
|
"loss": 0.8652,
|
|
"mean_token_accuracy": 0.7750141888856887,
|
|
"num_tokens": 849323114.0,
|
|
"step": 8490
|
|
},
|
|
{
|
|
"entropy": 0.8515625,
|
|
"epoch": 1.1616893323310327,
|
|
"grad_norm": 0.29896307947454887,
|
|
"learning_rate": 2.1611244187684937e-06,
|
|
"loss": 0.8433,
|
|
"mean_token_accuracy": 0.781935402750969,
|
|
"num_tokens": 850309798.0,
|
|
"step": 8500
|
|
},
|
|
{
|
|
"entropy": 0.86640625,
|
|
"epoch": 1.1630561060616416,
|
|
"grad_norm": 0.26853818794300127,
|
|
"learning_rate": 2.1576018035789773e-06,
|
|
"loss": 0.8626,
|
|
"mean_token_accuracy": 0.7772506386041641,
|
|
"num_tokens": 851352098.0,
|
|
"step": 8510
|
|
},
|
|
{
|
|
"entropy": 0.8494140625,
|
|
"epoch": 1.1644228797922505,
|
|
"grad_norm": 0.2953547767032317,
|
|
"learning_rate": 2.1540791883894604e-06,
|
|
"loss": 0.8468,
|
|
"mean_token_accuracy": 0.7776229351758956,
|
|
"num_tokens": 852295903.0,
|
|
"step": 8520
|
|
},
|
|
{
|
|
"entropy": 0.8919921875,
|
|
"epoch": 1.1657896535228593,
|
|
"grad_norm": 0.28057589805100064,
|
|
"learning_rate": 2.150556573199944e-06,
|
|
"loss": 0.898,
|
|
"mean_token_accuracy": 0.7717033296823501,
|
|
"num_tokens": 853277283.0,
|
|
"step": 8530
|
|
},
|
|
{
|
|
"entropy": 0.8779296875,
|
|
"epoch": 1.1671564272534682,
|
|
"grad_norm": 0.2530775586288463,
|
|
"learning_rate": 2.147033958010427e-06,
|
|
"loss": 0.8714,
|
|
"mean_token_accuracy": 0.774565526843071,
|
|
"num_tokens": 854271824.0,
|
|
"step": 8540
|
|
},
|
|
{
|
|
"entropy": 0.903125,
|
|
"epoch": 1.1685232009840771,
|
|
"grad_norm": 0.34544681139586714,
|
|
"learning_rate": 2.14351134282091e-06,
|
|
"loss": 0.8906,
|
|
"mean_token_accuracy": 0.770963904261589,
|
|
"num_tokens": 855285118.0,
|
|
"step": 8550
|
|
},
|
|
{
|
|
"entropy": 0.8515625,
|
|
"epoch": 1.169889974714686,
|
|
"grad_norm": 0.28264304124943934,
|
|
"learning_rate": 2.1399887276313937e-06,
|
|
"loss": 0.8318,
|
|
"mean_token_accuracy": 0.782940024137497,
|
|
"num_tokens": 856315835.0,
|
|
"step": 8560
|
|
},
|
|
{
|
|
"entropy": 0.91015625,
|
|
"epoch": 1.171256748445295,
|
|
"grad_norm": 0.2748247961925148,
|
|
"learning_rate": 2.136466112441877e-06,
|
|
"loss": 0.9092,
|
|
"mean_token_accuracy": 0.7672925025224686,
|
|
"num_tokens": 857328374.0,
|
|
"step": 8570
|
|
},
|
|
{
|
|
"entropy": 0.868359375,
|
|
"epoch": 1.1726235221759038,
|
|
"grad_norm": 0.2954906777157409,
|
|
"learning_rate": 2.1329434972523604e-06,
|
|
"loss": 0.8626,
|
|
"mean_token_accuracy": 0.7766576707363129,
|
|
"num_tokens": 858312562.0,
|
|
"step": 8580
|
|
},
|
|
{
|
|
"entropy": 0.8763671875,
|
|
"epoch": 1.1739902959065127,
|
|
"grad_norm": 0.30194491461811546,
|
|
"learning_rate": 2.1294208820628435e-06,
|
|
"loss": 0.866,
|
|
"mean_token_accuracy": 0.7756797432899475,
|
|
"num_tokens": 859285077.0,
|
|
"step": 8590
|
|
},
|
|
{
|
|
"entropy": 0.9158203125,
|
|
"epoch": 1.1753570696371216,
|
|
"grad_norm": 0.28695395246749955,
|
|
"learning_rate": 2.125898266873327e-06,
|
|
"loss": 0.9094,
|
|
"mean_token_accuracy": 0.763212776184082,
|
|
"num_tokens": 860348335.0,
|
|
"step": 8600
|
|
},
|
|
{
|
|
"entropy": 0.905859375,
|
|
"epoch": 1.1767238433677305,
|
|
"grad_norm": 0.2780168930509453,
|
|
"learning_rate": 2.12237565168381e-06,
|
|
"loss": 0.8923,
|
|
"mean_token_accuracy": 0.7695642530918121,
|
|
"num_tokens": 861353583.0,
|
|
"step": 8610
|
|
},
|
|
{
|
|
"entropy": 0.8904296875,
|
|
"epoch": 1.1780906170983394,
|
|
"grad_norm": 0.2808632982421028,
|
|
"learning_rate": 2.1188530364942938e-06,
|
|
"loss": 0.8864,
|
|
"mean_token_accuracy": 0.7716062903404236,
|
|
"num_tokens": 862358162.0,
|
|
"step": 8620
|
|
},
|
|
{
|
|
"entropy": 0.83515625,
|
|
"epoch": 1.1794573908289483,
|
|
"grad_norm": 0.2568351032980569,
|
|
"learning_rate": 2.115330421304777e-06,
|
|
"loss": 0.8332,
|
|
"mean_token_accuracy": 0.7814292162656784,
|
|
"num_tokens": 863351324.0,
|
|
"step": 8630
|
|
},
|
|
{
|
|
"entropy": 0.8546875,
|
|
"epoch": 1.1808241645595572,
|
|
"grad_norm": 0.2781252136318616,
|
|
"learning_rate": 2.1118078061152604e-06,
|
|
"loss": 0.8653,
|
|
"mean_token_accuracy": 0.7765821278095245,
|
|
"num_tokens": 864410088.0,
|
|
"step": 8640
|
|
},
|
|
{
|
|
"entropy": 0.8513671875,
|
|
"epoch": 1.182190938290166,
|
|
"grad_norm": 0.274210870225196,
|
|
"learning_rate": 2.1082851909257435e-06,
|
|
"loss": 0.8543,
|
|
"mean_token_accuracy": 0.7777923494577408,
|
|
"num_tokens": 865418307.0,
|
|
"step": 8650
|
|
},
|
|
{
|
|
"entropy": 0.875390625,
|
|
"epoch": 1.183557712020775,
|
|
"grad_norm": 0.2544995245560745,
|
|
"learning_rate": 2.1047625757362267e-06,
|
|
"loss": 0.8722,
|
|
"mean_token_accuracy": 0.775250768661499,
|
|
"num_tokens": 866455039.0,
|
|
"step": 8660
|
|
},
|
|
{
|
|
"entropy": 0.8765625,
|
|
"epoch": 1.1849244857513839,
|
|
"grad_norm": 0.304055083277078,
|
|
"learning_rate": 2.1012399605467102e-06,
|
|
"loss": 0.8825,
|
|
"mean_token_accuracy": 0.771222734451294,
|
|
"num_tokens": 867417207.0,
|
|
"step": 8670
|
|
},
|
|
{
|
|
"entropy": 0.9052734375,
|
|
"epoch": 1.1862912594819928,
|
|
"grad_norm": 0.2696270499669348,
|
|
"learning_rate": 2.0977173453571933e-06,
|
|
"loss": 0.9074,
|
|
"mean_token_accuracy": 0.767414927482605,
|
|
"num_tokens": 868377084.0,
|
|
"step": 8680
|
|
},
|
|
{
|
|
"entropy": 0.915625,
|
|
"epoch": 1.1876580332126017,
|
|
"grad_norm": 0.4237088248867688,
|
|
"learning_rate": 2.0941947301676765e-06,
|
|
"loss": 0.9171,
|
|
"mean_token_accuracy": 0.7651487052440643,
|
|
"num_tokens": 869321948.0,
|
|
"step": 8690
|
|
},
|
|
{
|
|
"entropy": 0.8552734375,
|
|
"epoch": 1.1890248069432106,
|
|
"grad_norm": 0.27133224594650235,
|
|
"learning_rate": 2.09067211497816e-06,
|
|
"loss": 0.847,
|
|
"mean_token_accuracy": 0.777655553817749,
|
|
"num_tokens": 870346923.0,
|
|
"step": 8700
|
|
},
|
|
{
|
|
"entropy": 0.8810546875,
|
|
"epoch": 1.1903915806738194,
|
|
"grad_norm": 0.2942182799809781,
|
|
"learning_rate": 2.087149499788643e-06,
|
|
"loss": 0.8778,
|
|
"mean_token_accuracy": 0.771362042427063,
|
|
"num_tokens": 871336095.0,
|
|
"step": 8710
|
|
},
|
|
{
|
|
"entropy": 0.880078125,
|
|
"epoch": 1.1917583544044283,
|
|
"grad_norm": 0.3855522425835558,
|
|
"learning_rate": 2.0836268845991263e-06,
|
|
"loss": 0.8768,
|
|
"mean_token_accuracy": 0.772871059179306,
|
|
"num_tokens": 872342415.0,
|
|
"step": 8720
|
|
},
|
|
{
|
|
"entropy": 0.8859375,
|
|
"epoch": 1.1931251281350372,
|
|
"grad_norm": 0.32513479036702647,
|
|
"learning_rate": 2.08010426940961e-06,
|
|
"loss": 0.8905,
|
|
"mean_token_accuracy": 0.7711873173713684,
|
|
"num_tokens": 873374465.0,
|
|
"step": 8730
|
|
},
|
|
{
|
|
"entropy": 0.905078125,
|
|
"epoch": 1.1944919018656461,
|
|
"grad_norm": 0.3152621683669878,
|
|
"learning_rate": 2.076581654220093e-06,
|
|
"loss": 0.9123,
|
|
"mean_token_accuracy": 0.7674855053424835,
|
|
"num_tokens": 874374048.0,
|
|
"step": 8740
|
|
},
|
|
{
|
|
"entropy": 0.8541015625,
|
|
"epoch": 1.195858675596255,
|
|
"grad_norm": 0.32203878810733166,
|
|
"learning_rate": 2.0730590390305765e-06,
|
|
"loss": 0.8483,
|
|
"mean_token_accuracy": 0.7792962521314621,
|
|
"num_tokens": 875363342.0,
|
|
"step": 8750
|
|
},
|
|
{
|
|
"entropy": 0.886328125,
|
|
"epoch": 1.197225449326864,
|
|
"grad_norm": 0.3106942149047768,
|
|
"learning_rate": 2.0695364238410596e-06,
|
|
"loss": 0.8637,
|
|
"mean_token_accuracy": 0.7743999153375626,
|
|
"num_tokens": 876424286.0,
|
|
"step": 8760
|
|
},
|
|
{
|
|
"entropy": 0.87265625,
|
|
"epoch": 1.1985922230574728,
|
|
"grad_norm": 0.25260367287281077,
|
|
"learning_rate": 2.066013808651543e-06,
|
|
"loss": 0.8722,
|
|
"mean_token_accuracy": 0.7735497713088989,
|
|
"num_tokens": 877423113.0,
|
|
"step": 8770
|
|
},
|
|
{
|
|
"entropy": 0.9107421875,
|
|
"epoch": 1.1999589967880817,
|
|
"grad_norm": 0.3305651888393956,
|
|
"learning_rate": 2.0624911934620263e-06,
|
|
"loss": 0.9088,
|
|
"mean_token_accuracy": 0.7677461117506027,
|
|
"num_tokens": 878373113.0,
|
|
"step": 8780
|
|
},
|
|
{
|
|
"entropy": 0.898046875,
|
|
"epoch": 1.2013257705186906,
|
|
"grad_norm": 0.2808961192217918,
|
|
"learning_rate": 2.05896857827251e-06,
|
|
"loss": 0.8967,
|
|
"mean_token_accuracy": 0.7692672610282898,
|
|
"num_tokens": 879428569.0,
|
|
"step": 8790
|
|
},
|
|
{
|
|
"entropy": 0.8427734375,
|
|
"epoch": 1.2026925442492995,
|
|
"grad_norm": 0.29000049617018775,
|
|
"learning_rate": 2.055445963082993e-06,
|
|
"loss": 0.8333,
|
|
"mean_token_accuracy": 0.7826966732740402,
|
|
"num_tokens": 880424358.0,
|
|
"step": 8800
|
|
},
|
|
{
|
|
"entropy": 0.855859375,
|
|
"epoch": 1.2040593179799084,
|
|
"grad_norm": 0.2695101520356307,
|
|
"learning_rate": 2.0519233478934765e-06,
|
|
"loss": 0.8595,
|
|
"mean_token_accuracy": 0.7768262982368469,
|
|
"num_tokens": 881459047.0,
|
|
"step": 8810
|
|
},
|
|
{
|
|
"entropy": 0.887109375,
|
|
"epoch": 1.2054260917105173,
|
|
"grad_norm": 0.2946064596261865,
|
|
"learning_rate": 2.0484007327039596e-06,
|
|
"loss": 0.8878,
|
|
"mean_token_accuracy": 0.7716862231492996,
|
|
"num_tokens": 882445406.0,
|
|
"step": 8820
|
|
},
|
|
{
|
|
"entropy": 0.8833984375,
|
|
"epoch": 1.2067928654411262,
|
|
"grad_norm": 0.3268126449202719,
|
|
"learning_rate": 2.0448781175144427e-06,
|
|
"loss": 0.8715,
|
|
"mean_token_accuracy": 0.7741323858499527,
|
|
"num_tokens": 883435853.0,
|
|
"step": 8830
|
|
},
|
|
{
|
|
"entropy": 0.8697265625,
|
|
"epoch": 1.208159639171735,
|
|
"grad_norm": 0.3210336964377021,
|
|
"learning_rate": 2.0413555023249263e-06,
|
|
"loss": 0.8643,
|
|
"mean_token_accuracy": 0.776130297780037,
|
|
"num_tokens": 884446319.0,
|
|
"step": 8840
|
|
},
|
|
{
|
|
"entropy": 0.8775390625,
|
|
"epoch": 1.209526412902344,
|
|
"grad_norm": 0.2962784170102282,
|
|
"learning_rate": 2.0378328871354094e-06,
|
|
"loss": 0.8756,
|
|
"mean_token_accuracy": 0.7733410179615021,
|
|
"num_tokens": 885488707.0,
|
|
"step": 8850
|
|
},
|
|
{
|
|
"entropy": 0.8658203125,
|
|
"epoch": 1.2108931866329529,
|
|
"grad_norm": 0.30913240381507245,
|
|
"learning_rate": 2.034310271945893e-06,
|
|
"loss": 0.8629,
|
|
"mean_token_accuracy": 0.7776019871234894,
|
|
"num_tokens": 886537296.0,
|
|
"step": 8860
|
|
},
|
|
{
|
|
"entropy": 0.87890625,
|
|
"epoch": 1.2122599603635618,
|
|
"grad_norm": 0.28487486580420135,
|
|
"learning_rate": 2.030787656756376e-06,
|
|
"loss": 0.8726,
|
|
"mean_token_accuracy": 0.7727652668952942,
|
|
"num_tokens": 887518408.0,
|
|
"step": 8870
|
|
},
|
|
{
|
|
"entropy": 0.8578125,
|
|
"epoch": 1.2136267340941707,
|
|
"grad_norm": 0.27543412025935426,
|
|
"learning_rate": 2.027265041566859e-06,
|
|
"loss": 0.8685,
|
|
"mean_token_accuracy": 0.7757952272891998,
|
|
"num_tokens": 888512569.0,
|
|
"step": 8880
|
|
},
|
|
{
|
|
"entropy": 0.868359375,
|
|
"epoch": 1.2149935078247796,
|
|
"grad_norm": 0.29319106983320553,
|
|
"learning_rate": 2.0237424263773427e-06,
|
|
"loss": 0.8589,
|
|
"mean_token_accuracy": 0.7756124794483185,
|
|
"num_tokens": 889478735.0,
|
|
"step": 8890
|
|
},
|
|
{
|
|
"entropy": 0.84453125,
|
|
"epoch": 1.2163602815553884,
|
|
"grad_norm": 0.24953290735170106,
|
|
"learning_rate": 2.020219811187826e-06,
|
|
"loss": 0.8409,
|
|
"mean_token_accuracy": 0.7818514466285705,
|
|
"num_tokens": 890498254.0,
|
|
"step": 8900
|
|
},
|
|
{
|
|
"entropy": 0.89296875,
|
|
"epoch": 1.2177270552859973,
|
|
"grad_norm": 0.2813475300349863,
|
|
"learning_rate": 2.0166971959983094e-06,
|
|
"loss": 0.8981,
|
|
"mean_token_accuracy": 0.7680205345153809,
|
|
"num_tokens": 891510123.0,
|
|
"step": 8910
|
|
},
|
|
{
|
|
"entropy": 0.8458984375,
|
|
"epoch": 1.2190938290166062,
|
|
"grad_norm": 0.2774025401790624,
|
|
"learning_rate": 2.0131745808087925e-06,
|
|
"loss": 0.8489,
|
|
"mean_token_accuracy": 0.7807881861925126,
|
|
"num_tokens": 892518225.0,
|
|
"step": 8920
|
|
},
|
|
{
|
|
"entropy": 0.854296875,
|
|
"epoch": 1.2204606027472151,
|
|
"grad_norm": 0.2953623137060924,
|
|
"learning_rate": 2.009651965619276e-06,
|
|
"loss": 0.8565,
|
|
"mean_token_accuracy": 0.7782915621995926,
|
|
"num_tokens": 893518989.0,
|
|
"step": 8930
|
|
},
|
|
{
|
|
"entropy": 0.8916015625,
|
|
"epoch": 1.221827376477824,
|
|
"grad_norm": 0.29284169912434205,
|
|
"learning_rate": 2.0061293504297592e-06,
|
|
"loss": 0.8989,
|
|
"mean_token_accuracy": 0.7692197233438491,
|
|
"num_tokens": 894547776.0,
|
|
"step": 8940
|
|
},
|
|
{
|
|
"entropy": 0.8287109375,
|
|
"epoch": 1.223194150208433,
|
|
"grad_norm": 0.26331580937964716,
|
|
"learning_rate": 2.0026067352402428e-06,
|
|
"loss": 0.8212,
|
|
"mean_token_accuracy": 0.7847041428089142,
|
|
"num_tokens": 895565993.0,
|
|
"step": 8950
|
|
},
|
|
{
|
|
"entropy": 0.887890625,
|
|
"epoch": 1.2245609239390418,
|
|
"grad_norm": 0.2703605135285083,
|
|
"learning_rate": 1.999084120050726e-06,
|
|
"loss": 0.8896,
|
|
"mean_token_accuracy": 0.7713199764490127,
|
|
"num_tokens": 896569160.0,
|
|
"step": 8960
|
|
},
|
|
{
|
|
"entropy": 0.8525390625,
|
|
"epoch": 1.2259276976696507,
|
|
"grad_norm": 0.30967106825552837,
|
|
"learning_rate": 1.9955615048612094e-06,
|
|
"loss": 0.8466,
|
|
"mean_token_accuracy": 0.7799480080604553,
|
|
"num_tokens": 897558719.0,
|
|
"step": 8970
|
|
},
|
|
{
|
|
"entropy": 0.87109375,
|
|
"epoch": 1.2272944714002596,
|
|
"grad_norm": 0.29673475202429495,
|
|
"learning_rate": 1.9920388896716926e-06,
|
|
"loss": 0.8706,
|
|
"mean_token_accuracy": 0.7759987533092498,
|
|
"num_tokens": 898576322.0,
|
|
"step": 8980
|
|
},
|
|
{
|
|
"entropy": 0.8802734375,
|
|
"epoch": 1.2286612451308685,
|
|
"grad_norm": 0.273112081016348,
|
|
"learning_rate": 1.9885162744821757e-06,
|
|
"loss": 0.8818,
|
|
"mean_token_accuracy": 0.7754036575555802,
|
|
"num_tokens": 899578542.0,
|
|
"step": 8990
|
|
},
|
|
{
|
|
"entropy": 0.86484375,
|
|
"epoch": 1.2300280188614774,
|
|
"grad_norm": 0.2904699721010549,
|
|
"learning_rate": 1.9849936592926592e-06,
|
|
"loss": 0.859,
|
|
"mean_token_accuracy": 0.7758816838264465,
|
|
"num_tokens": 900519968.0,
|
|
"step": 9000
|
|
},
|
|
{
|
|
"entropy": 0.8587890625,
|
|
"epoch": 1.2313947925920863,
|
|
"grad_norm": 0.2661063227959392,
|
|
"learning_rate": 1.9814710441031424e-06,
|
|
"loss": 0.8527,
|
|
"mean_token_accuracy": 0.7772663623094559,
|
|
"num_tokens": 901510031.0,
|
|
"step": 9010
|
|
},
|
|
{
|
|
"entropy": 0.855859375,
|
|
"epoch": 1.2327615663226954,
|
|
"grad_norm": 0.288105082706228,
|
|
"learning_rate": 1.9779484289136255e-06,
|
|
"loss": 0.84,
|
|
"mean_token_accuracy": 0.7816647380590439,
|
|
"num_tokens": 902496483.0,
|
|
"step": 9020
|
|
},
|
|
{
|
|
"entropy": 0.8978515625,
|
|
"epoch": 1.2341283400533043,
|
|
"grad_norm": 0.29271768430136214,
|
|
"learning_rate": 1.974425813724109e-06,
|
|
"loss": 0.8898,
|
|
"mean_token_accuracy": 0.7704720109701156,
|
|
"num_tokens": 903504215.0,
|
|
"step": 9030
|
|
},
|
|
{
|
|
"entropy": 0.848046875,
|
|
"epoch": 1.2354951137839132,
|
|
"grad_norm": 0.28444691682996587,
|
|
"learning_rate": 1.970903198534592e-06,
|
|
"loss": 0.8382,
|
|
"mean_token_accuracy": 0.7817092567682267,
|
|
"num_tokens": 904526445.0,
|
|
"step": 9040
|
|
},
|
|
{
|
|
"entropy": 0.8771484375,
|
|
"epoch": 1.236861887514522,
|
|
"grad_norm": 0.30512208447470646,
|
|
"learning_rate": 1.9673805833450753e-06,
|
|
"loss": 0.872,
|
|
"mean_token_accuracy": 0.7733955621719361,
|
|
"num_tokens": 905545886.0,
|
|
"step": 9050
|
|
},
|
|
{
|
|
"entropy": 0.891015625,
|
|
"epoch": 1.238228661245131,
|
|
"grad_norm": 0.2668341120269385,
|
|
"learning_rate": 1.963857968155559e-06,
|
|
"loss": 0.885,
|
|
"mean_token_accuracy": 0.7716741800308228,
|
|
"num_tokens": 906568445.0,
|
|
"step": 9060
|
|
},
|
|
{
|
|
"entropy": 0.8673828125,
|
|
"epoch": 1.2395954349757399,
|
|
"grad_norm": 0.38283248220483496,
|
|
"learning_rate": 1.960335352966042e-06,
|
|
"loss": 0.858,
|
|
"mean_token_accuracy": 0.7790177077054977,
|
|
"num_tokens": 907570386.0,
|
|
"step": 9070
|
|
},
|
|
{
|
|
"entropy": 0.889453125,
|
|
"epoch": 1.2409622087063488,
|
|
"grad_norm": 0.26622767176886525,
|
|
"learning_rate": 1.9568127377765255e-06,
|
|
"loss": 0.8836,
|
|
"mean_token_accuracy": 0.7694226264953613,
|
|
"num_tokens": 908598389.0,
|
|
"step": 9080
|
|
},
|
|
{
|
|
"entropy": 0.8376953125,
|
|
"epoch": 1.2423289824369577,
|
|
"grad_norm": 0.2667449351444791,
|
|
"learning_rate": 1.9532901225870086e-06,
|
|
"loss": 0.8331,
|
|
"mean_token_accuracy": 0.7823696911334992,
|
|
"num_tokens": 909666054.0,
|
|
"step": 9090
|
|
},
|
|
{
|
|
"entropy": 0.8716796875,
|
|
"epoch": 1.2436957561675666,
|
|
"grad_norm": 0.3327957898402581,
|
|
"learning_rate": 1.949767507397492e-06,
|
|
"loss": 0.8694,
|
|
"mean_token_accuracy": 0.7762972742319107,
|
|
"num_tokens": 910633216.0,
|
|
"step": 9100
|
|
},
|
|
{
|
|
"entropy": 0.8935546875,
|
|
"epoch": 1.2450625298981755,
|
|
"grad_norm": 0.2839416411781784,
|
|
"learning_rate": 1.9462448922079753e-06,
|
|
"loss": 0.8901,
|
|
"mean_token_accuracy": 0.7720283597707749,
|
|
"num_tokens": 911614061.0,
|
|
"step": 9110
|
|
},
|
|
{
|
|
"entropy": 0.8810546875,
|
|
"epoch": 1.2464293036287843,
|
|
"grad_norm": 0.2654573802873806,
|
|
"learning_rate": 1.942722277018459e-06,
|
|
"loss": 0.8787,
|
|
"mean_token_accuracy": 0.7740931302309036,
|
|
"num_tokens": 912593168.0,
|
|
"step": 9120
|
|
},
|
|
{
|
|
"entropy": 0.91015625,
|
|
"epoch": 1.2477960773593932,
|
|
"grad_norm": 0.2973851565001578,
|
|
"learning_rate": 1.939199661828942e-06,
|
|
"loss": 0.9134,
|
|
"mean_token_accuracy": 0.7661606281995773,
|
|
"num_tokens": 913577159.0,
|
|
"step": 9130
|
|
},
|
|
{
|
|
"entropy": 0.848828125,
|
|
"epoch": 1.2491628510900021,
|
|
"grad_norm": 0.2835949595704016,
|
|
"learning_rate": 1.9356770466394255e-06,
|
|
"loss": 0.8467,
|
|
"mean_token_accuracy": 0.7802795022726059,
|
|
"num_tokens": 914502010.0,
|
|
"step": 9140
|
|
},
|
|
{
|
|
"entropy": 0.8384765625,
|
|
"epoch": 1.250529624820611,
|
|
"grad_norm": 0.2655032423083131,
|
|
"learning_rate": 1.9321544314499086e-06,
|
|
"loss": 0.8328,
|
|
"mean_token_accuracy": 0.7820462614297867,
|
|
"num_tokens": 915493825.0,
|
|
"step": 9150
|
|
},
|
|
{
|
|
"entropy": 0.8708984375,
|
|
"epoch": 1.25189639855122,
|
|
"grad_norm": 0.32258967425297197,
|
|
"learning_rate": 1.9286318162603917e-06,
|
|
"loss": 0.871,
|
|
"mean_token_accuracy": 0.7735952287912369,
|
|
"num_tokens": 916504508.0,
|
|
"step": 9160
|
|
},
|
|
{
|
|
"entropy": 0.8375,
|
|
"epoch": 1.2532631722818288,
|
|
"grad_norm": 0.2612004554262283,
|
|
"learning_rate": 1.9251092010708753e-06,
|
|
"loss": 0.8378,
|
|
"mean_token_accuracy": 0.782549825310707,
|
|
"num_tokens": 917504219.0,
|
|
"step": 9170
|
|
},
|
|
{
|
|
"entropy": 0.887890625,
|
|
"epoch": 1.2546299460124377,
|
|
"grad_norm": 0.23597870927731035,
|
|
"learning_rate": 1.9215865858813584e-06,
|
|
"loss": 0.8883,
|
|
"mean_token_accuracy": 0.7714750617742538,
|
|
"num_tokens": 918499654.0,
|
|
"step": 9180
|
|
},
|
|
{
|
|
"entropy": 0.8912109375,
|
|
"epoch": 1.2559967197430466,
|
|
"grad_norm": 0.32479753889001944,
|
|
"learning_rate": 1.918063970691842e-06,
|
|
"loss": 0.8902,
|
|
"mean_token_accuracy": 0.769848495721817,
|
|
"num_tokens": 919501683.0,
|
|
"step": 9190
|
|
},
|
|
{
|
|
"entropy": 0.8728515625,
|
|
"epoch": 1.2573634934736555,
|
|
"grad_norm": 0.2878168384897092,
|
|
"learning_rate": 1.914541355502325e-06,
|
|
"loss": 0.8636,
|
|
"mean_token_accuracy": 0.7735674947500228,
|
|
"num_tokens": 920459327.0,
|
|
"step": 9200
|
|
},
|
|
{
|
|
"entropy": 0.8755859375,
|
|
"epoch": 1.2587302672042644,
|
|
"grad_norm": 0.26216708254332544,
|
|
"learning_rate": 1.9110187403128082e-06,
|
|
"loss": 0.8675,
|
|
"mean_token_accuracy": 0.7733732491731644,
|
|
"num_tokens": 921446659.0,
|
|
"step": 9210
|
|
},
|
|
{
|
|
"entropy": 0.868359375,
|
|
"epoch": 1.2600970409348733,
|
|
"grad_norm": 0.2872357551794017,
|
|
"learning_rate": 1.9074961251232918e-06,
|
|
"loss": 0.868,
|
|
"mean_token_accuracy": 0.7748338848352432,
|
|
"num_tokens": 922490909.0,
|
|
"step": 9220
|
|
},
|
|
{
|
|
"entropy": 0.8666015625,
|
|
"epoch": 1.2614638146654822,
|
|
"grad_norm": 0.31240808211046717,
|
|
"learning_rate": 1.903973509933775e-06,
|
|
"loss": 0.8595,
|
|
"mean_token_accuracy": 0.7748129427433014,
|
|
"num_tokens": 923472599.0,
|
|
"step": 9230
|
|
},
|
|
{
|
|
"entropy": 0.9005859375,
|
|
"epoch": 1.262830588396091,
|
|
"grad_norm": 0.2853257288891705,
|
|
"learning_rate": 1.9004508947442582e-06,
|
|
"loss": 0.8884,
|
|
"mean_token_accuracy": 0.7695591032505036,
|
|
"num_tokens": 924433510.0,
|
|
"step": 9240
|
|
},
|
|
{
|
|
"entropy": 0.8408203125,
|
|
"epoch": 1.2641973621267,
|
|
"grad_norm": 0.31460544818540304,
|
|
"learning_rate": 1.8969282795547418e-06,
|
|
"loss": 0.8393,
|
|
"mean_token_accuracy": 0.7821477174758911,
|
|
"num_tokens": 925407361.0,
|
|
"step": 9250
|
|
},
|
|
{
|
|
"entropy": 0.8650390625,
|
|
"epoch": 1.2655641358573089,
|
|
"grad_norm": 0.27758515726770133,
|
|
"learning_rate": 1.8934056643652249e-06,
|
|
"loss": 0.8625,
|
|
"mean_token_accuracy": 0.7759598553180694,
|
|
"num_tokens": 926389273.0,
|
|
"step": 9260
|
|
},
|
|
{
|
|
"entropy": 0.86171875,
|
|
"epoch": 1.2669309095879178,
|
|
"grad_norm": 0.27098063718144316,
|
|
"learning_rate": 1.889883049175708e-06,
|
|
"loss": 0.8575,
|
|
"mean_token_accuracy": 0.7767011612653733,
|
|
"num_tokens": 927373762.0,
|
|
"step": 9270
|
|
},
|
|
{
|
|
"entropy": 0.825390625,
|
|
"epoch": 1.2682976833185267,
|
|
"grad_norm": 0.2743542995015736,
|
|
"learning_rate": 1.8863604339861916e-06,
|
|
"loss": 0.8224,
|
|
"mean_token_accuracy": 0.784211739897728,
|
|
"num_tokens": 928371720.0,
|
|
"step": 9280
|
|
},
|
|
{
|
|
"entropy": 0.9015625,
|
|
"epoch": 1.2696644570491356,
|
|
"grad_norm": 0.28878610425697254,
|
|
"learning_rate": 1.8828378187966747e-06,
|
|
"loss": 0.9031,
|
|
"mean_token_accuracy": 0.7681753396987915,
|
|
"num_tokens": 929393058.0,
|
|
"step": 9290
|
|
},
|
|
{
|
|
"entropy": 0.854296875,
|
|
"epoch": 1.2710312307797444,
|
|
"grad_norm": 0.28971539015410286,
|
|
"learning_rate": 1.8793152036071582e-06,
|
|
"loss": 0.854,
|
|
"mean_token_accuracy": 0.7777727007865906,
|
|
"num_tokens": 930384776.0,
|
|
"step": 9300
|
|
},
|
|
{
|
|
"entropy": 0.8345703125,
|
|
"epoch": 1.2723980045103533,
|
|
"grad_norm": 0.2677724870933524,
|
|
"learning_rate": 1.8757925884176414e-06,
|
|
"loss": 0.8325,
|
|
"mean_token_accuracy": 0.7827881515026093,
|
|
"num_tokens": 931403568.0,
|
|
"step": 9310
|
|
},
|
|
{
|
|
"entropy": 0.8990234375,
|
|
"epoch": 1.2737647782409622,
|
|
"grad_norm": 0.27673491152291624,
|
|
"learning_rate": 1.8722699732281247e-06,
|
|
"loss": 0.9007,
|
|
"mean_token_accuracy": 0.7659657835960388,
|
|
"num_tokens": 932482171.0,
|
|
"step": 9320
|
|
},
|
|
{
|
|
"entropy": 0.9068359375,
|
|
"epoch": 1.2751315519715711,
|
|
"grad_norm": 0.27610533233984197,
|
|
"learning_rate": 1.868747358038608e-06,
|
|
"loss": 0.8951,
|
|
"mean_token_accuracy": 0.7694515109062194,
|
|
"num_tokens": 933468489.0,
|
|
"step": 9330
|
|
},
|
|
{
|
|
"entropy": 0.8533203125,
|
|
"epoch": 1.27649832570218,
|
|
"grad_norm": 0.27677323977663304,
|
|
"learning_rate": 1.8652247428490914e-06,
|
|
"loss": 0.8509,
|
|
"mean_token_accuracy": 0.7791351258754731,
|
|
"num_tokens": 934463214.0,
|
|
"step": 9340
|
|
},
|
|
{
|
|
"entropy": 0.8927734375,
|
|
"epoch": 1.277865099432789,
|
|
"grad_norm": 0.29908301571585316,
|
|
"learning_rate": 1.8617021276595745e-06,
|
|
"loss": 0.8938,
|
|
"mean_token_accuracy": 0.7709990113973617,
|
|
"num_tokens": 935448538.0,
|
|
"step": 9350
|
|
},
|
|
{
|
|
"entropy": 0.8732421875,
|
|
"epoch": 1.2792318731633978,
|
|
"grad_norm": 0.3257356655057338,
|
|
"learning_rate": 1.858179512470058e-06,
|
|
"loss": 0.8703,
|
|
"mean_token_accuracy": 0.7766079872846603,
|
|
"num_tokens": 936432921.0,
|
|
"step": 9360
|
|
},
|
|
{
|
|
"entropy": 0.8576171875,
|
|
"epoch": 1.2805986468940067,
|
|
"grad_norm": 0.25802828583719345,
|
|
"learning_rate": 1.8546568972805412e-06,
|
|
"loss": 0.8586,
|
|
"mean_token_accuracy": 0.778261935710907,
|
|
"num_tokens": 937500981.0,
|
|
"step": 9370
|
|
},
|
|
{
|
|
"entropy": 0.854296875,
|
|
"epoch": 1.2819654206246156,
|
|
"grad_norm": 0.2761886097668249,
|
|
"learning_rate": 1.8511342820910245e-06,
|
|
"loss": 0.8429,
|
|
"mean_token_accuracy": 0.7793655425310135,
|
|
"num_tokens": 938490489.0,
|
|
"step": 9380
|
|
},
|
|
{
|
|
"entropy": 0.8435546875,
|
|
"epoch": 1.2833321943552245,
|
|
"grad_norm": 0.32080949477596504,
|
|
"learning_rate": 1.8476116669015078e-06,
|
|
"loss": 0.8414,
|
|
"mean_token_accuracy": 0.7819464147090912,
|
|
"num_tokens": 939512159.0,
|
|
"step": 9390
|
|
},
|
|
{
|
|
"entropy": 0.85078125,
|
|
"epoch": 1.2846989680858334,
|
|
"grad_norm": 0.3169447036713485,
|
|
"learning_rate": 1.8440890517119912e-06,
|
|
"loss": 0.8404,
|
|
"mean_token_accuracy": 0.7806135803461075,
|
|
"num_tokens": 940527992.0,
|
|
"step": 9400
|
|
},
|
|
{
|
|
"entropy": 0.8685546875,
|
|
"epoch": 1.2860657418164423,
|
|
"grad_norm": 0.3212304988400311,
|
|
"learning_rate": 1.8405664365224743e-06,
|
|
"loss": 0.8611,
|
|
"mean_token_accuracy": 0.7764210283756257,
|
|
"num_tokens": 941494357.0,
|
|
"step": 9410
|
|
},
|
|
{
|
|
"entropy": 0.8796875,
|
|
"epoch": 1.2874325155470512,
|
|
"grad_norm": 0.29241965287303845,
|
|
"learning_rate": 1.8370438213329578e-06,
|
|
"loss": 0.8839,
|
|
"mean_token_accuracy": 0.7744765937328338,
|
|
"num_tokens": 942546843.0,
|
|
"step": 9420
|
|
},
|
|
{
|
|
"entropy": 0.894140625,
|
|
"epoch": 1.28879928927766,
|
|
"grad_norm": 0.2893543240859387,
|
|
"learning_rate": 1.833521206143441e-06,
|
|
"loss": 0.8814,
|
|
"mean_token_accuracy": 0.7718048334121704,
|
|
"num_tokens": 943552862.0,
|
|
"step": 9430
|
|
},
|
|
{
|
|
"entropy": 0.917578125,
|
|
"epoch": 1.290166063008269,
|
|
"grad_norm": 0.3405855592973438,
|
|
"learning_rate": 1.8299985909539245e-06,
|
|
"loss": 0.9095,
|
|
"mean_token_accuracy": 0.765747657418251,
|
|
"num_tokens": 944513500.0,
|
|
"step": 9440
|
|
},
|
|
{
|
|
"entropy": 0.88515625,
|
|
"epoch": 1.2915328367388779,
|
|
"grad_norm": 0.30440997492977945,
|
|
"learning_rate": 1.8264759757644076e-06,
|
|
"loss": 0.8769,
|
|
"mean_token_accuracy": 0.7744852572679519,
|
|
"num_tokens": 945501679.0,
|
|
"step": 9450
|
|
},
|
|
{
|
|
"entropy": 0.8515625,
|
|
"epoch": 1.2928996104694868,
|
|
"grad_norm": 0.3100216836081891,
|
|
"learning_rate": 1.822953360574891e-06,
|
|
"loss": 0.8493,
|
|
"mean_token_accuracy": 0.7785258650779724,
|
|
"num_tokens": 946450831.0,
|
|
"step": 9460
|
|
},
|
|
{
|
|
"entropy": 0.8533203125,
|
|
"epoch": 1.2942663842000957,
|
|
"grad_norm": 0.24350792225577814,
|
|
"learning_rate": 1.8194307453853743e-06,
|
|
"loss": 0.8426,
|
|
"mean_token_accuracy": 0.7810440123081207,
|
|
"num_tokens": 947437906.0,
|
|
"step": 9470
|
|
},
|
|
{
|
|
"entropy": 0.9009765625,
|
|
"epoch": 1.2956331579307045,
|
|
"grad_norm": 0.3235131767011773,
|
|
"learning_rate": 1.8159081301958576e-06,
|
|
"loss": 0.9027,
|
|
"mean_token_accuracy": 0.7692883968353271,
|
|
"num_tokens": 948462736.0,
|
|
"step": 9480
|
|
},
|
|
{
|
|
"entropy": 0.85625,
|
|
"epoch": 1.2969999316613134,
|
|
"grad_norm": 0.27664586948184866,
|
|
"learning_rate": 1.8123855150063408e-06,
|
|
"loss": 0.8492,
|
|
"mean_token_accuracy": 0.7808725893497467,
|
|
"num_tokens": 949480422.0,
|
|
"step": 9490
|
|
},
|
|
{
|
|
"entropy": 0.887109375,
|
|
"epoch": 1.2983667053919223,
|
|
"grad_norm": 0.2733516980875827,
|
|
"learning_rate": 1.8088628998168243e-06,
|
|
"loss": 0.8933,
|
|
"mean_token_accuracy": 0.7705414295196533,
|
|
"num_tokens": 950447938.0,
|
|
"step": 9500
|
|
},
|
|
{
|
|
"entropy": 0.8416015625,
|
|
"epoch": 1.2997334791225312,
|
|
"grad_norm": 0.2899920072658007,
|
|
"learning_rate": 1.8053402846273074e-06,
|
|
"loss": 0.8316,
|
|
"mean_token_accuracy": 0.7810243606567383,
|
|
"num_tokens": 951435110.0,
|
|
"step": 9510
|
|
},
|
|
{
|
|
"entropy": 0.8197265625,
|
|
"epoch": 1.3011002528531401,
|
|
"grad_norm": 0.25289007051027473,
|
|
"learning_rate": 1.8018176694377906e-06,
|
|
"loss": 0.8187,
|
|
"mean_token_accuracy": 0.7852918386459351,
|
|
"num_tokens": 952417964.0,
|
|
"step": 9520
|
|
},
|
|
{
|
|
"entropy": 0.8572265625,
|
|
"epoch": 1.302467026583749,
|
|
"grad_norm": 0.2737966318859512,
|
|
"learning_rate": 1.798295054248274e-06,
|
|
"loss": 0.8621,
|
|
"mean_token_accuracy": 0.7760204344987869,
|
|
"num_tokens": 953393628.0,
|
|
"step": 9530
|
|
},
|
|
{
|
|
"entropy": 0.8630859375,
|
|
"epoch": 1.303833800314358,
|
|
"grad_norm": 0.3137055599820381,
|
|
"learning_rate": 1.7947724390587572e-06,
|
|
"loss": 0.8516,
|
|
"mean_token_accuracy": 0.7781952768564224,
|
|
"num_tokens": 954373520.0,
|
|
"step": 9540
|
|
},
|
|
{
|
|
"entropy": 0.8876953125,
|
|
"epoch": 1.3052005740449668,
|
|
"grad_norm": 0.4786409264137401,
|
|
"learning_rate": 1.7912498238692408e-06,
|
|
"loss": 0.8884,
|
|
"mean_token_accuracy": 0.7708891332149506,
|
|
"num_tokens": 955395506.0,
|
|
"step": 9550
|
|
},
|
|
{
|
|
"entropy": 0.8580078125,
|
|
"epoch": 1.3065673477755757,
|
|
"grad_norm": 0.28160079007951816,
|
|
"learning_rate": 1.7877272086797239e-06,
|
|
"loss": 0.8659,
|
|
"mean_token_accuracy": 0.777359414100647,
|
|
"num_tokens": 956397900.0,
|
|
"step": 9560
|
|
},
|
|
{
|
|
"entropy": 0.874609375,
|
|
"epoch": 1.3079341215061846,
|
|
"grad_norm": 0.2756322650552585,
|
|
"learning_rate": 1.7842045934902072e-06,
|
|
"loss": 0.8709,
|
|
"mean_token_accuracy": 0.7730977296829223,
|
|
"num_tokens": 957416254.0,
|
|
"step": 9570
|
|
},
|
|
{
|
|
"entropy": 0.8755859375,
|
|
"epoch": 1.3093008952367935,
|
|
"grad_norm": 0.31315893411125917,
|
|
"learning_rate": 1.7806819783006908e-06,
|
|
"loss": 0.8677,
|
|
"mean_token_accuracy": 0.7749348819255829,
|
|
"num_tokens": 958405858.0,
|
|
"step": 9580
|
|
},
|
|
{
|
|
"entropy": 0.86953125,
|
|
"epoch": 1.3106676689674024,
|
|
"grad_norm": 0.27122604105011255,
|
|
"learning_rate": 1.777159363111174e-06,
|
|
"loss": 0.8751,
|
|
"mean_token_accuracy": 0.774826142191887,
|
|
"num_tokens": 959399206.0,
|
|
"step": 9590
|
|
},
|
|
{
|
|
"entropy": 0.8615234375,
|
|
"epoch": 1.3120344426980113,
|
|
"grad_norm": 0.2649236484867626,
|
|
"learning_rate": 1.773636747921657e-06,
|
|
"loss": 0.8458,
|
|
"mean_token_accuracy": 0.7790809571743011,
|
|
"num_tokens": 960346958.0,
|
|
"step": 9600
|
|
},
|
|
{
|
|
"entropy": 0.865625,
|
|
"epoch": 1.3134012164286202,
|
|
"grad_norm": 0.24733843483002937,
|
|
"learning_rate": 1.7701141327321406e-06,
|
|
"loss": 0.8514,
|
|
"mean_token_accuracy": 0.7794793486595154,
|
|
"num_tokens": 961380624.0,
|
|
"step": 9610
|
|
},
|
|
{
|
|
"entropy": 0.8701171875,
|
|
"epoch": 1.314767990159229,
|
|
"grad_norm": 0.2922829870197953,
|
|
"learning_rate": 1.7665915175426237e-06,
|
|
"loss": 0.8645,
|
|
"mean_token_accuracy": 0.7761024296283722,
|
|
"num_tokens": 962431264.0,
|
|
"step": 9620
|
|
},
|
|
{
|
|
"entropy": 0.855859375,
|
|
"epoch": 1.316134763889838,
|
|
"grad_norm": 0.27139512571121044,
|
|
"learning_rate": 1.763068902353107e-06,
|
|
"loss": 0.8586,
|
|
"mean_token_accuracy": 0.7759068310260773,
|
|
"num_tokens": 963408348.0,
|
|
"step": 9630
|
|
},
|
|
{
|
|
"entropy": 0.896484375,
|
|
"epoch": 1.3175015376204469,
|
|
"grad_norm": 0.28151716511572417,
|
|
"learning_rate": 1.7595462871635904e-06,
|
|
"loss": 0.8969,
|
|
"mean_token_accuracy": 0.7688102573156357,
|
|
"num_tokens": 964383286.0,
|
|
"step": 9640
|
|
},
|
|
{
|
|
"entropy": 0.8314453125,
|
|
"epoch": 1.3188683113510558,
|
|
"grad_norm": 0.27142446869520703,
|
|
"learning_rate": 1.7560236719740737e-06,
|
|
"loss": 0.8255,
|
|
"mean_token_accuracy": 0.7830659687519074,
|
|
"num_tokens": 965401787.0,
|
|
"step": 9650
|
|
},
|
|
{
|
|
"entropy": 0.8830078125,
|
|
"epoch": 1.3202350850816646,
|
|
"grad_norm": 0.265779923809107,
|
|
"learning_rate": 1.752501056784557e-06,
|
|
"loss": 0.8699,
|
|
"mean_token_accuracy": 0.7777509748935699,
|
|
"num_tokens": 966428379.0,
|
|
"step": 9660
|
|
},
|
|
{
|
|
"entropy": 0.869921875,
|
|
"epoch": 1.3216018588122735,
|
|
"grad_norm": 0.28878914372956976,
|
|
"learning_rate": 1.7489784415950404e-06,
|
|
"loss": 0.8674,
|
|
"mean_token_accuracy": 0.7736344426870346,
|
|
"num_tokens": 967389830.0,
|
|
"step": 9670
|
|
},
|
|
{
|
|
"entropy": 0.8482421875,
|
|
"epoch": 1.3229686325428824,
|
|
"grad_norm": 0.28150383792295774,
|
|
"learning_rate": 1.7454558264055235e-06,
|
|
"loss": 0.8358,
|
|
"mean_token_accuracy": 0.7794744908809662,
|
|
"num_tokens": 968365035.0,
|
|
"step": 9680
|
|
},
|
|
{
|
|
"entropy": 0.86484375,
|
|
"epoch": 1.3243354062734913,
|
|
"grad_norm": 0.3650738480441285,
|
|
"learning_rate": 1.741933211216007e-06,
|
|
"loss": 0.8609,
|
|
"mean_token_accuracy": 0.7793535441160202,
|
|
"num_tokens": 969327581.0,
|
|
"step": 9690
|
|
},
|
|
{
|
|
"entropy": 0.88203125,
|
|
"epoch": 1.3257021800041002,
|
|
"grad_norm": 0.28328549343218073,
|
|
"learning_rate": 1.7384105960264902e-06,
|
|
"loss": 0.8756,
|
|
"mean_token_accuracy": 0.7754467606544495,
|
|
"num_tokens": 970334798.0,
|
|
"step": 9700
|
|
},
|
|
{
|
|
"entropy": 0.9080078125,
|
|
"epoch": 1.3270689537347091,
|
|
"grad_norm": 0.3319907150281096,
|
|
"learning_rate": 1.7348879808369735e-06,
|
|
"loss": 0.8993,
|
|
"mean_token_accuracy": 0.7663931667804718,
|
|
"num_tokens": 971361769.0,
|
|
"step": 9710
|
|
},
|
|
{
|
|
"entropy": 0.8556640625,
|
|
"epoch": 1.328435727465318,
|
|
"grad_norm": 0.2964947073771662,
|
|
"learning_rate": 1.7313653656474568e-06,
|
|
"loss": 0.8665,
|
|
"mean_token_accuracy": 0.7754402935504914,
|
|
"num_tokens": 972383063.0,
|
|
"step": 9720
|
|
},
|
|
{
|
|
"entropy": 0.890625,
|
|
"epoch": 1.329802501195927,
|
|
"grad_norm": 0.32857281179190895,
|
|
"learning_rate": 1.7278427504579402e-06,
|
|
"loss": 0.8925,
|
|
"mean_token_accuracy": 0.7712470620870591,
|
|
"num_tokens": 973414724.0,
|
|
"step": 9730
|
|
},
|
|
{
|
|
"entropy": 0.8849609375,
|
|
"epoch": 1.3311692749265358,
|
|
"grad_norm": 0.31343374092761156,
|
|
"learning_rate": 1.7243201352684233e-06,
|
|
"loss": 0.8957,
|
|
"mean_token_accuracy": 0.7692920953035355,
|
|
"num_tokens": 974456506.0,
|
|
"step": 9740
|
|
},
|
|
{
|
|
"entropy": 0.8861328125,
|
|
"epoch": 1.3325360486571447,
|
|
"grad_norm": 0.30863582143399093,
|
|
"learning_rate": 1.7207975200789068e-06,
|
|
"loss": 0.8837,
|
|
"mean_token_accuracy": 0.7712534755468369,
|
|
"num_tokens": 975457144.0,
|
|
"step": 9750
|
|
},
|
|
{
|
|
"entropy": 0.86015625,
|
|
"epoch": 1.3339028223877536,
|
|
"grad_norm": 0.31999143831684135,
|
|
"learning_rate": 1.71727490488939e-06,
|
|
"loss": 0.8534,
|
|
"mean_token_accuracy": 0.7781889319419861,
|
|
"num_tokens": 976407050.0,
|
|
"step": 9760
|
|
},
|
|
{
|
|
"entropy": 0.8587890625,
|
|
"epoch": 1.3352695961183625,
|
|
"grad_norm": 0.310762394125521,
|
|
"learning_rate": 1.7137522896998735e-06,
|
|
"loss": 0.8707,
|
|
"mean_token_accuracy": 0.7744028240442276,
|
|
"num_tokens": 977416497.0,
|
|
"step": 9770
|
|
},
|
|
{
|
|
"entropy": 0.8392578125,
|
|
"epoch": 1.3366363698489714,
|
|
"grad_norm": 0.29291565063600467,
|
|
"learning_rate": 1.7102296745103566e-06,
|
|
"loss": 0.8426,
|
|
"mean_token_accuracy": 0.7815774261951447,
|
|
"num_tokens": 978381112.0,
|
|
"step": 9780
|
|
},
|
|
{
|
|
"entropy": 0.8828125,
|
|
"epoch": 1.3380031435795803,
|
|
"grad_norm": 0.2936200386021605,
|
|
"learning_rate": 1.7067070593208398e-06,
|
|
"loss": 0.8754,
|
|
"mean_token_accuracy": 0.7726123780012131,
|
|
"num_tokens": 979413506.0,
|
|
"step": 9790
|
|
},
|
|
{
|
|
"entropy": 0.8740234375,
|
|
"epoch": 1.3393699173101892,
|
|
"grad_norm": 0.294195465068907,
|
|
"learning_rate": 1.7031844441313233e-06,
|
|
"loss": 0.8715,
|
|
"mean_token_accuracy": 0.7759861558675766,
|
|
"num_tokens": 980416508.0,
|
|
"step": 9800
|
|
},
|
|
{
|
|
"entropy": 0.85859375,
|
|
"epoch": 1.340736691040798,
|
|
"grad_norm": 0.27957733800245366,
|
|
"learning_rate": 1.6996618289418064e-06,
|
|
"loss": 0.8612,
|
|
"mean_token_accuracy": 0.7772588312625885,
|
|
"num_tokens": 981432229.0,
|
|
"step": 9810
|
|
},
|
|
{
|
|
"entropy": 0.8697265625,
|
|
"epoch": 1.342103464771407,
|
|
"grad_norm": 0.27873855769285805,
|
|
"learning_rate": 1.6961392137522898e-06,
|
|
"loss": 0.8629,
|
|
"mean_token_accuracy": 0.7747573733329773,
|
|
"num_tokens": 982384005.0,
|
|
"step": 9820
|
|
},
|
|
{
|
|
"entropy": 0.83828125,
|
|
"epoch": 1.3434702385020159,
|
|
"grad_norm": 0.28787528419872754,
|
|
"learning_rate": 1.6926165985627733e-06,
|
|
"loss": 0.8263,
|
|
"mean_token_accuracy": 0.7841740727424622,
|
|
"num_tokens": 983332625.0,
|
|
"step": 9830
|
|
},
|
|
{
|
|
"entropy": 0.88359375,
|
|
"epoch": 1.3448370122326248,
|
|
"grad_norm": 0.278602231761231,
|
|
"learning_rate": 1.6890939833732564e-06,
|
|
"loss": 0.8895,
|
|
"mean_token_accuracy": 0.7705972284078598,
|
|
"num_tokens": 984373257.0,
|
|
"step": 9840
|
|
},
|
|
{
|
|
"entropy": 0.8875,
|
|
"epoch": 1.3462037859632339,
|
|
"grad_norm": 0.29000784155829434,
|
|
"learning_rate": 1.6855713681837396e-06,
|
|
"loss": 0.8896,
|
|
"mean_token_accuracy": 0.7704685389995575,
|
|
"num_tokens": 985386348.0,
|
|
"step": 9850
|
|
},
|
|
{
|
|
"entropy": 0.848828125,
|
|
"epoch": 1.3475705596938428,
|
|
"grad_norm": 0.319053826481834,
|
|
"learning_rate": 1.6820487529942231e-06,
|
|
"loss": 0.8322,
|
|
"mean_token_accuracy": 0.7823724687099457,
|
|
"num_tokens": 986380984.0,
|
|
"step": 9860
|
|
},
|
|
{
|
|
"entropy": 0.8453125,
|
|
"epoch": 1.3489373334244517,
|
|
"grad_norm": 0.313293237851821,
|
|
"learning_rate": 1.6785261378047062e-06,
|
|
"loss": 0.8328,
|
|
"mean_token_accuracy": 0.7815274357795715,
|
|
"num_tokens": 987380186.0,
|
|
"step": 9870
|
|
},
|
|
{
|
|
"entropy": 0.862890625,
|
|
"epoch": 1.3503041071550606,
|
|
"grad_norm": 0.2929695464136778,
|
|
"learning_rate": 1.6750035226151898e-06,
|
|
"loss": 0.8522,
|
|
"mean_token_accuracy": 0.7781436920166016,
|
|
"num_tokens": 988334846.0,
|
|
"step": 9880
|
|
},
|
|
{
|
|
"entropy": 0.8625,
|
|
"epoch": 1.3516708808856694,
|
|
"grad_norm": 0.33173261938980997,
|
|
"learning_rate": 1.671480907425673e-06,
|
|
"loss": 0.8541,
|
|
"mean_token_accuracy": 0.7772812575101853,
|
|
"num_tokens": 989349103.0,
|
|
"step": 9890
|
|
},
|
|
{
|
|
"entropy": 0.879296875,
|
|
"epoch": 1.3530376546162783,
|
|
"grad_norm": 0.26973709329863693,
|
|
"learning_rate": 1.6679582922361562e-06,
|
|
"loss": 0.8806,
|
|
"mean_token_accuracy": 0.7745280742645264,
|
|
"num_tokens": 990316665.0,
|
|
"step": 9900
|
|
},
|
|
{
|
|
"entropy": 0.8666015625,
|
|
"epoch": 1.3544044283468872,
|
|
"grad_norm": 0.2811051591441551,
|
|
"learning_rate": 1.6644356770466396e-06,
|
|
"loss": 0.8659,
|
|
"mean_token_accuracy": 0.7750137120485305,
|
|
"num_tokens": 991331791.0,
|
|
"step": 9910
|
|
},
|
|
{
|
|
"entropy": 0.872265625,
|
|
"epoch": 1.3557712020774961,
|
|
"grad_norm": 0.30177110111942934,
|
|
"learning_rate": 1.660913061857123e-06,
|
|
"loss": 0.8673,
|
|
"mean_token_accuracy": 0.7742047369480133,
|
|
"num_tokens": 992342639.0,
|
|
"step": 9920
|
|
},
|
|
{
|
|
"entropy": 0.8333984375,
|
|
"epoch": 1.357137975808105,
|
|
"grad_norm": 0.2851483868243473,
|
|
"learning_rate": 1.657390446667606e-06,
|
|
"loss": 0.8197,
|
|
"mean_token_accuracy": 0.7857209980487824,
|
|
"num_tokens": 993343880.0,
|
|
"step": 9930
|
|
},
|
|
{
|
|
"entropy": 0.85546875,
|
|
"epoch": 1.358504749538714,
|
|
"grad_norm": 0.2846482055598666,
|
|
"learning_rate": 1.6538678314780896e-06,
|
|
"loss": 0.8558,
|
|
"mean_token_accuracy": 0.7787387520074844,
|
|
"num_tokens": 994330896.0,
|
|
"step": 9940
|
|
},
|
|
{
|
|
"entropy": 0.9255859375,
|
|
"epoch": 1.3598715232693228,
|
|
"grad_norm": 0.3124966784644721,
|
|
"learning_rate": 1.6503452162885727e-06,
|
|
"loss": 0.9242,
|
|
"mean_token_accuracy": 0.762260690331459,
|
|
"num_tokens": 995303202.0,
|
|
"step": 9950
|
|
},
|
|
{
|
|
"entropy": 0.8724609375,
|
|
"epoch": 1.3612382969999317,
|
|
"grad_norm": 0.30776940061163877,
|
|
"learning_rate": 1.646822601099056e-06,
|
|
"loss": 0.8625,
|
|
"mean_token_accuracy": 0.7753304958343505,
|
|
"num_tokens": 996303291.0,
|
|
"step": 9960
|
|
},
|
|
{
|
|
"entropy": 0.853515625,
|
|
"epoch": 1.3626050707305406,
|
|
"grad_norm": 0.295032497556634,
|
|
"learning_rate": 1.6432999859095394e-06,
|
|
"loss": 0.8629,
|
|
"mean_token_accuracy": 0.7779766619205475,
|
|
"num_tokens": 997303788.0,
|
|
"step": 9970
|
|
},
|
|
{
|
|
"entropy": 0.9103515625,
|
|
"epoch": 1.3639718444611495,
|
|
"grad_norm": 0.3236329896592833,
|
|
"learning_rate": 1.6397773707200227e-06,
|
|
"loss": 0.9047,
|
|
"mean_token_accuracy": 0.7660177528858185,
|
|
"num_tokens": 998319382.0,
|
|
"step": 9980
|
|
},
|
|
{
|
|
"entropy": 0.8544921875,
|
|
"epoch": 1.3653386181917584,
|
|
"grad_norm": 0.2731444059585204,
|
|
"learning_rate": 1.636254755530506e-06,
|
|
"loss": 0.856,
|
|
"mean_token_accuracy": 0.7763866126537323,
|
|
"num_tokens": 999281687.0,
|
|
"step": 9990
|
|
},
|
|
{
|
|
"entropy": 0.803125,
|
|
"epoch": 1.3667053919223673,
|
|
"grad_norm": 0.29863456760805873,
|
|
"learning_rate": 1.6327321403409894e-06,
|
|
"loss": 0.8044,
|
|
"mean_token_accuracy": 0.7882795959711075,
|
|
"num_tokens": 1000281300.0,
|
|
"step": 10000
|
|
},
|
|
{
|
|
"entropy": 0.87734375,
|
|
"epoch": 1.3680721656529762,
|
|
"grad_norm": 0.3302527317732476,
|
|
"learning_rate": 1.6292095251514725e-06,
|
|
"loss": 0.8823,
|
|
"mean_token_accuracy": 0.7731898188591003,
|
|
"num_tokens": 1001263809.0,
|
|
"step": 10010
|
|
},
|
|
{
|
|
"entropy": 0.8748046875,
|
|
"epoch": 1.369438939383585,
|
|
"grad_norm": 0.2812491482101027,
|
|
"learning_rate": 1.625686909961956e-06,
|
|
"loss": 0.8763,
|
|
"mean_token_accuracy": 0.7726812720298767,
|
|
"num_tokens": 1002285599.0,
|
|
"step": 10020
|
|
},
|
|
{
|
|
"entropy": 0.8837890625,
|
|
"epoch": 1.370805713114194,
|
|
"grad_norm": 0.32222958265221385,
|
|
"learning_rate": 1.6221642947724392e-06,
|
|
"loss": 0.8815,
|
|
"mean_token_accuracy": 0.7722633183002472,
|
|
"num_tokens": 1003235746.0,
|
|
"step": 10030
|
|
},
|
|
{
|
|
"entropy": 0.90546875,
|
|
"epoch": 1.3721724868448029,
|
|
"grad_norm": 0.3283893489535146,
|
|
"learning_rate": 1.6186416795829223e-06,
|
|
"loss": 0.8911,
|
|
"mean_token_accuracy": 0.7702903926372529,
|
|
"num_tokens": 1004259293.0,
|
|
"step": 10040
|
|
},
|
|
{
|
|
"entropy": 0.8744140625,
|
|
"epoch": 1.3735392605754118,
|
|
"grad_norm": 0.29814463034837946,
|
|
"learning_rate": 1.6151190643934058e-06,
|
|
"loss": 0.869,
|
|
"mean_token_accuracy": 0.77351393699646,
|
|
"num_tokens": 1005283339.0,
|
|
"step": 10050
|
|
},
|
|
{
|
|
"entropy": 0.8501953125,
|
|
"epoch": 1.3749060343060207,
|
|
"grad_norm": 0.3067566338316263,
|
|
"learning_rate": 1.6115964492038892e-06,
|
|
"loss": 0.8435,
|
|
"mean_token_accuracy": 0.7788958221673965,
|
|
"num_tokens": 1006277819.0,
|
|
"step": 10060
|
|
},
|
|
{
|
|
"entropy": 0.866796875,
|
|
"epoch": 1.3762728080366295,
|
|
"grad_norm": 0.29026815953332324,
|
|
"learning_rate": 1.6080738340143723e-06,
|
|
"loss": 0.8688,
|
|
"mean_token_accuracy": 0.7771556466817856,
|
|
"num_tokens": 1007297790.0,
|
|
"step": 10070
|
|
},
|
|
{
|
|
"entropy": 0.8439453125,
|
|
"epoch": 1.3776395817672384,
|
|
"grad_norm": 0.30088801990975333,
|
|
"learning_rate": 1.6045512188248559e-06,
|
|
"loss": 0.8437,
|
|
"mean_token_accuracy": 0.7811595350503922,
|
|
"num_tokens": 1008240040.0,
|
|
"step": 10080
|
|
},
|
|
{
|
|
"entropy": 0.8869140625,
|
|
"epoch": 1.3790063554978473,
|
|
"grad_norm": 0.2733823201742263,
|
|
"learning_rate": 1.601028603635339e-06,
|
|
"loss": 0.881,
|
|
"mean_token_accuracy": 0.7738099932670593,
|
|
"num_tokens": 1009207461.0,
|
|
"step": 10090
|
|
},
|
|
{
|
|
"entropy": 0.8689453125,
|
|
"epoch": 1.3803731292284562,
|
|
"grad_norm": 0.2931943712643613,
|
|
"learning_rate": 1.5975059884458225e-06,
|
|
"loss": 0.8747,
|
|
"mean_token_accuracy": 0.773636981844902,
|
|
"num_tokens": 1010260369.0,
|
|
"step": 10100
|
|
},
|
|
{
|
|
"entropy": 0.86015625,
|
|
"epoch": 1.3817399029590651,
|
|
"grad_norm": 0.28745274621637523,
|
|
"learning_rate": 1.5939833732563056e-06,
|
|
"loss": 0.8561,
|
|
"mean_token_accuracy": 0.7771642923355102,
|
|
"num_tokens": 1011266680.0,
|
|
"step": 10110
|
|
},
|
|
{
|
|
"entropy": 0.8640625,
|
|
"epoch": 1.383106676689674,
|
|
"grad_norm": 0.2514385624245379,
|
|
"learning_rate": 1.5904607580667888e-06,
|
|
"loss": 0.8599,
|
|
"mean_token_accuracy": 0.7761237561702728,
|
|
"num_tokens": 1012303037.0,
|
|
"step": 10120
|
|
},
|
|
{
|
|
"entropy": 0.8716796875,
|
|
"epoch": 1.384473450420283,
|
|
"grad_norm": 0.32152716960888383,
|
|
"learning_rate": 1.5869381428772723e-06,
|
|
"loss": 0.879,
|
|
"mean_token_accuracy": 0.7742128610610962,
|
|
"num_tokens": 1013288875.0,
|
|
"step": 10130
|
|
},
|
|
{
|
|
"entropy": 0.8841796875,
|
|
"epoch": 1.3858402241508918,
|
|
"grad_norm": 0.3161240860549004,
|
|
"learning_rate": 1.5834155276877554e-06,
|
|
"loss": 0.8795,
|
|
"mean_token_accuracy": 0.7741805225610733,
|
|
"num_tokens": 1014269825.0,
|
|
"step": 10140
|
|
},
|
|
{
|
|
"entropy": 0.8751953125,
|
|
"epoch": 1.3872069978815007,
|
|
"grad_norm": 0.276351423837556,
|
|
"learning_rate": 1.5798929124982388e-06,
|
|
"loss": 0.8665,
|
|
"mean_token_accuracy": 0.7734689742326737,
|
|
"num_tokens": 1015229258.0,
|
|
"step": 10150
|
|
},
|
|
{
|
|
"entropy": 0.86328125,
|
|
"epoch": 1.3885737716121096,
|
|
"grad_norm": 0.32405675215798696,
|
|
"learning_rate": 1.5763702973087221e-06,
|
|
"loss": 0.8669,
|
|
"mean_token_accuracy": 0.774903291463852,
|
|
"num_tokens": 1016201586.0,
|
|
"step": 10160
|
|
},
|
|
{
|
|
"entropy": 0.8310546875,
|
|
"epoch": 1.3899405453427185,
|
|
"grad_norm": 0.3389254229104075,
|
|
"learning_rate": 1.5728476821192054e-06,
|
|
"loss": 0.8304,
|
|
"mean_token_accuracy": 0.7840936541557312,
|
|
"num_tokens": 1017189084.0,
|
|
"step": 10170
|
|
},
|
|
{
|
|
"entropy": 0.90390625,
|
|
"epoch": 1.3913073190733274,
|
|
"grad_norm": 0.2681045410818905,
|
|
"learning_rate": 1.5693250669296886e-06,
|
|
"loss": 0.9028,
|
|
"mean_token_accuracy": 0.7675277024507523,
|
|
"num_tokens": 1018229400.0,
|
|
"step": 10180
|
|
},
|
|
{
|
|
"entropy": 0.8525390625,
|
|
"epoch": 1.3926740928039363,
|
|
"grad_norm": 0.28957304773648884,
|
|
"learning_rate": 1.5658024517401721e-06,
|
|
"loss": 0.8483,
|
|
"mean_token_accuracy": 0.7786338150501251,
|
|
"num_tokens": 1019223740.0,
|
|
"step": 10190
|
|
},
|
|
{
|
|
"entropy": 0.8330078125,
|
|
"epoch": 1.3940408665345452,
|
|
"grad_norm": 0.24934903540438788,
|
|
"learning_rate": 1.5622798365506552e-06,
|
|
"loss": 0.8331,
|
|
"mean_token_accuracy": 0.7836787343025208,
|
|
"num_tokens": 1020248644.0,
|
|
"step": 10200
|
|
},
|
|
{
|
|
"entropy": 0.8671875,
|
|
"epoch": 1.395407640265154,
|
|
"grad_norm": 0.2832573468241549,
|
|
"learning_rate": 1.5587572213611388e-06,
|
|
"loss": 0.8563,
|
|
"mean_token_accuracy": 0.7777988493442536,
|
|
"num_tokens": 1021242917.0,
|
|
"step": 10210
|
|
},
|
|
{
|
|
"entropy": 0.8984375,
|
|
"epoch": 1.396774413995763,
|
|
"grad_norm": 0.3027301740173215,
|
|
"learning_rate": 1.555234606171622e-06,
|
|
"loss": 0.8842,
|
|
"mean_token_accuracy": 0.7725995630025864,
|
|
"num_tokens": 1022258132.0,
|
|
"step": 10220
|
|
},
|
|
{
|
|
"entropy": 0.8732421875,
|
|
"epoch": 1.3981411877263719,
|
|
"grad_norm": 0.2641353191035651,
|
|
"learning_rate": 1.5517119909821052e-06,
|
|
"loss": 0.8662,
|
|
"mean_token_accuracy": 0.7748925298452377,
|
|
"num_tokens": 1023297650.0,
|
|
"step": 10230
|
|
},
|
|
{
|
|
"entropy": 0.8826171875,
|
|
"epoch": 1.3995079614569808,
|
|
"grad_norm": 0.32469469510928667,
|
|
"learning_rate": 1.5481893757925886e-06,
|
|
"loss": 0.8802,
|
|
"mean_token_accuracy": 0.7720128536224365,
|
|
"num_tokens": 1024320939.0,
|
|
"step": 10240
|
|
},
|
|
{
|
|
"entropy": 0.883203125,
|
|
"epoch": 1.4008747351875896,
|
|
"grad_norm": 0.32420811369467356,
|
|
"learning_rate": 1.544666760603072e-06,
|
|
"loss": 0.8771,
|
|
"mean_token_accuracy": 0.7725533664226532,
|
|
"num_tokens": 1025274770.0,
|
|
"step": 10250
|
|
},
|
|
{
|
|
"entropy": 0.9087890625,
|
|
"epoch": 1.4022415089181985,
|
|
"grad_norm": 0.31739679873516785,
|
|
"learning_rate": 1.541144145413555e-06,
|
|
"loss": 0.9113,
|
|
"mean_token_accuracy": 0.7657405793666839,
|
|
"num_tokens": 1026299910.0,
|
|
"step": 10260
|
|
},
|
|
{
|
|
"entropy": 0.8701171875,
|
|
"epoch": 1.4036082826488074,
|
|
"grad_norm": 0.2770452466086168,
|
|
"learning_rate": 1.5376215302240386e-06,
|
|
"loss": 0.8678,
|
|
"mean_token_accuracy": 0.7774371236562729,
|
|
"num_tokens": 1027370475.0,
|
|
"step": 10270
|
|
},
|
|
{
|
|
"entropy": 0.8654296875,
|
|
"epoch": 1.4049750563794163,
|
|
"grad_norm": 0.2652954517440594,
|
|
"learning_rate": 1.5340989150345217e-06,
|
|
"loss": 0.8683,
|
|
"mean_token_accuracy": 0.7762093275785447,
|
|
"num_tokens": 1028382536.0,
|
|
"step": 10280
|
|
},
|
|
{
|
|
"entropy": 0.830859375,
|
|
"epoch": 1.4063418301100252,
|
|
"grad_norm": 0.259451170425897,
|
|
"learning_rate": 1.5305762998450048e-06,
|
|
"loss": 0.8358,
|
|
"mean_token_accuracy": 0.7829228818416596,
|
|
"num_tokens": 1029368876.0,
|
|
"step": 10290
|
|
},
|
|
{
|
|
"entropy": 0.9041015625,
|
|
"epoch": 1.4077086038406341,
|
|
"grad_norm": 0.31457437866273447,
|
|
"learning_rate": 1.5270536846554884e-06,
|
|
"loss": 0.8997,
|
|
"mean_token_accuracy": 0.7691920280456543,
|
|
"num_tokens": 1030363887.0,
|
|
"step": 10300
|
|
},
|
|
{
|
|
"entropy": 0.8873046875,
|
|
"epoch": 1.409075377571243,
|
|
"grad_norm": 0.32217900139085415,
|
|
"learning_rate": 1.5235310694659717e-06,
|
|
"loss": 0.8848,
|
|
"mean_token_accuracy": 0.7700747460126877,
|
|
"num_tokens": 1031382465.0,
|
|
"step": 10310
|
|
},
|
|
{
|
|
"entropy": 0.8466796875,
|
|
"epoch": 1.410442151301852,
|
|
"grad_norm": 0.2919668658930878,
|
|
"learning_rate": 1.5200084542764548e-06,
|
|
"loss": 0.8303,
|
|
"mean_token_accuracy": 0.7817565530538559,
|
|
"num_tokens": 1032411055.0,
|
|
"step": 10320
|
|
},
|
|
{
|
|
"entropy": 0.8673828125,
|
|
"epoch": 1.4118089250324608,
|
|
"grad_norm": 0.3019031254166279,
|
|
"learning_rate": 1.5164858390869384e-06,
|
|
"loss": 0.8644,
|
|
"mean_token_accuracy": 0.7750849276781082,
|
|
"num_tokens": 1033412543.0,
|
|
"step": 10330
|
|
},
|
|
{
|
|
"entropy": 0.9390625,
|
|
"epoch": 1.4131756987630697,
|
|
"grad_norm": 0.3164585938479043,
|
|
"learning_rate": 1.5129632238974215e-06,
|
|
"loss": 0.9373,
|
|
"mean_token_accuracy": 0.7646792441606521,
|
|
"num_tokens": 1034371319.0,
|
|
"step": 10340
|
|
},
|
|
{
|
|
"entropy": 0.866015625,
|
|
"epoch": 1.4145424724936786,
|
|
"grad_norm": 0.29499241425933503,
|
|
"learning_rate": 1.509440608707905e-06,
|
|
"loss": 0.859,
|
|
"mean_token_accuracy": 0.7765154212713241,
|
|
"num_tokens": 1035403626.0,
|
|
"step": 10350
|
|
},
|
|
{
|
|
"entropy": 0.907421875,
|
|
"epoch": 1.4159092462242875,
|
|
"grad_norm": 0.2888862094031244,
|
|
"learning_rate": 1.5059179935183882e-06,
|
|
"loss": 0.9054,
|
|
"mean_token_accuracy": 0.7669157952070236,
|
|
"num_tokens": 1036396025.0,
|
|
"step": 10360
|
|
},
|
|
{
|
|
"entropy": 0.887109375,
|
|
"epoch": 1.4172760199548966,
|
|
"grad_norm": 0.2745766284807638,
|
|
"learning_rate": 1.5023953783288713e-06,
|
|
"loss": 0.892,
|
|
"mean_token_accuracy": 0.772228917479515,
|
|
"num_tokens": 1037404050.0,
|
|
"step": 10370
|
|
},
|
|
{
|
|
"entropy": 0.8744140625,
|
|
"epoch": 1.4186427936855055,
|
|
"grad_norm": 0.27007581998134383,
|
|
"learning_rate": 1.4988727631393549e-06,
|
|
"loss": 0.8704,
|
|
"mean_token_accuracy": 0.7730152010917664,
|
|
"num_tokens": 1038390015.0,
|
|
"step": 10380
|
|
},
|
|
{
|
|
"entropy": 0.8837890625,
|
|
"epoch": 1.4200095674161144,
|
|
"grad_norm": 0.3440321635094779,
|
|
"learning_rate": 1.495350147949838e-06,
|
|
"loss": 0.8815,
|
|
"mean_token_accuracy": 0.7734723627567291,
|
|
"num_tokens": 1039429565.0,
|
|
"step": 10390
|
|
},
|
|
{
|
|
"entropy": 0.8462890625,
|
|
"epoch": 1.4213763411467233,
|
|
"grad_norm": 0.2968419858596835,
|
|
"learning_rate": 1.4918275327603213e-06,
|
|
"loss": 0.8453,
|
|
"mean_token_accuracy": 0.7818903744220733,
|
|
"num_tokens": 1040377026.0,
|
|
"step": 10400
|
|
},
|
|
{
|
|
"entropy": 0.843359375,
|
|
"epoch": 1.4227431148773322,
|
|
"grad_norm": 0.3325327852429789,
|
|
"learning_rate": 1.4883049175708047e-06,
|
|
"loss": 0.8469,
|
|
"mean_token_accuracy": 0.7788823693990707,
|
|
"num_tokens": 1041327045.0,
|
|
"step": 10410
|
|
},
|
|
{
|
|
"entropy": 0.877734375,
|
|
"epoch": 1.424109888607941,
|
|
"grad_norm": 0.28207032446873415,
|
|
"learning_rate": 1.484782302381288e-06,
|
|
"loss": 0.8649,
|
|
"mean_token_accuracy": 0.7739810794591904,
|
|
"num_tokens": 1042337962.0,
|
|
"step": 10420
|
|
},
|
|
{
|
|
"entropy": 0.8875,
|
|
"epoch": 1.42547666233855,
|
|
"grad_norm": 0.3354768623236146,
|
|
"learning_rate": 1.4812596871917711e-06,
|
|
"loss": 0.8858,
|
|
"mean_token_accuracy": 0.77273530960083,
|
|
"num_tokens": 1043321165.0,
|
|
"step": 10430
|
|
},
|
|
{
|
|
"entropy": 0.855078125,
|
|
"epoch": 1.4268434360691589,
|
|
"grad_norm": 0.2792704434643619,
|
|
"learning_rate": 1.4777370720022547e-06,
|
|
"loss": 0.8505,
|
|
"mean_token_accuracy": 0.77798490524292,
|
|
"num_tokens": 1044311980.0,
|
|
"step": 10440
|
|
},
|
|
{
|
|
"entropy": 0.87109375,
|
|
"epoch": 1.4282102097997678,
|
|
"grad_norm": 0.26837085979112213,
|
|
"learning_rate": 1.4742144568127378e-06,
|
|
"loss": 0.8621,
|
|
"mean_token_accuracy": 0.7761414378881455,
|
|
"num_tokens": 1045346330.0,
|
|
"step": 10450
|
|
},
|
|
{
|
|
"entropy": 0.8396484375,
|
|
"epoch": 1.4295769835303767,
|
|
"grad_norm": 0.296231082116614,
|
|
"learning_rate": 1.4706918416232213e-06,
|
|
"loss": 0.8309,
|
|
"mean_token_accuracy": 0.7836050003767013,
|
|
"num_tokens": 1046325414.0,
|
|
"step": 10460
|
|
},
|
|
{
|
|
"entropy": 0.864453125,
|
|
"epoch": 1.4309437572609855,
|
|
"grad_norm": 0.2764023806869321,
|
|
"learning_rate": 1.4671692264337045e-06,
|
|
"loss": 0.8546,
|
|
"mean_token_accuracy": 0.7772158682346344,
|
|
"num_tokens": 1047362651.0,
|
|
"step": 10470
|
|
},
|
|
{
|
|
"entropy": 0.855078125,
|
|
"epoch": 1.4323105309915944,
|
|
"grad_norm": 0.2559584388457542,
|
|
"learning_rate": 1.4636466112441878e-06,
|
|
"loss": 0.85,
|
|
"mean_token_accuracy": 0.7775700837373734,
|
|
"num_tokens": 1048404198.0,
|
|
"step": 10480
|
|
},
|
|
{
|
|
"entropy": 0.839453125,
|
|
"epoch": 1.4336773047222033,
|
|
"grad_norm": 0.2597804862730114,
|
|
"learning_rate": 1.4601239960546711e-06,
|
|
"loss": 0.8317,
|
|
"mean_token_accuracy": 0.7827340632677078,
|
|
"num_tokens": 1049402313.0,
|
|
"step": 10490
|
|
},
|
|
{
|
|
"entropy": 0.8791015625,
|
|
"epoch": 1.4350440784528122,
|
|
"grad_norm": 0.27926848936690957,
|
|
"learning_rate": 1.4566013808651545e-06,
|
|
"loss": 0.8679,
|
|
"mean_token_accuracy": 0.7735607624053955,
|
|
"num_tokens": 1050455539.0,
|
|
"step": 10500
|
|
},
|
|
{
|
|
"entropy": 0.8376953125,
|
|
"epoch": 1.4364108521834211,
|
|
"grad_norm": 0.2698103539415137,
|
|
"learning_rate": 1.4530787656756376e-06,
|
|
"loss": 0.8414,
|
|
"mean_token_accuracy": 0.7798713564872741,
|
|
"num_tokens": 1051486678.0,
|
|
"step": 10510
|
|
},
|
|
{
|
|
"entropy": 0.845703125,
|
|
"epoch": 1.43777762591403,
|
|
"grad_norm": 0.31314806346126556,
|
|
"learning_rate": 1.4495561504861211e-06,
|
|
"loss": 0.8341,
|
|
"mean_token_accuracy": 0.781172975897789,
|
|
"num_tokens": 1052531712.0,
|
|
"step": 10520
|
|
},
|
|
{
|
|
"entropy": 0.8787109375,
|
|
"epoch": 1.439144399644639,
|
|
"grad_norm": 0.2662594553576196,
|
|
"learning_rate": 1.4460335352966043e-06,
|
|
"loss": 0.875,
|
|
"mean_token_accuracy": 0.7766796737909317,
|
|
"num_tokens": 1053517602.0,
|
|
"step": 10530
|
|
},
|
|
{
|
|
"entropy": 0.9220703125,
|
|
"epoch": 1.4405111733752478,
|
|
"grad_norm": 0.2846024319493406,
|
|
"learning_rate": 1.4425109201070876e-06,
|
|
"loss": 0.9136,
|
|
"mean_token_accuracy": 0.7659439861774444,
|
|
"num_tokens": 1054603242.0,
|
|
"step": 10540
|
|
},
|
|
{
|
|
"entropy": 0.87578125,
|
|
"epoch": 1.4418779471058567,
|
|
"grad_norm": 0.3011218729314113,
|
|
"learning_rate": 1.438988304917571e-06,
|
|
"loss": 0.8684,
|
|
"mean_token_accuracy": 0.7742231547832489,
|
|
"num_tokens": 1055631508.0,
|
|
"step": 10550
|
|
},
|
|
{
|
|
"entropy": 0.8703125,
|
|
"epoch": 1.4432447208364656,
|
|
"grad_norm": 0.2688450964018825,
|
|
"learning_rate": 1.4354656897280543e-06,
|
|
"loss": 0.8736,
|
|
"mean_token_accuracy": 0.774561008810997,
|
|
"num_tokens": 1056647059.0,
|
|
"step": 10560
|
|
},
|
|
{
|
|
"entropy": 0.858984375,
|
|
"epoch": 1.4446114945670745,
|
|
"grad_norm": 0.278558606996924,
|
|
"learning_rate": 1.4319430745385376e-06,
|
|
"loss": 0.8531,
|
|
"mean_token_accuracy": 0.7769123136997222,
|
|
"num_tokens": 1057642547.0,
|
|
"step": 10570
|
|
},
|
|
{
|
|
"entropy": 0.8654296875,
|
|
"epoch": 1.4459782682976834,
|
|
"grad_norm": 0.2699502118719808,
|
|
"learning_rate": 1.428420459349021e-06,
|
|
"loss": 0.8594,
|
|
"mean_token_accuracy": 0.7777760267257691,
|
|
"num_tokens": 1058636834.0,
|
|
"step": 10580
|
|
},
|
|
{
|
|
"entropy": 0.833203125,
|
|
"epoch": 1.4473450420282923,
|
|
"grad_norm": 0.3091919656540769,
|
|
"learning_rate": 1.424897844159504e-06,
|
|
"loss": 0.832,
|
|
"mean_token_accuracy": 0.7834525495767594,
|
|
"num_tokens": 1059632469.0,
|
|
"step": 10590
|
|
},
|
|
{
|
|
"entropy": 0.8548828125,
|
|
"epoch": 1.4487118157589012,
|
|
"grad_norm": 0.40628237408823725,
|
|
"learning_rate": 1.4213752289699876e-06,
|
|
"loss": 0.8497,
|
|
"mean_token_accuracy": 0.7787525236606598,
|
|
"num_tokens": 1060625056.0,
|
|
"step": 10600
|
|
},
|
|
{
|
|
"entropy": 0.8490234375,
|
|
"epoch": 1.45007858948951,
|
|
"grad_norm": 0.2549457821322234,
|
|
"learning_rate": 1.4178526137804707e-06,
|
|
"loss": 0.8464,
|
|
"mean_token_accuracy": 0.7795799493789672,
|
|
"num_tokens": 1061599213.0,
|
|
"step": 10610
|
|
},
|
|
{
|
|
"entropy": 0.85,
|
|
"epoch": 1.451445363220119,
|
|
"grad_norm": 0.2765979013567755,
|
|
"learning_rate": 1.4143299985909538e-06,
|
|
"loss": 0.8402,
|
|
"mean_token_accuracy": 0.7830659836530686,
|
|
"num_tokens": 1062562905.0,
|
|
"step": 10620
|
|
},
|
|
{
|
|
"entropy": 0.8583984375,
|
|
"epoch": 1.4528121369507279,
|
|
"grad_norm": 0.30228190154935475,
|
|
"learning_rate": 1.4108073834014374e-06,
|
|
"loss": 0.8454,
|
|
"mean_token_accuracy": 0.779165831208229,
|
|
"num_tokens": 1063588219.0,
|
|
"step": 10630
|
|
},
|
|
{
|
|
"entropy": 0.8560546875,
|
|
"epoch": 1.4541789106813368,
|
|
"grad_norm": 0.31309745511048215,
|
|
"learning_rate": 1.4072847682119205e-06,
|
|
"loss": 0.8585,
|
|
"mean_token_accuracy": 0.7773567765951157,
|
|
"num_tokens": 1064608630.0,
|
|
"step": 10640
|
|
},
|
|
{
|
|
"entropy": 0.8513671875,
|
|
"epoch": 1.4555456844119457,
|
|
"grad_norm": 0.27661817261182714,
|
|
"learning_rate": 1.4037621530224039e-06,
|
|
"loss": 0.8451,
|
|
"mean_token_accuracy": 0.7801015704870224,
|
|
"num_tokens": 1065665294.0,
|
|
"step": 10650
|
|
},
|
|
{
|
|
"entropy": 0.844140625,
|
|
"epoch": 1.4569124581425545,
|
|
"grad_norm": 0.2686250002931729,
|
|
"learning_rate": 1.4002395378328872e-06,
|
|
"loss": 0.8432,
|
|
"mean_token_accuracy": 0.7785424023866654,
|
|
"num_tokens": 1066709379.0,
|
|
"step": 10660
|
|
},
|
|
{
|
|
"entropy": 0.833984375,
|
|
"epoch": 1.4582792318731634,
|
|
"grad_norm": 0.289527881011814,
|
|
"learning_rate": 1.3967169226433705e-06,
|
|
"loss": 0.8282,
|
|
"mean_token_accuracy": 0.7839784532785415,
|
|
"num_tokens": 1067711140.0,
|
|
"step": 10670
|
|
},
|
|
{
|
|
"entropy": 0.882421875,
|
|
"epoch": 1.4596460056037723,
|
|
"grad_norm": 0.2775137508282416,
|
|
"learning_rate": 1.393194307453854e-06,
|
|
"loss": 0.8746,
|
|
"mean_token_accuracy": 0.7743588328361511,
|
|
"num_tokens": 1068730040.0,
|
|
"step": 10680
|
|
},
|
|
{
|
|
"entropy": 0.899609375,
|
|
"epoch": 1.4610127793343812,
|
|
"grad_norm": 0.2737194247036859,
|
|
"learning_rate": 1.3896716922643372e-06,
|
|
"loss": 0.8882,
|
|
"mean_token_accuracy": 0.7687178999185562,
|
|
"num_tokens": 1069711609.0,
|
|
"step": 10690
|
|
},
|
|
{
|
|
"entropy": 0.84921875,
|
|
"epoch": 1.4623795530649901,
|
|
"grad_norm": 0.26332911860028346,
|
|
"learning_rate": 1.3861490770748203e-06,
|
|
"loss": 0.85,
|
|
"mean_token_accuracy": 0.7789302170276642,
|
|
"num_tokens": 1070725876.0,
|
|
"step": 10700
|
|
},
|
|
{
|
|
"entropy": 0.875,
|
|
"epoch": 1.463746326795599,
|
|
"grad_norm": 0.32070509758113175,
|
|
"learning_rate": 1.3826264618853039e-06,
|
|
"loss": 0.8725,
|
|
"mean_token_accuracy": 0.7732052206993103,
|
|
"num_tokens": 1071668147.0,
|
|
"step": 10710
|
|
},
|
|
{
|
|
"entropy": 0.8859375,
|
|
"epoch": 1.465113100526208,
|
|
"grad_norm": 0.25816830774513616,
|
|
"learning_rate": 1.379103846695787e-06,
|
|
"loss": 0.8885,
|
|
"mean_token_accuracy": 0.7707754164934159,
|
|
"num_tokens": 1072743410.0,
|
|
"step": 10720
|
|
},
|
|
{
|
|
"entropy": 0.880859375,
|
|
"epoch": 1.4664798742568168,
|
|
"grad_norm": 0.30010829067753947,
|
|
"learning_rate": 1.3755812315062703e-06,
|
|
"loss": 0.8834,
|
|
"mean_token_accuracy": 0.771925276517868,
|
|
"num_tokens": 1073736824.0,
|
|
"step": 10730
|
|
},
|
|
{
|
|
"entropy": 0.8455078125,
|
|
"epoch": 1.4678466479874257,
|
|
"grad_norm": 0.2724116341581341,
|
|
"learning_rate": 1.3720586163167537e-06,
|
|
"loss": 0.8489,
|
|
"mean_token_accuracy": 0.777256777882576,
|
|
"num_tokens": 1074729635.0,
|
|
"step": 10740
|
|
},
|
|
{
|
|
"entropy": 0.8990234375,
|
|
"epoch": 1.4692134217180346,
|
|
"grad_norm": 0.35867023643596596,
|
|
"learning_rate": 1.368536001127237e-06,
|
|
"loss": 0.9032,
|
|
"mean_token_accuracy": 0.7697528690099716,
|
|
"num_tokens": 1075697703.0,
|
|
"step": 10750
|
|
},
|
|
{
|
|
"entropy": 0.83046875,
|
|
"epoch": 1.4705801954486435,
|
|
"grad_norm": 0.2677465212549416,
|
|
"learning_rate": 1.3650133859377201e-06,
|
|
"loss": 0.8208,
|
|
"mean_token_accuracy": 0.7827497392892837,
|
|
"num_tokens": 1076694087.0,
|
|
"step": 10760
|
|
},
|
|
{
|
|
"entropy": 0.8771484375,
|
|
"epoch": 1.4719469691792524,
|
|
"grad_norm": 0.3022206903131519,
|
|
"learning_rate": 1.3614907707482037e-06,
|
|
"loss": 0.8681,
|
|
"mean_token_accuracy": 0.7751612037420272,
|
|
"num_tokens": 1077667027.0,
|
|
"step": 10770
|
|
},
|
|
{
|
|
"entropy": 0.9330078125,
|
|
"epoch": 1.4733137429098613,
|
|
"grad_norm": 0.2999463712257374,
|
|
"learning_rate": 1.3579681555586868e-06,
|
|
"loss": 0.9163,
|
|
"mean_token_accuracy": 0.765855661034584,
|
|
"num_tokens": 1078690083.0,
|
|
"step": 10780
|
|
},
|
|
{
|
|
"entropy": 0.8708984375,
|
|
"epoch": 1.4746805166404702,
|
|
"grad_norm": 0.2563696976644131,
|
|
"learning_rate": 1.3544455403691703e-06,
|
|
"loss": 0.8645,
|
|
"mean_token_accuracy": 0.7758054822683335,
|
|
"num_tokens": 1079723737.0,
|
|
"step": 10790
|
|
},
|
|
{
|
|
"entropy": 0.8583984375,
|
|
"epoch": 1.476047290371079,
|
|
"grad_norm": 0.2904473432706252,
|
|
"learning_rate": 1.3509229251796535e-06,
|
|
"loss": 0.8465,
|
|
"mean_token_accuracy": 0.7796220630407333,
|
|
"num_tokens": 1080680630.0,
|
|
"step": 10800
|
|
},
|
|
{
|
|
"entropy": 0.874609375,
|
|
"epoch": 1.477414064101688,
|
|
"grad_norm": 0.30539169209075323,
|
|
"learning_rate": 1.3474003099901368e-06,
|
|
"loss": 0.8641,
|
|
"mean_token_accuracy": 0.7752112507820129,
|
|
"num_tokens": 1081660055.0,
|
|
"step": 10810
|
|
},
|
|
{
|
|
"entropy": 0.853515625,
|
|
"epoch": 1.4787808378322969,
|
|
"grad_norm": 0.2802226552563608,
|
|
"learning_rate": 1.3438776948006201e-06,
|
|
"loss": 0.8375,
|
|
"mean_token_accuracy": 0.7818818300962448,
|
|
"num_tokens": 1082647817.0,
|
|
"step": 10820
|
|
},
|
|
{
|
|
"entropy": 0.8345703125,
|
|
"epoch": 1.4801476115629058,
|
|
"grad_norm": 0.2620922994036499,
|
|
"learning_rate": 1.3403550796111035e-06,
|
|
"loss": 0.8326,
|
|
"mean_token_accuracy": 0.7841039150953293,
|
|
"num_tokens": 1083655902.0,
|
|
"step": 10830
|
|
},
|
|
{
|
|
"entropy": 0.8798828125,
|
|
"epoch": 1.4815143852935146,
|
|
"grad_norm": 0.2753337479007652,
|
|
"learning_rate": 1.3368324644215866e-06,
|
|
"loss": 0.8651,
|
|
"mean_token_accuracy": 0.7753425061702728,
|
|
"num_tokens": 1084658261.0,
|
|
"step": 10840
|
|
},
|
|
{
|
|
"entropy": 0.82265625,
|
|
"epoch": 1.4828811590241235,
|
|
"grad_norm": 0.2680592045169861,
|
|
"learning_rate": 1.3333098492320701e-06,
|
|
"loss": 0.8176,
|
|
"mean_token_accuracy": 0.7848248302936554,
|
|
"num_tokens": 1085681109.0,
|
|
"step": 10850
|
|
},
|
|
{
|
|
"entropy": 0.8865234375,
|
|
"epoch": 1.4842479327547324,
|
|
"grad_norm": 0.3101376622568282,
|
|
"learning_rate": 1.3297872340425533e-06,
|
|
"loss": 0.8743,
|
|
"mean_token_accuracy": 0.772421944141388,
|
|
"num_tokens": 1086654251.0,
|
|
"step": 10860
|
|
},
|
|
{
|
|
"entropy": 0.8927734375,
|
|
"epoch": 1.4856147064853413,
|
|
"grad_norm": 0.2796953364432048,
|
|
"learning_rate": 1.3262646188530364e-06,
|
|
"loss": 0.8947,
|
|
"mean_token_accuracy": 0.7702005475759506,
|
|
"num_tokens": 1087700800.0,
|
|
"step": 10870
|
|
},
|
|
{
|
|
"entropy": 0.8486328125,
|
|
"epoch": 1.4869814802159502,
|
|
"grad_norm": 0.2586238285701018,
|
|
"learning_rate": 1.32274200366352e-06,
|
|
"loss": 0.8493,
|
|
"mean_token_accuracy": 0.7808002680540085,
|
|
"num_tokens": 1088698909.0,
|
|
"step": 10880
|
|
},
|
|
{
|
|
"entropy": 0.8466796875,
|
|
"epoch": 1.4883482539465591,
|
|
"grad_norm": 0.29299920692092096,
|
|
"learning_rate": 1.319219388474003e-06,
|
|
"loss": 0.8469,
|
|
"mean_token_accuracy": 0.7812377452850342,
|
|
"num_tokens": 1089661133.0,
|
|
"step": 10890
|
|
},
|
|
{
|
|
"entropy": 0.891015625,
|
|
"epoch": 1.489715027677168,
|
|
"grad_norm": 0.26916979010717124,
|
|
"learning_rate": 1.3156967732844866e-06,
|
|
"loss": 0.8903,
|
|
"mean_token_accuracy": 0.771129384636879,
|
|
"num_tokens": 1090668495.0,
|
|
"step": 10900
|
|
},
|
|
{
|
|
"entropy": 0.833203125,
|
|
"epoch": 1.491081801407777,
|
|
"grad_norm": 0.3014457432142704,
|
|
"learning_rate": 1.31217415809497e-06,
|
|
"loss": 0.826,
|
|
"mean_token_accuracy": 0.7844257771968841,
|
|
"num_tokens": 1091695695.0,
|
|
"step": 10910
|
|
},
|
|
{
|
|
"entropy": 0.890234375,
|
|
"epoch": 1.4924485751383858,
|
|
"grad_norm": 0.27447151211401183,
|
|
"learning_rate": 1.308651542905453e-06,
|
|
"loss": 0.8994,
|
|
"mean_token_accuracy": 0.7682587534189225,
|
|
"num_tokens": 1092761308.0,
|
|
"step": 10920
|
|
},
|
|
{
|
|
"entropy": 0.8609375,
|
|
"epoch": 1.4938153488689947,
|
|
"grad_norm": 0.3631190424149563,
|
|
"learning_rate": 1.3051289277159366e-06,
|
|
"loss": 0.8594,
|
|
"mean_token_accuracy": 0.777430072426796,
|
|
"num_tokens": 1093736671.0,
|
|
"step": 10930
|
|
},
|
|
{
|
|
"entropy": 0.869921875,
|
|
"epoch": 1.4951821225996036,
|
|
"grad_norm": 0.28717421117205405,
|
|
"learning_rate": 1.3016063125264197e-06,
|
|
"loss": 0.8719,
|
|
"mean_token_accuracy": 0.7759289056062698,
|
|
"num_tokens": 1094707485.0,
|
|
"step": 10940
|
|
},
|
|
{
|
|
"entropy": 0.8857421875,
|
|
"epoch": 1.4965488963302125,
|
|
"grad_norm": 0.30349812750310895,
|
|
"learning_rate": 1.2980836973369029e-06,
|
|
"loss": 0.8801,
|
|
"mean_token_accuracy": 0.7711030870676041,
|
|
"num_tokens": 1095682234.0,
|
|
"step": 10950
|
|
},
|
|
{
|
|
"entropy": 0.856640625,
|
|
"epoch": 1.4979156700608214,
|
|
"grad_norm": 0.30665800878232197,
|
|
"learning_rate": 1.2945610821473864e-06,
|
|
"loss": 0.8561,
|
|
"mean_token_accuracy": 0.7781047821044922,
|
|
"num_tokens": 1096636241.0,
|
|
"step": 10960
|
|
},
|
|
{
|
|
"entropy": 0.8423828125,
|
|
"epoch": 1.4992824437914303,
|
|
"grad_norm": 0.2827149603730888,
|
|
"learning_rate": 1.2910384669578695e-06,
|
|
"loss": 0.8365,
|
|
"mean_token_accuracy": 0.7843181908130645,
|
|
"num_tokens": 1097641314.0,
|
|
"step": 10970
|
|
},
|
|
{
|
|
"entropy": 0.8421875,
|
|
"epoch": 1.5006492175220392,
|
|
"grad_norm": 0.3069189245422636,
|
|
"learning_rate": 1.2875158517683529e-06,
|
|
"loss": 0.8435,
|
|
"mean_token_accuracy": 0.779144087433815,
|
|
"num_tokens": 1098615649.0,
|
|
"step": 10980
|
|
},
|
|
{
|
|
"entropy": 0.8939453125,
|
|
"epoch": 1.502015991252648,
|
|
"grad_norm": 0.3068060025509964,
|
|
"learning_rate": 1.2839932365788362e-06,
|
|
"loss": 0.8867,
|
|
"mean_token_accuracy": 0.7705550909042358,
|
|
"num_tokens": 1099606073.0,
|
|
"step": 10990
|
|
},
|
|
{
|
|
"entropy": 0.8734375,
|
|
"epoch": 1.503382764983257,
|
|
"grad_norm": 0.26063039630664536,
|
|
"learning_rate": 1.2804706213893195e-06,
|
|
"loss": 0.8697,
|
|
"mean_token_accuracy": 0.7759454995393753,
|
|
"num_tokens": 1100631986.0,
|
|
"step": 11000
|
|
},
|
|
{
|
|
"entropy": 0.900390625,
|
|
"epoch": 1.5047495387138659,
|
|
"grad_norm": 0.28501926678239226,
|
|
"learning_rate": 1.2769480061998029e-06,
|
|
"loss": 0.8999,
|
|
"mean_token_accuracy": 0.7710356414318085,
|
|
"num_tokens": 1101630371.0,
|
|
"step": 11010
|
|
},
|
|
{
|
|
"entropy": 0.834765625,
|
|
"epoch": 1.5061163124444747,
|
|
"grad_norm": 0.2789466236250796,
|
|
"learning_rate": 1.2734253910102862e-06,
|
|
"loss": 0.8318,
|
|
"mean_token_accuracy": 0.7830664843320847,
|
|
"num_tokens": 1102685721.0,
|
|
"step": 11020
|
|
},
|
|
{
|
|
"entropy": 0.8623046875,
|
|
"epoch": 1.5074830861750836,
|
|
"grad_norm": 0.2884337799841642,
|
|
"learning_rate": 1.2699027758207693e-06,
|
|
"loss": 0.8598,
|
|
"mean_token_accuracy": 0.7770796060562134,
|
|
"num_tokens": 1103679241.0,
|
|
"step": 11030
|
|
},
|
|
{
|
|
"entropy": 0.8947265625,
|
|
"epoch": 1.5088498599056925,
|
|
"grad_norm": 0.30093412087818056,
|
|
"learning_rate": 1.2663801606312529e-06,
|
|
"loss": 0.8879,
|
|
"mean_token_accuracy": 0.7692087441682816,
|
|
"num_tokens": 1104760350.0,
|
|
"step": 11040
|
|
},
|
|
{
|
|
"entropy": 0.8384765625,
|
|
"epoch": 1.5102166336363014,
|
|
"grad_norm": 0.2910906218143732,
|
|
"learning_rate": 1.262857545441736e-06,
|
|
"loss": 0.8328,
|
|
"mean_token_accuracy": 0.7831022411584854,
|
|
"num_tokens": 1105710063.0,
|
|
"step": 11050
|
|
},
|
|
{
|
|
"entropy": 0.87265625,
|
|
"epoch": 1.5115834073669103,
|
|
"grad_norm": 0.26714630447226956,
|
|
"learning_rate": 1.2593349302522193e-06,
|
|
"loss": 0.8694,
|
|
"mean_token_accuracy": 0.7731814622879029,
|
|
"num_tokens": 1106711978.0,
|
|
"step": 11060
|
|
},
|
|
{
|
|
"entropy": 0.8734375,
|
|
"epoch": 1.5129501810975192,
|
|
"grad_norm": 0.29973971058550014,
|
|
"learning_rate": 1.2558123150627027e-06,
|
|
"loss": 0.8745,
|
|
"mean_token_accuracy": 0.7750522345304489,
|
|
"num_tokens": 1107727919.0,
|
|
"step": 11070
|
|
},
|
|
{
|
|
"entropy": 0.901953125,
|
|
"epoch": 1.5143169548281281,
|
|
"grad_norm": 0.33113399783344166,
|
|
"learning_rate": 1.252289699873186e-06,
|
|
"loss": 0.9081,
|
|
"mean_token_accuracy": 0.7666845381259918,
|
|
"num_tokens": 1108733663.0,
|
|
"step": 11080
|
|
},
|
|
{
|
|
"entropy": 0.8955078125,
|
|
"epoch": 1.515683728558737,
|
|
"grad_norm": 0.30460773335216423,
|
|
"learning_rate": 1.2487670846836693e-06,
|
|
"loss": 0.8892,
|
|
"mean_token_accuracy": 0.7701234757900238,
|
|
"num_tokens": 1109745998.0,
|
|
"step": 11090
|
|
},
|
|
{
|
|
"entropy": 0.933984375,
|
|
"epoch": 1.517050502289346,
|
|
"grad_norm": 0.3347373291406093,
|
|
"learning_rate": 1.2452444694941527e-06,
|
|
"loss": 0.9344,
|
|
"mean_token_accuracy": 0.7606715589761734,
|
|
"num_tokens": 1110716442.0,
|
|
"step": 11100
|
|
},
|
|
{
|
|
"entropy": 0.890625,
|
|
"epoch": 1.5184172760199548,
|
|
"grad_norm": 0.2588791064962792,
|
|
"learning_rate": 1.2417218543046358e-06,
|
|
"loss": 0.8789,
|
|
"mean_token_accuracy": 0.7750590920448304,
|
|
"num_tokens": 1111713189.0,
|
|
"step": 11110
|
|
},
|
|
{
|
|
"entropy": 0.86484375,
|
|
"epoch": 1.5197840497505637,
|
|
"grad_norm": 0.2705598181620554,
|
|
"learning_rate": 1.2381992391151191e-06,
|
|
"loss": 0.8627,
|
|
"mean_token_accuracy": 0.7782370388507843,
|
|
"num_tokens": 1112704038.0,
|
|
"step": 11120
|
|
},
|
|
{
|
|
"entropy": 0.8701171875,
|
|
"epoch": 1.5211508234811726,
|
|
"grad_norm": 0.31220100393743544,
|
|
"learning_rate": 1.2346766239256025e-06,
|
|
"loss": 0.8811,
|
|
"mean_token_accuracy": 0.772601455450058,
|
|
"num_tokens": 1113734181.0,
|
|
"step": 11130
|
|
},
|
|
{
|
|
"entropy": 0.848046875,
|
|
"epoch": 1.5225175972117815,
|
|
"grad_norm": 0.2814714988147779,
|
|
"learning_rate": 1.2311540087360858e-06,
|
|
"loss": 0.8481,
|
|
"mean_token_accuracy": 0.7787964075803757,
|
|
"num_tokens": 1114781257.0,
|
|
"step": 11140
|
|
},
|
|
{
|
|
"entropy": 0.8533203125,
|
|
"epoch": 1.5238843709423904,
|
|
"grad_norm": 0.31209234366418287,
|
|
"learning_rate": 1.2276313935465691e-06,
|
|
"loss": 0.835,
|
|
"mean_token_accuracy": 0.7822346270084382,
|
|
"num_tokens": 1115775580.0,
|
|
"step": 11150
|
|
},
|
|
{
|
|
"entropy": 0.836328125,
|
|
"epoch": 1.5252511446729993,
|
|
"grad_norm": 0.3024918498224262,
|
|
"learning_rate": 1.2241087783570525e-06,
|
|
"loss": 0.8334,
|
|
"mean_token_accuracy": 0.7822325885295868,
|
|
"num_tokens": 1116778443.0,
|
|
"step": 11160
|
|
},
|
|
{
|
|
"entropy": 0.8732421875,
|
|
"epoch": 1.5266179184036082,
|
|
"grad_norm": 0.26398450244921357,
|
|
"learning_rate": 1.2205861631675358e-06,
|
|
"loss": 0.8739,
|
|
"mean_token_accuracy": 0.7742313861846923,
|
|
"num_tokens": 1117785103.0,
|
|
"step": 11170
|
|
},
|
|
{
|
|
"entropy": 0.8890625,
|
|
"epoch": 1.527984692134217,
|
|
"grad_norm": 0.3267016008339695,
|
|
"learning_rate": 1.217063547978019e-06,
|
|
"loss": 0.8841,
|
|
"mean_token_accuracy": 0.7711778312921524,
|
|
"num_tokens": 1118765464.0,
|
|
"step": 11180
|
|
},
|
|
{
|
|
"entropy": 0.8599609375,
|
|
"epoch": 1.529351465864826,
|
|
"grad_norm": 0.2816162462072086,
|
|
"learning_rate": 1.2135409327885023e-06,
|
|
"loss": 0.8639,
|
|
"mean_token_accuracy": 0.7760516703128815,
|
|
"num_tokens": 1119764058.0,
|
|
"step": 11190
|
|
},
|
|
{
|
|
"entropy": 0.85078125,
|
|
"epoch": 1.5307182395954348,
|
|
"grad_norm": 0.2933267038816799,
|
|
"learning_rate": 1.2100183175989856e-06,
|
|
"loss": 0.8426,
|
|
"mean_token_accuracy": 0.7811099171638489,
|
|
"num_tokens": 1120738890.0,
|
|
"step": 11200
|
|
},
|
|
{
|
|
"entropy": 0.83125,
|
|
"epoch": 1.5320850133260437,
|
|
"grad_norm": 0.28473499930350976,
|
|
"learning_rate": 1.206495702409469e-06,
|
|
"loss": 0.8271,
|
|
"mean_token_accuracy": 0.782834866642952,
|
|
"num_tokens": 1121719357.0,
|
|
"step": 11210
|
|
},
|
|
{
|
|
"entropy": 0.820703125,
|
|
"epoch": 1.5334517870566526,
|
|
"grad_norm": 0.30107604938817395,
|
|
"learning_rate": 1.202973087219952e-06,
|
|
"loss": 0.8127,
|
|
"mean_token_accuracy": 0.785349115729332,
|
|
"num_tokens": 1122718939.0,
|
|
"step": 11220
|
|
},
|
|
{
|
|
"entropy": 0.846875,
|
|
"epoch": 1.5348185607872615,
|
|
"grad_norm": 0.29223258711264116,
|
|
"learning_rate": 1.1994504720304354e-06,
|
|
"loss": 0.8553,
|
|
"mean_token_accuracy": 0.7791976749897003,
|
|
"num_tokens": 1123745255.0,
|
|
"step": 11230
|
|
},
|
|
{
|
|
"entropy": 0.8384765625,
|
|
"epoch": 1.5361853345178704,
|
|
"grad_norm": 0.3336996388434604,
|
|
"learning_rate": 1.1959278568409187e-06,
|
|
"loss": 0.8329,
|
|
"mean_token_accuracy": 0.7810836613178254,
|
|
"num_tokens": 1124780242.0,
|
|
"step": 11240
|
|
},
|
|
{
|
|
"entropy": 0.8384765625,
|
|
"epoch": 1.5375521082484793,
|
|
"grad_norm": 0.28984821114084985,
|
|
"learning_rate": 1.192405241651402e-06,
|
|
"loss": 0.8355,
|
|
"mean_token_accuracy": 0.7808184146881103,
|
|
"num_tokens": 1125787926.0,
|
|
"step": 11250
|
|
},
|
|
{
|
|
"entropy": 0.8669921875,
|
|
"epoch": 1.5389188819790882,
|
|
"grad_norm": 0.3322236215803577,
|
|
"learning_rate": 1.1888826264618854e-06,
|
|
"loss": 0.867,
|
|
"mean_token_accuracy": 0.7744541555643082,
|
|
"num_tokens": 1126778760.0,
|
|
"step": 11260
|
|
},
|
|
{
|
|
"entropy": 0.8537109375,
|
|
"epoch": 1.540285655709697,
|
|
"grad_norm": 0.25635561018137204,
|
|
"learning_rate": 1.1853600112723687e-06,
|
|
"loss": 0.8468,
|
|
"mean_token_accuracy": 0.7795037209987641,
|
|
"num_tokens": 1127803831.0,
|
|
"step": 11270
|
|
},
|
|
{
|
|
"entropy": 0.8435546875,
|
|
"epoch": 1.541652429440306,
|
|
"grad_norm": 0.32931041755481244,
|
|
"learning_rate": 1.181837396082852e-06,
|
|
"loss": 0.8378,
|
|
"mean_token_accuracy": 0.7805411249399186,
|
|
"num_tokens": 1128824935.0,
|
|
"step": 11280
|
|
},
|
|
{
|
|
"entropy": 0.8498046875,
|
|
"epoch": 1.543019203170915,
|
|
"grad_norm": 0.28823769743893835,
|
|
"learning_rate": 1.1783147808933352e-06,
|
|
"loss": 0.8441,
|
|
"mean_token_accuracy": 0.7798783004283905,
|
|
"num_tokens": 1129830752.0,
|
|
"step": 11290
|
|
},
|
|
{
|
|
"entropy": 0.9267578125,
|
|
"epoch": 1.5443859769015238,
|
|
"grad_norm": 0.28409649867419046,
|
|
"learning_rate": 1.1747921657038185e-06,
|
|
"loss": 0.9298,
|
|
"mean_token_accuracy": 0.7632709085941315,
|
|
"num_tokens": 1130826790.0,
|
|
"step": 11300
|
|
},
|
|
{
|
|
"entropy": 0.8421875,
|
|
"epoch": 1.5457527506321327,
|
|
"grad_norm": 0.2878058130030118,
|
|
"learning_rate": 1.1712695505143019e-06,
|
|
"loss": 0.8344,
|
|
"mean_token_accuracy": 0.7802106499671936,
|
|
"num_tokens": 1131808066.0,
|
|
"step": 11310
|
|
},
|
|
{
|
|
"entropy": 0.87890625,
|
|
"epoch": 1.5471195243627416,
|
|
"grad_norm": 0.28409780413795677,
|
|
"learning_rate": 1.1677469353247852e-06,
|
|
"loss": 0.8665,
|
|
"mean_token_accuracy": 0.7766708254814148,
|
|
"num_tokens": 1132754963.0,
|
|
"step": 11320
|
|
},
|
|
{
|
|
"entropy": 0.8908203125,
|
|
"epoch": 1.5484862980933507,
|
|
"grad_norm": 0.2870171622190749,
|
|
"learning_rate": 1.1642243201352685e-06,
|
|
"loss": 0.8912,
|
|
"mean_token_accuracy": 0.769545441865921,
|
|
"num_tokens": 1133788644.0,
|
|
"step": 11330
|
|
},
|
|
{
|
|
"entropy": 0.8498046875,
|
|
"epoch": 1.5498530718239596,
|
|
"grad_norm": 0.3109283360939281,
|
|
"learning_rate": 1.1607017049457519e-06,
|
|
"loss": 0.8459,
|
|
"mean_token_accuracy": 0.7787738114595413,
|
|
"num_tokens": 1134812730.0,
|
|
"step": 11340
|
|
},
|
|
{
|
|
"entropy": 0.8923828125,
|
|
"epoch": 1.5512198455545685,
|
|
"grad_norm": 0.3006627722254283,
|
|
"learning_rate": 1.1571790897562352e-06,
|
|
"loss": 0.8913,
|
|
"mean_token_accuracy": 0.7713357836008072,
|
|
"num_tokens": 1135773833.0,
|
|
"step": 11350
|
|
},
|
|
{
|
|
"entropy": 0.8521484375,
|
|
"epoch": 1.5525866192851774,
|
|
"grad_norm": 0.2899613866349312,
|
|
"learning_rate": 1.1536564745667183e-06,
|
|
"loss": 0.8428,
|
|
"mean_token_accuracy": 0.7804517447948456,
|
|
"num_tokens": 1136755524.0,
|
|
"step": 11360
|
|
},
|
|
{
|
|
"entropy": 0.8525390625,
|
|
"epoch": 1.5539533930157863,
|
|
"grad_norm": 0.2900321741835512,
|
|
"learning_rate": 1.1501338593772017e-06,
|
|
"loss": 0.8459,
|
|
"mean_token_accuracy": 0.7799200624227524,
|
|
"num_tokens": 1137730738.0,
|
|
"step": 11370
|
|
},
|
|
{
|
|
"entropy": 0.8794921875,
|
|
"epoch": 1.5553201667463952,
|
|
"grad_norm": 0.30763715646105555,
|
|
"learning_rate": 1.146611244187685e-06,
|
|
"loss": 0.8743,
|
|
"mean_token_accuracy": 0.776023006439209,
|
|
"num_tokens": 1138761002.0,
|
|
"step": 11380
|
|
},
|
|
{
|
|
"entropy": 0.880078125,
|
|
"epoch": 1.556686940477004,
|
|
"grad_norm": 0.2833992910636164,
|
|
"learning_rate": 1.1430886289981683e-06,
|
|
"loss": 0.8789,
|
|
"mean_token_accuracy": 0.7701349139213562,
|
|
"num_tokens": 1139762019.0,
|
|
"step": 11390
|
|
},
|
|
{
|
|
"entropy": 0.825,
|
|
"epoch": 1.558053714207613,
|
|
"grad_norm": 0.3137979537478371,
|
|
"learning_rate": 1.1395660138086517e-06,
|
|
"loss": 0.8158,
|
|
"mean_token_accuracy": 0.7859611570835113,
|
|
"num_tokens": 1140702582.0,
|
|
"step": 11400
|
|
},
|
|
{
|
|
"entropy": 0.8400390625,
|
|
"epoch": 1.5594204879382219,
|
|
"grad_norm": 0.3074566653805055,
|
|
"learning_rate": 1.136043398619135e-06,
|
|
"loss": 0.8372,
|
|
"mean_token_accuracy": 0.7826174288988114,
|
|
"num_tokens": 1141683207.0,
|
|
"step": 11410
|
|
},
|
|
{
|
|
"entropy": 0.911328125,
|
|
"epoch": 1.5607872616688307,
|
|
"grad_norm": 0.31092984612294183,
|
|
"learning_rate": 1.1325207834296184e-06,
|
|
"loss": 0.8997,
|
|
"mean_token_accuracy": 0.7682123512029648,
|
|
"num_tokens": 1142702501.0,
|
|
"step": 11420
|
|
},
|
|
{
|
|
"entropy": 0.833203125,
|
|
"epoch": 1.5621540353994396,
|
|
"grad_norm": 0.2786736019101562,
|
|
"learning_rate": 1.1289981682401017e-06,
|
|
"loss": 0.8264,
|
|
"mean_token_accuracy": 0.7829974681138993,
|
|
"num_tokens": 1143668224.0,
|
|
"step": 11430
|
|
},
|
|
{
|
|
"entropy": 0.8576171875,
|
|
"epoch": 1.5635208091300485,
|
|
"grad_norm": 0.2599497331094021,
|
|
"learning_rate": 1.1254755530505848e-06,
|
|
"loss": 0.8489,
|
|
"mean_token_accuracy": 0.7782787352800369,
|
|
"num_tokens": 1144665372.0,
|
|
"step": 11440
|
|
},
|
|
{
|
|
"entropy": 0.8884765625,
|
|
"epoch": 1.5648875828606574,
|
|
"grad_norm": 0.3148767766144308,
|
|
"learning_rate": 1.1219529378610681e-06,
|
|
"loss": 0.8825,
|
|
"mean_token_accuracy": 0.7710226088762283,
|
|
"num_tokens": 1145660388.0,
|
|
"step": 11450
|
|
},
|
|
{
|
|
"entropy": 0.8830078125,
|
|
"epoch": 1.5662543565912663,
|
|
"grad_norm": 0.2712776410883267,
|
|
"learning_rate": 1.1184303226715515e-06,
|
|
"loss": 0.8798,
|
|
"mean_token_accuracy": 0.7731872975826264,
|
|
"num_tokens": 1146681677.0,
|
|
"step": 11460
|
|
},
|
|
{
|
|
"entropy": 0.8869140625,
|
|
"epoch": 1.5676211303218752,
|
|
"grad_norm": 0.3112699232648531,
|
|
"learning_rate": 1.1149077074820346e-06,
|
|
"loss": 0.8795,
|
|
"mean_token_accuracy": 0.7727049082517624,
|
|
"num_tokens": 1147684038.0,
|
|
"step": 11470
|
|
},
|
|
{
|
|
"entropy": 0.915234375,
|
|
"epoch": 1.5689879040524841,
|
|
"grad_norm": 0.3241984168005189,
|
|
"learning_rate": 1.111385092292518e-06,
|
|
"loss": 0.9103,
|
|
"mean_token_accuracy": 0.7683268517255784,
|
|
"num_tokens": 1148667886.0,
|
|
"step": 11480
|
|
},
|
|
{
|
|
"entropy": 0.8642578125,
|
|
"epoch": 1.570354677783093,
|
|
"grad_norm": 0.2593983777827075,
|
|
"learning_rate": 1.1078624771030013e-06,
|
|
"loss": 0.8653,
|
|
"mean_token_accuracy": 0.7760563850402832,
|
|
"num_tokens": 1149668552.0,
|
|
"step": 11490
|
|
},
|
|
{
|
|
"entropy": 0.8779296875,
|
|
"epoch": 1.571721451513702,
|
|
"grad_norm": 0.27484239102119307,
|
|
"learning_rate": 1.1043398619134846e-06,
|
|
"loss": 0.8726,
|
|
"mean_token_accuracy": 0.7719525337219239,
|
|
"num_tokens": 1150708766.0,
|
|
"step": 11500
|
|
},
|
|
{
|
|
"entropy": 0.8849609375,
|
|
"epoch": 1.5730882252443108,
|
|
"grad_norm": 0.29976024503625787,
|
|
"learning_rate": 1.100817246723968e-06,
|
|
"loss": 0.8796,
|
|
"mean_token_accuracy": 0.7723476469516755,
|
|
"num_tokens": 1151695889.0,
|
|
"step": 11510
|
|
},
|
|
{
|
|
"entropy": 0.884375,
|
|
"epoch": 1.5744549989749197,
|
|
"grad_norm": 0.2495767386550892,
|
|
"learning_rate": 1.0972946315344513e-06,
|
|
"loss": 0.8869,
|
|
"mean_token_accuracy": 0.772763466835022,
|
|
"num_tokens": 1152737798.0,
|
|
"step": 11520
|
|
},
|
|
{
|
|
"entropy": 0.8517578125,
|
|
"epoch": 1.5758217727055286,
|
|
"grad_norm": 0.28622049362701774,
|
|
"learning_rate": 1.0937720163449346e-06,
|
|
"loss": 0.8498,
|
|
"mean_token_accuracy": 0.7809607744216919,
|
|
"num_tokens": 1153721870.0,
|
|
"step": 11530
|
|
},
|
|
{
|
|
"entropy": 0.865234375,
|
|
"epoch": 1.5771885464361375,
|
|
"grad_norm": 0.28156950518744017,
|
|
"learning_rate": 1.0902494011554177e-06,
|
|
"loss": 0.8628,
|
|
"mean_token_accuracy": 0.7752205312252045,
|
|
"num_tokens": 1154734151.0,
|
|
"step": 11540
|
|
},
|
|
{
|
|
"entropy": 0.8421875,
|
|
"epoch": 1.5785553201667464,
|
|
"grad_norm": 0.28891648617299315,
|
|
"learning_rate": 1.086726785965901e-06,
|
|
"loss": 0.8445,
|
|
"mean_token_accuracy": 0.779954394698143,
|
|
"num_tokens": 1155783238.0,
|
|
"step": 11550
|
|
},
|
|
{
|
|
"entropy": 0.9037109375,
|
|
"epoch": 1.5799220938973553,
|
|
"grad_norm": 0.30360163246287336,
|
|
"learning_rate": 1.0832041707763844e-06,
|
|
"loss": 0.8993,
|
|
"mean_token_accuracy": 0.7698518693447113,
|
|
"num_tokens": 1156782441.0,
|
|
"step": 11560
|
|
},
|
|
{
|
|
"entropy": 0.898828125,
|
|
"epoch": 1.5812888676279642,
|
|
"grad_norm": 0.29862724788835926,
|
|
"learning_rate": 1.0796815555868678e-06,
|
|
"loss": 0.9011,
|
|
"mean_token_accuracy": 0.7717364966869354,
|
|
"num_tokens": 1157846256.0,
|
|
"step": 11570
|
|
},
|
|
{
|
|
"entropy": 0.9046875,
|
|
"epoch": 1.582655641358573,
|
|
"grad_norm": 0.257504776244002,
|
|
"learning_rate": 1.076158940397351e-06,
|
|
"loss": 0.9022,
|
|
"mean_token_accuracy": 0.7683009624481201,
|
|
"num_tokens": 1158851215.0,
|
|
"step": 11580
|
|
},
|
|
{
|
|
"entropy": 0.88046875,
|
|
"epoch": 1.584022415089182,
|
|
"grad_norm": 0.2765814776043368,
|
|
"learning_rate": 1.0726363252078344e-06,
|
|
"loss": 0.8816,
|
|
"mean_token_accuracy": 0.7731819421052932,
|
|
"num_tokens": 1159807777.0,
|
|
"step": 11590
|
|
},
|
|
{
|
|
"entropy": 0.8279296875,
|
|
"epoch": 1.5853891888197909,
|
|
"grad_norm": 0.2611522896429713,
|
|
"learning_rate": 1.0691137100183178e-06,
|
|
"loss": 0.819,
|
|
"mean_token_accuracy": 0.7861458510160446,
|
|
"num_tokens": 1160799680.0,
|
|
"step": 11600
|
|
},
|
|
{
|
|
"entropy": 0.8341796875,
|
|
"epoch": 1.5867559625503997,
|
|
"grad_norm": 0.2939340693028046,
|
|
"learning_rate": 1.065591094828801e-06,
|
|
"loss": 0.8349,
|
|
"mean_token_accuracy": 0.7818466395139694,
|
|
"num_tokens": 1161858193.0,
|
|
"step": 11610
|
|
},
|
|
{
|
|
"entropy": 0.8515625,
|
|
"epoch": 1.5881227362810086,
|
|
"grad_norm": 0.2844808992456468,
|
|
"learning_rate": 1.0620684796392842e-06,
|
|
"loss": 0.8434,
|
|
"mean_token_accuracy": 0.778670871257782,
|
|
"num_tokens": 1162797263.0,
|
|
"step": 11620
|
|
},
|
|
{
|
|
"entropy": 0.807421875,
|
|
"epoch": 1.5894895100116175,
|
|
"grad_norm": 0.2657501434188124,
|
|
"learning_rate": 1.0585458644497676e-06,
|
|
"loss": 0.8024,
|
|
"mean_token_accuracy": 0.7873925983905792,
|
|
"num_tokens": 1163764902.0,
|
|
"step": 11630
|
|
},
|
|
{
|
|
"entropy": 0.88984375,
|
|
"epoch": 1.5908562837422264,
|
|
"grad_norm": 0.3091897923381258,
|
|
"learning_rate": 1.0550232492602509e-06,
|
|
"loss": 0.8897,
|
|
"mean_token_accuracy": 0.7701803117990493,
|
|
"num_tokens": 1164750238.0,
|
|
"step": 11640
|
|
},
|
|
{
|
|
"entropy": 0.890234375,
|
|
"epoch": 1.5922230574728353,
|
|
"grad_norm": 0.2728311573825055,
|
|
"learning_rate": 1.0515006340707342e-06,
|
|
"loss": 0.881,
|
|
"mean_token_accuracy": 0.7742711663246155,
|
|
"num_tokens": 1165727780.0,
|
|
"step": 11650
|
|
},
|
|
{
|
|
"entropy": 0.832421875,
|
|
"epoch": 1.5935898312034442,
|
|
"grad_norm": 0.2981606117493079,
|
|
"learning_rate": 1.0479780188812176e-06,
|
|
"loss": 0.8329,
|
|
"mean_token_accuracy": 0.7823011040687561,
|
|
"num_tokens": 1166675519.0,
|
|
"step": 11660
|
|
},
|
|
{
|
|
"entropy": 0.8654296875,
|
|
"epoch": 1.5949566049340531,
|
|
"grad_norm": 0.30130842997763957,
|
|
"learning_rate": 1.0444554036917009e-06,
|
|
"loss": 0.8667,
|
|
"mean_token_accuracy": 0.7750510692596435,
|
|
"num_tokens": 1167685434.0,
|
|
"step": 11670
|
|
},
|
|
{
|
|
"entropy": 0.886328125,
|
|
"epoch": 1.5963233786646622,
|
|
"grad_norm": 0.27493563685162636,
|
|
"learning_rate": 1.0409327885021842e-06,
|
|
"loss": 0.8866,
|
|
"mean_token_accuracy": 0.7686771124601364,
|
|
"num_tokens": 1168705229.0,
|
|
"step": 11680
|
|
},
|
|
{
|
|
"entropy": 0.840625,
|
|
"epoch": 1.5976901523952711,
|
|
"grad_norm": 0.2789733492198398,
|
|
"learning_rate": 1.0374101733126674e-06,
|
|
"loss": 0.8455,
|
|
"mean_token_accuracy": 0.7818532347679138,
|
|
"num_tokens": 1169700686.0,
|
|
"step": 11690
|
|
},
|
|
{
|
|
"entropy": 0.825,
|
|
"epoch": 1.59905692612588,
|
|
"grad_norm": 0.32776334484518227,
|
|
"learning_rate": 1.0338875581231507e-06,
|
|
"loss": 0.8124,
|
|
"mean_token_accuracy": 0.7857388645410538,
|
|
"num_tokens": 1170639619.0,
|
|
"step": 11700
|
|
},
|
|
{
|
|
"entropy": 0.8857421875,
|
|
"epoch": 1.600423699856489,
|
|
"grad_norm": 0.2881398067718224,
|
|
"learning_rate": 1.030364942933634e-06,
|
|
"loss": 0.8842,
|
|
"mean_token_accuracy": 0.7693279027938843,
|
|
"num_tokens": 1171615484.0,
|
|
"step": 11710
|
|
},
|
|
{
|
|
"entropy": 0.8853515625,
|
|
"epoch": 1.6017904735870978,
|
|
"grad_norm": 0.2734297556292038,
|
|
"learning_rate": 1.0268423277441174e-06,
|
|
"loss": 0.8826,
|
|
"mean_token_accuracy": 0.7715979903936386,
|
|
"num_tokens": 1172578943.0,
|
|
"step": 11720
|
|
},
|
|
{
|
|
"entropy": 0.8724609375,
|
|
"epoch": 1.6031572473177067,
|
|
"grad_norm": 0.2907085352264695,
|
|
"learning_rate": 1.0233197125546005e-06,
|
|
"loss": 0.8618,
|
|
"mean_token_accuracy": 0.7772665202617646,
|
|
"num_tokens": 1173520988.0,
|
|
"step": 11730
|
|
},
|
|
{
|
|
"entropy": 0.8701171875,
|
|
"epoch": 1.6045240210483156,
|
|
"grad_norm": 0.29383103671609756,
|
|
"learning_rate": 1.0197970973650838e-06,
|
|
"loss": 0.8687,
|
|
"mean_token_accuracy": 0.7749045014381408,
|
|
"num_tokens": 1174523999.0,
|
|
"step": 11740
|
|
},
|
|
{
|
|
"entropy": 0.863671875,
|
|
"epoch": 1.6058907947789245,
|
|
"grad_norm": 0.29609485133607166,
|
|
"learning_rate": 1.0162744821755674e-06,
|
|
"loss": 0.866,
|
|
"mean_token_accuracy": 0.7764067351818085,
|
|
"num_tokens": 1175543448.0,
|
|
"step": 11750
|
|
},
|
|
{
|
|
"entropy": 0.8501953125,
|
|
"epoch": 1.6072575685095334,
|
|
"grad_norm": 0.30235767758376325,
|
|
"learning_rate": 1.0127518669860505e-06,
|
|
"loss": 0.8438,
|
|
"mean_token_accuracy": 0.780577152967453,
|
|
"num_tokens": 1176532582.0,
|
|
"step": 11760
|
|
},
|
|
{
|
|
"entropy": 0.90234375,
|
|
"epoch": 1.6086243422401423,
|
|
"grad_norm": 0.3232096336587165,
|
|
"learning_rate": 1.0092292517965338e-06,
|
|
"loss": 0.8969,
|
|
"mean_token_accuracy": 0.7707766234874726,
|
|
"num_tokens": 1177564469.0,
|
|
"step": 11770
|
|
},
|
|
{
|
|
"entropy": 0.873046875,
|
|
"epoch": 1.6099911159707512,
|
|
"grad_norm": 0.2888200453017122,
|
|
"learning_rate": 1.0057066366070172e-06,
|
|
"loss": 0.8691,
|
|
"mean_token_accuracy": 0.77649105489254,
|
|
"num_tokens": 1178563420.0,
|
|
"step": 11780
|
|
},
|
|
{
|
|
"entropy": 0.8400390625,
|
|
"epoch": 1.61135788970136,
|
|
"grad_norm": 0.27069821589351356,
|
|
"learning_rate": 1.0021840214175005e-06,
|
|
"loss": 0.8318,
|
|
"mean_token_accuracy": 0.7833161413669586,
|
|
"num_tokens": 1179550273.0,
|
|
"step": 11790
|
|
},
|
|
{
|
|
"entropy": 0.87109375,
|
|
"epoch": 1.612724663431969,
|
|
"grad_norm": 0.29195656431143163,
|
|
"learning_rate": 9.986614062279836e-07,
|
|
"loss": 0.8742,
|
|
"mean_token_accuracy": 0.7750835359096527,
|
|
"num_tokens": 1180561479.0,
|
|
"step": 11800
|
|
},
|
|
{
|
|
"entropy": 0.8255859375,
|
|
"epoch": 1.6140914371625779,
|
|
"grad_norm": 0.2835611521468528,
|
|
"learning_rate": 9.95138791038467e-07,
|
|
"loss": 0.8191,
|
|
"mean_token_accuracy": 0.7856638163328171,
|
|
"num_tokens": 1181568038.0,
|
|
"step": 11810
|
|
},
|
|
{
|
|
"entropy": 0.8484375,
|
|
"epoch": 1.6154582108931868,
|
|
"grad_norm": 0.29588055101479027,
|
|
"learning_rate": 9.916161758489503e-07,
|
|
"loss": 0.8361,
|
|
"mean_token_accuracy": 0.7833306968212128,
|
|
"num_tokens": 1182566485.0,
|
|
"step": 11820
|
|
},
|
|
{
|
|
"entropy": 0.84765625,
|
|
"epoch": 1.6168249846237956,
|
|
"grad_norm": 0.293573437824845,
|
|
"learning_rate": 9.880935606594336e-07,
|
|
"loss": 0.8506,
|
|
"mean_token_accuracy": 0.7789759337902069,
|
|
"num_tokens": 1183634690.0,
|
|
"step": 11830
|
|
},
|
|
{
|
|
"entropy": 0.85703125,
|
|
"epoch": 1.6181917583544045,
|
|
"grad_norm": 0.26079687214372577,
|
|
"learning_rate": 9.84570945469917e-07,
|
|
"loss": 0.864,
|
|
"mean_token_accuracy": 0.7771445631980896,
|
|
"num_tokens": 1184633173.0,
|
|
"step": 11840
|
|
},
|
|
{
|
|
"entropy": 0.87265625,
|
|
"epoch": 1.6195585320850134,
|
|
"grad_norm": 0.29874312320427604,
|
|
"learning_rate": 9.810483302804003e-07,
|
|
"loss": 0.8587,
|
|
"mean_token_accuracy": 0.7762014895677567,
|
|
"num_tokens": 1185632867.0,
|
|
"step": 11850
|
|
},
|
|
{
|
|
"entropy": 0.8638671875,
|
|
"epoch": 1.6209253058156223,
|
|
"grad_norm": 0.3234909225851195,
|
|
"learning_rate": 9.775257150908836e-07,
|
|
"loss": 0.853,
|
|
"mean_token_accuracy": 0.7774585545063019,
|
|
"num_tokens": 1186611170.0,
|
|
"step": 11860
|
|
},
|
|
{
|
|
"entropy": 0.858203125,
|
|
"epoch": 1.6222920795462312,
|
|
"grad_norm": 0.27988802823122666,
|
|
"learning_rate": 9.740030999013668e-07,
|
|
"loss": 0.8599,
|
|
"mean_token_accuracy": 0.7779110550880433,
|
|
"num_tokens": 1187586294.0,
|
|
"step": 11870
|
|
},
|
|
{
|
|
"entropy": 0.843359375,
|
|
"epoch": 1.6236588532768401,
|
|
"grad_norm": 0.28059326851636046,
|
|
"learning_rate": 9.7048048471185e-07,
|
|
"loss": 0.8332,
|
|
"mean_token_accuracy": 0.7801219522953033,
|
|
"num_tokens": 1188555265.0,
|
|
"step": 11880
|
|
},
|
|
{
|
|
"entropy": 0.886328125,
|
|
"epoch": 1.625025627007449,
|
|
"grad_norm": 0.2566122620851746,
|
|
"learning_rate": 9.669578695223334e-07,
|
|
"loss": 0.8846,
|
|
"mean_token_accuracy": 0.7714350163936615,
|
|
"num_tokens": 1189573652.0,
|
|
"step": 11890
|
|
},
|
|
{
|
|
"entropy": 0.8708984375,
|
|
"epoch": 1.626392400738058,
|
|
"grad_norm": 0.278372327383698,
|
|
"learning_rate": 9.634352543328168e-07,
|
|
"loss": 0.865,
|
|
"mean_token_accuracy": 0.7742911785840988,
|
|
"num_tokens": 1190541519.0,
|
|
"step": 11900
|
|
},
|
|
{
|
|
"entropy": 0.865625,
|
|
"epoch": 1.6277591744686668,
|
|
"grad_norm": 0.3026119006108552,
|
|
"learning_rate": 9.599126391433e-07,
|
|
"loss": 0.8624,
|
|
"mean_token_accuracy": 0.7760124951601028,
|
|
"num_tokens": 1191547906.0,
|
|
"step": 11910
|
|
},
|
|
{
|
|
"entropy": 0.8861328125,
|
|
"epoch": 1.6291259481992757,
|
|
"grad_norm": 0.2849391885088169,
|
|
"learning_rate": 9.563900239537834e-07,
|
|
"loss": 0.8802,
|
|
"mean_token_accuracy": 0.7726823687553406,
|
|
"num_tokens": 1192539483.0,
|
|
"step": 11920
|
|
},
|
|
{
|
|
"entropy": 0.8419921875,
|
|
"epoch": 1.6304927219298846,
|
|
"grad_norm": 0.3007375082457199,
|
|
"learning_rate": 9.528674087642667e-07,
|
|
"loss": 0.8476,
|
|
"mean_token_accuracy": 0.7790239781141282,
|
|
"num_tokens": 1193562218.0,
|
|
"step": 11930
|
|
},
|
|
{
|
|
"entropy": 0.8349609375,
|
|
"epoch": 1.6318594956604935,
|
|
"grad_norm": 0.3967277318392668,
|
|
"learning_rate": 9.4934479357475e-07,
|
|
"loss": 0.8219,
|
|
"mean_token_accuracy": 0.7849721938371659,
|
|
"num_tokens": 1194576671.0,
|
|
"step": 11940
|
|
},
|
|
{
|
|
"entropy": 0.8833984375,
|
|
"epoch": 1.6332262693911024,
|
|
"grad_norm": 0.269190141537747,
|
|
"learning_rate": 9.458221783852332e-07,
|
|
"loss": 0.8741,
|
|
"mean_token_accuracy": 0.7728091567754746,
|
|
"num_tokens": 1195622767.0,
|
|
"step": 11950
|
|
},
|
|
{
|
|
"entropy": 0.8578125,
|
|
"epoch": 1.6345930431217113,
|
|
"grad_norm": 0.30907282929996494,
|
|
"learning_rate": 9.422995631957166e-07,
|
|
"loss": 0.8555,
|
|
"mean_token_accuracy": 0.7779740929603577,
|
|
"num_tokens": 1196573023.0,
|
|
"step": 11960
|
|
},
|
|
{
|
|
"entropy": 0.8892578125,
|
|
"epoch": 1.6359598168523202,
|
|
"grad_norm": 0.33127401647816335,
|
|
"learning_rate": 9.387769480061999e-07,
|
|
"loss": 0.883,
|
|
"mean_token_accuracy": 0.772578302025795,
|
|
"num_tokens": 1197559281.0,
|
|
"step": 11970
|
|
},
|
|
{
|
|
"entropy": 0.860546875,
|
|
"epoch": 1.637326590582929,
|
|
"grad_norm": 0.30297107461820383,
|
|
"learning_rate": 9.352543328166831e-07,
|
|
"loss": 0.8525,
|
|
"mean_token_accuracy": 0.777357217669487,
|
|
"num_tokens": 1198597388.0,
|
|
"step": 11980
|
|
},
|
|
{
|
|
"entropy": 0.8701171875,
|
|
"epoch": 1.638693364313538,
|
|
"grad_norm": 0.26902547092764767,
|
|
"learning_rate": 9.317317176271665e-07,
|
|
"loss": 0.87,
|
|
"mean_token_accuracy": 0.7740891069173813,
|
|
"num_tokens": 1199552097.0,
|
|
"step": 11990
|
|
},
|
|
{
|
|
"entropy": 0.89609375,
|
|
"epoch": 1.6400601380441469,
|
|
"grad_norm": 0.26798744096626975,
|
|
"learning_rate": 9.282091024376498e-07,
|
|
"loss": 0.8982,
|
|
"mean_token_accuracy": 0.7679164946079254,
|
|
"num_tokens": 1200618875.0,
|
|
"step": 12000
|
|
},
|
|
{
|
|
"entropy": 0.9033203125,
|
|
"epoch": 1.6414269117747557,
|
|
"grad_norm": 0.32385101444018416,
|
|
"learning_rate": 9.246864872481331e-07,
|
|
"loss": 0.8978,
|
|
"mean_token_accuracy": 0.7665571004152298,
|
|
"num_tokens": 1201586598.0,
|
|
"step": 12010
|
|
},
|
|
{
|
|
"entropy": 0.8498046875,
|
|
"epoch": 1.6427936855053646,
|
|
"grad_norm": 0.3445829840912683,
|
|
"learning_rate": 9.211638720586164e-07,
|
|
"loss": 0.8443,
|
|
"mean_token_accuracy": 0.7787088453769684,
|
|
"num_tokens": 1202539920.0,
|
|
"step": 12020
|
|
},
|
|
{
|
|
"entropy": 0.8537109375,
|
|
"epoch": 1.6441604592359735,
|
|
"grad_norm": 0.3055940384869851,
|
|
"learning_rate": 9.176412568690997e-07,
|
|
"loss": 0.8429,
|
|
"mean_token_accuracy": 0.7801574826240539,
|
|
"num_tokens": 1203531737.0,
|
|
"step": 12030
|
|
},
|
|
{
|
|
"entropy": 0.8814453125,
|
|
"epoch": 1.6455272329665824,
|
|
"grad_norm": 0.3184161839814361,
|
|
"learning_rate": 9.14118641679583e-07,
|
|
"loss": 0.8727,
|
|
"mean_token_accuracy": 0.7715849578380585,
|
|
"num_tokens": 1204557940.0,
|
|
"step": 12040
|
|
},
|
|
{
|
|
"entropy": 0.84609375,
|
|
"epoch": 1.6468940066971913,
|
|
"grad_norm": 0.2956671772236001,
|
|
"learning_rate": 9.105960264900663e-07,
|
|
"loss": 0.8469,
|
|
"mean_token_accuracy": 0.7789913535118103,
|
|
"num_tokens": 1205503711.0,
|
|
"step": 12050
|
|
},
|
|
{
|
|
"entropy": 0.86015625,
|
|
"epoch": 1.6482607804278002,
|
|
"grad_norm": 0.29322091845561943,
|
|
"learning_rate": 9.070734113005496e-07,
|
|
"loss": 0.8662,
|
|
"mean_token_accuracy": 0.7774302005767822,
|
|
"num_tokens": 1206546593.0,
|
|
"step": 12060
|
|
},
|
|
{
|
|
"entropy": 0.9271484375,
|
|
"epoch": 1.6496275541584091,
|
|
"grad_norm": 0.28225194792237424,
|
|
"learning_rate": 9.035507961110329e-07,
|
|
"loss": 0.9191,
|
|
"mean_token_accuracy": 0.7643861204385758,
|
|
"num_tokens": 1207524826.0,
|
|
"step": 12070
|
|
},
|
|
{
|
|
"entropy": 0.8484375,
|
|
"epoch": 1.650994327889018,
|
|
"grad_norm": 0.2729296396370077,
|
|
"learning_rate": 9.000281809215163e-07,
|
|
"loss": 0.8467,
|
|
"mean_token_accuracy": 0.779317033290863,
|
|
"num_tokens": 1208517371.0,
|
|
"step": 12080
|
|
},
|
|
{
|
|
"entropy": 0.8267578125,
|
|
"epoch": 1.652361101619627,
|
|
"grad_norm": 0.29034265476279375,
|
|
"learning_rate": 8.965055657319995e-07,
|
|
"loss": 0.8271,
|
|
"mean_token_accuracy": 0.7843029797077179,
|
|
"num_tokens": 1209515477.0,
|
|
"step": 12090
|
|
},
|
|
{
|
|
"entropy": 0.8490234375,
|
|
"epoch": 1.6537278753502358,
|
|
"grad_norm": 0.2580749658385215,
|
|
"learning_rate": 8.929829505424828e-07,
|
|
"loss": 0.8425,
|
|
"mean_token_accuracy": 0.780238863825798,
|
|
"num_tokens": 1210518784.0,
|
|
"step": 12100
|
|
},
|
|
{
|
|
"entropy": 0.8462890625,
|
|
"epoch": 1.6550946490808447,
|
|
"grad_norm": 0.27510447797500215,
|
|
"learning_rate": 8.894603353529662e-07,
|
|
"loss": 0.843,
|
|
"mean_token_accuracy": 0.780693593621254,
|
|
"num_tokens": 1211510407.0,
|
|
"step": 12110
|
|
},
|
|
{
|
|
"entropy": 0.8236328125,
|
|
"epoch": 1.6564614228114536,
|
|
"grad_norm": 0.3196420446558396,
|
|
"learning_rate": 8.859377201634495e-07,
|
|
"loss": 0.8223,
|
|
"mean_token_accuracy": 0.7837882548570633,
|
|
"num_tokens": 1212553005.0,
|
|
"step": 12120
|
|
},
|
|
{
|
|
"entropy": 0.9064453125,
|
|
"epoch": 1.6578281965420625,
|
|
"grad_norm": 0.2710811281367604,
|
|
"learning_rate": 8.824151049739326e-07,
|
|
"loss": 0.8987,
|
|
"mean_token_accuracy": 0.7682785570621491,
|
|
"num_tokens": 1213565725.0,
|
|
"step": 12130
|
|
},
|
|
{
|
|
"entropy": 0.8353515625,
|
|
"epoch": 1.6591949702726714,
|
|
"grad_norm": 0.2713378857851186,
|
|
"learning_rate": 8.788924897844161e-07,
|
|
"loss": 0.834,
|
|
"mean_token_accuracy": 0.7835211813449859,
|
|
"num_tokens": 1214603670.0,
|
|
"step": 12140
|
|
},
|
|
{
|
|
"entropy": 0.8373046875,
|
|
"epoch": 1.6605617440032803,
|
|
"grad_norm": 0.289967458807556,
|
|
"learning_rate": 8.753698745948994e-07,
|
|
"loss": 0.8273,
|
|
"mean_token_accuracy": 0.7841713279485703,
|
|
"num_tokens": 1215635088.0,
|
|
"step": 12150
|
|
},
|
|
{
|
|
"entropy": 0.8501953125,
|
|
"epoch": 1.6619285177338892,
|
|
"grad_norm": 0.2926160521309634,
|
|
"learning_rate": 8.718472594053825e-07,
|
|
"loss": 0.8384,
|
|
"mean_token_accuracy": 0.7823783993721009,
|
|
"num_tokens": 1216580894.0,
|
|
"step": 12160
|
|
},
|
|
{
|
|
"entropy": 0.85546875,
|
|
"epoch": 1.663295291464498,
|
|
"grad_norm": 0.29001587867980866,
|
|
"learning_rate": 8.683246442158659e-07,
|
|
"loss": 0.8511,
|
|
"mean_token_accuracy": 0.777684798836708,
|
|
"num_tokens": 1217558954.0,
|
|
"step": 12170
|
|
},
|
|
{
|
|
"entropy": 0.833984375,
|
|
"epoch": 1.664662065195107,
|
|
"grad_norm": 0.27592928768268643,
|
|
"learning_rate": 8.648020290263492e-07,
|
|
"loss": 0.8262,
|
|
"mean_token_accuracy": 0.7842383831739426,
|
|
"num_tokens": 1218517193.0,
|
|
"step": 12180
|
|
},
|
|
{
|
|
"entropy": 0.8626953125,
|
|
"epoch": 1.6660288389257158,
|
|
"grad_norm": 0.28602250077495234,
|
|
"learning_rate": 8.612794138368325e-07,
|
|
"loss": 0.8588,
|
|
"mean_token_accuracy": 0.7779427886009216,
|
|
"num_tokens": 1219524296.0,
|
|
"step": 12190
|
|
},
|
|
{
|
|
"entropy": 0.87890625,
|
|
"epoch": 1.6673956126563247,
|
|
"grad_norm": 0.30870016010713175,
|
|
"learning_rate": 8.577567986473158e-07,
|
|
"loss": 0.8703,
|
|
"mean_token_accuracy": 0.7763081014156341,
|
|
"num_tokens": 1220486825.0,
|
|
"step": 12200
|
|
},
|
|
{
|
|
"entropy": 0.851953125,
|
|
"epoch": 1.6687623863869336,
|
|
"grad_norm": 0.2893179717439089,
|
|
"learning_rate": 8.542341834577991e-07,
|
|
"loss": 0.8421,
|
|
"mean_token_accuracy": 0.7805455714464188,
|
|
"num_tokens": 1221470776.0,
|
|
"step": 12210
|
|
},
|
|
{
|
|
"entropy": 0.8556640625,
|
|
"epoch": 1.6701291601175425,
|
|
"grad_norm": 0.29236793390764215,
|
|
"learning_rate": 8.507115682682824e-07,
|
|
"loss": 0.8536,
|
|
"mean_token_accuracy": 0.7786176413297653,
|
|
"num_tokens": 1222444866.0,
|
|
"step": 12220
|
|
},
|
|
{
|
|
"entropy": 0.9021484375,
|
|
"epoch": 1.6714959338481514,
|
|
"grad_norm": 0.2894534281184185,
|
|
"learning_rate": 8.471889530787658e-07,
|
|
"loss": 0.8987,
|
|
"mean_token_accuracy": 0.7691145002841949,
|
|
"num_tokens": 1223472539.0,
|
|
"step": 12230
|
|
},
|
|
{
|
|
"entropy": 0.8490234375,
|
|
"epoch": 1.6728627075787603,
|
|
"grad_norm": 0.27265193538590243,
|
|
"learning_rate": 8.43666337889249e-07,
|
|
"loss": 0.8424,
|
|
"mean_token_accuracy": 0.7791900873184204,
|
|
"num_tokens": 1224457678.0,
|
|
"step": 12240
|
|
},
|
|
{
|
|
"entropy": 0.8595703125,
|
|
"epoch": 1.6742294813093692,
|
|
"grad_norm": 0.2799182019731549,
|
|
"learning_rate": 8.401437226997323e-07,
|
|
"loss": 0.8608,
|
|
"mean_token_accuracy": 0.778246796131134,
|
|
"num_tokens": 1225479026.0,
|
|
"step": 12250
|
|
},
|
|
{
|
|
"entropy": 0.8546875,
|
|
"epoch": 1.675596255039978,
|
|
"grad_norm": 0.29714342438967356,
|
|
"learning_rate": 8.366211075102157e-07,
|
|
"loss": 0.8465,
|
|
"mean_token_accuracy": 0.7796145081520081,
|
|
"num_tokens": 1226404809.0,
|
|
"step": 12260
|
|
},
|
|
{
|
|
"entropy": 0.858984375,
|
|
"epoch": 1.676963028770587,
|
|
"grad_norm": 0.2996177139631168,
|
|
"learning_rate": 8.330984923206989e-07,
|
|
"loss": 0.8454,
|
|
"mean_token_accuracy": 0.7799507975578308,
|
|
"num_tokens": 1227427431.0,
|
|
"step": 12270
|
|
},
|
|
{
|
|
"entropy": 0.9001953125,
|
|
"epoch": 1.678329802501196,
|
|
"grad_norm": 0.31659758081487455,
|
|
"learning_rate": 8.295758771311822e-07,
|
|
"loss": 0.9039,
|
|
"mean_token_accuracy": 0.7681001245975494,
|
|
"num_tokens": 1228361534.0,
|
|
"step": 12280
|
|
},
|
|
{
|
|
"entropy": 0.879296875,
|
|
"epoch": 1.6796965762318048,
|
|
"grad_norm": 0.2547183661125712,
|
|
"learning_rate": 8.260532619416656e-07,
|
|
"loss": 0.8726,
|
|
"mean_token_accuracy": 0.7729923963546753,
|
|
"num_tokens": 1229330180.0,
|
|
"step": 12290
|
|
},
|
|
{
|
|
"entropy": 0.85234375,
|
|
"epoch": 1.6810633499624137,
|
|
"grad_norm": 0.27179733144996343,
|
|
"learning_rate": 8.225306467521489e-07,
|
|
"loss": 0.8439,
|
|
"mean_token_accuracy": 0.7811705976724624,
|
|
"num_tokens": 1230329561.0,
|
|
"step": 12300
|
|
},
|
|
{
|
|
"entropy": 0.8794921875,
|
|
"epoch": 1.6824301236930226,
|
|
"grad_norm": 0.283971137818226,
|
|
"learning_rate": 8.190080315626321e-07,
|
|
"loss": 0.8825,
|
|
"mean_token_accuracy": 0.7736949980258941,
|
|
"num_tokens": 1231293719.0,
|
|
"step": 12310
|
|
},
|
|
{
|
|
"entropy": 0.8666015625,
|
|
"epoch": 1.6837968974236315,
|
|
"grad_norm": 0.27059616399801584,
|
|
"learning_rate": 8.154854163731155e-07,
|
|
"loss": 0.8552,
|
|
"mean_token_accuracy": 0.7778140068054199,
|
|
"num_tokens": 1232326259.0,
|
|
"step": 12320
|
|
},
|
|
{
|
|
"entropy": 0.862890625,
|
|
"epoch": 1.6851636711542404,
|
|
"grad_norm": 0.29558538535854156,
|
|
"learning_rate": 8.119628011835988e-07,
|
|
"loss": 0.8603,
|
|
"mean_token_accuracy": 0.7751149296760559,
|
|
"num_tokens": 1233306218.0,
|
|
"step": 12330
|
|
},
|
|
{
|
|
"entropy": 0.8572265625,
|
|
"epoch": 1.6865304448848493,
|
|
"grad_norm": 0.28942395310828617,
|
|
"learning_rate": 8.084401859940821e-07,
|
|
"loss": 0.8465,
|
|
"mean_token_accuracy": 0.7790455222129822,
|
|
"num_tokens": 1234278410.0,
|
|
"step": 12340
|
|
},
|
|
{
|
|
"entropy": 0.8974609375,
|
|
"epoch": 1.6878972186154582,
|
|
"grad_norm": 0.2880145334436267,
|
|
"learning_rate": 8.049175708045654e-07,
|
|
"loss": 0.9045,
|
|
"mean_token_accuracy": 0.7678204298019409,
|
|
"num_tokens": 1235291188.0,
|
|
"step": 12350
|
|
},
|
|
{
|
|
"entropy": 0.8310546875,
|
|
"epoch": 1.689263992346067,
|
|
"grad_norm": 0.2579580579542964,
|
|
"learning_rate": 8.013949556150487e-07,
|
|
"loss": 0.8236,
|
|
"mean_token_accuracy": 0.7826025754213333,
|
|
"num_tokens": 1236240073.0,
|
|
"step": 12360
|
|
},
|
|
{
|
|
"entropy": 0.8583984375,
|
|
"epoch": 1.690630766076676,
|
|
"grad_norm": 0.30296177216695963,
|
|
"learning_rate": 7.97872340425532e-07,
|
|
"loss": 0.851,
|
|
"mean_token_accuracy": 0.7787958353757858,
|
|
"num_tokens": 1237244925.0,
|
|
"step": 12370
|
|
},
|
|
{
|
|
"entropy": 0.87578125,
|
|
"epoch": 1.6919975398072848,
|
|
"grad_norm": 0.32144092094208615,
|
|
"learning_rate": 7.943497252360153e-07,
|
|
"loss": 0.8772,
|
|
"mean_token_accuracy": 0.7716000437736511,
|
|
"num_tokens": 1238216788.0,
|
|
"step": 12380
|
|
},
|
|
{
|
|
"entropy": 0.8796875,
|
|
"epoch": 1.6933643135378937,
|
|
"grad_norm": 0.28893961961146813,
|
|
"learning_rate": 7.908271100464986e-07,
|
|
"loss": 0.8802,
|
|
"mean_token_accuracy": 0.7713866263628006,
|
|
"num_tokens": 1239252019.0,
|
|
"step": 12390
|
|
},
|
|
{
|
|
"entropy": 0.872265625,
|
|
"epoch": 1.6947310872685026,
|
|
"grad_norm": 0.32874268974093307,
|
|
"learning_rate": 7.873044948569819e-07,
|
|
"loss": 0.8793,
|
|
"mean_token_accuracy": 0.7741204053163528,
|
|
"num_tokens": 1240249609.0,
|
|
"step": 12400
|
|
},
|
|
{
|
|
"entropy": 0.8767578125,
|
|
"epoch": 1.6960978609991115,
|
|
"grad_norm": 0.2658571424976563,
|
|
"learning_rate": 7.837818796674653e-07,
|
|
"loss": 0.8665,
|
|
"mean_token_accuracy": 0.7768232971429825,
|
|
"num_tokens": 1241251755.0,
|
|
"step": 12410
|
|
},
|
|
{
|
|
"entropy": 0.8591796875,
|
|
"epoch": 1.6974646347297204,
|
|
"grad_norm": 0.32829987453317633,
|
|
"learning_rate": 7.802592644779484e-07,
|
|
"loss": 0.8489,
|
|
"mean_token_accuracy": 0.7782013595104218,
|
|
"num_tokens": 1242296692.0,
|
|
"step": 12420
|
|
},
|
|
{
|
|
"entropy": 0.8802734375,
|
|
"epoch": 1.6988314084603293,
|
|
"grad_norm": 0.311172222714429,
|
|
"learning_rate": 7.767366492884317e-07,
|
|
"loss": 0.8809,
|
|
"mean_token_accuracy": 0.7719771295785904,
|
|
"num_tokens": 1243281814.0,
|
|
"step": 12430
|
|
},
|
|
{
|
|
"entropy": 0.8798828125,
|
|
"epoch": 1.7001981821909382,
|
|
"grad_norm": 0.2807078720149518,
|
|
"learning_rate": 7.732140340989152e-07,
|
|
"loss": 0.8758,
|
|
"mean_token_accuracy": 0.7737786293029785,
|
|
"num_tokens": 1244322236.0,
|
|
"step": 12440
|
|
},
|
|
{
|
|
"entropy": 0.8658203125,
|
|
"epoch": 1.701564955921547,
|
|
"grad_norm": 0.29211060190261856,
|
|
"learning_rate": 7.696914189093985e-07,
|
|
"loss": 0.8547,
|
|
"mean_token_accuracy": 0.777660584449768,
|
|
"num_tokens": 1245322367.0,
|
|
"step": 12450
|
|
},
|
|
{
|
|
"entropy": 0.8478515625,
|
|
"epoch": 1.702931729652156,
|
|
"grad_norm": 0.23949352390091777,
|
|
"learning_rate": 7.661688037198816e-07,
|
|
"loss": 0.8405,
|
|
"mean_token_accuracy": 0.7790974825620651,
|
|
"num_tokens": 1246334299.0,
|
|
"step": 12460
|
|
},
|
|
{
|
|
"entropy": 0.8392578125,
|
|
"epoch": 1.704298503382765,
|
|
"grad_norm": 0.27031768087365915,
|
|
"learning_rate": 7.62646188530365e-07,
|
|
"loss": 0.8337,
|
|
"mean_token_accuracy": 0.7822641432285309,
|
|
"num_tokens": 1247319119.0,
|
|
"step": 12470
|
|
},
|
|
{
|
|
"entropy": 0.8359375,
|
|
"epoch": 1.7056652771133738,
|
|
"grad_norm": 0.2708286767980263,
|
|
"learning_rate": 7.591235733408483e-07,
|
|
"loss": 0.8252,
|
|
"mean_token_accuracy": 0.7833973139524459,
|
|
"num_tokens": 1248330237.0,
|
|
"step": 12480
|
|
},
|
|
{
|
|
"entropy": 0.8181640625,
|
|
"epoch": 1.7070320508439827,
|
|
"grad_norm": 0.31047707698955007,
|
|
"learning_rate": 7.556009581513315e-07,
|
|
"loss": 0.8051,
|
|
"mean_token_accuracy": 0.7871794283390046,
|
|
"num_tokens": 1249326932.0,
|
|
"step": 12490
|
|
},
|
|
{
|
|
"entropy": 0.8234375,
|
|
"epoch": 1.7083988245745916,
|
|
"grad_norm": 0.3048451258146698,
|
|
"learning_rate": 7.520783429618149e-07,
|
|
"loss": 0.8307,
|
|
"mean_token_accuracy": 0.7835266321897507,
|
|
"num_tokens": 1250364367.0,
|
|
"step": 12500
|
|
},
|
|
{
|
|
"entropy": 0.8380859375,
|
|
"epoch": 1.7097655983052005,
|
|
"grad_norm": 0.293905529737557,
|
|
"learning_rate": 7.485557277722982e-07,
|
|
"loss": 0.84,
|
|
"mean_token_accuracy": 0.7825448125600815,
|
|
"num_tokens": 1251368785.0,
|
|
"step": 12510
|
|
},
|
|
{
|
|
"entropy": 0.865234375,
|
|
"epoch": 1.7111323720358094,
|
|
"grad_norm": 0.26245897386032035,
|
|
"learning_rate": 7.450331125827815e-07,
|
|
"loss": 0.863,
|
|
"mean_token_accuracy": 0.7757230252027512,
|
|
"num_tokens": 1252374565.0,
|
|
"step": 12520
|
|
},
|
|
{
|
|
"entropy": 0.8609375,
|
|
"epoch": 1.7124991457664183,
|
|
"grad_norm": 0.3242223533367697,
|
|
"learning_rate": 7.415104973932648e-07,
|
|
"loss": 0.8511,
|
|
"mean_token_accuracy": 0.7783118963241578,
|
|
"num_tokens": 1253381014.0,
|
|
"step": 12530
|
|
},
|
|
{
|
|
"entropy": 0.8357421875,
|
|
"epoch": 1.7138659194970272,
|
|
"grad_norm": 0.30664393370176746,
|
|
"learning_rate": 7.379878822037481e-07,
|
|
"loss": 0.8323,
|
|
"mean_token_accuracy": 0.7812796920537949,
|
|
"num_tokens": 1254317373.0,
|
|
"step": 12540
|
|
},
|
|
{
|
|
"entropy": 0.9029296875,
|
|
"epoch": 1.715232693227636,
|
|
"grad_norm": 0.28713617752898973,
|
|
"learning_rate": 7.344652670142314e-07,
|
|
"loss": 0.8912,
|
|
"mean_token_accuracy": 0.7701553255319595,
|
|
"num_tokens": 1255314246.0,
|
|
"step": 12550
|
|
},
|
|
{
|
|
"entropy": 0.8650390625,
|
|
"epoch": 1.716599466958245,
|
|
"grad_norm": 0.28847897785285403,
|
|
"learning_rate": 7.309426518247147e-07,
|
|
"loss": 0.8577,
|
|
"mean_token_accuracy": 0.7771687060594559,
|
|
"num_tokens": 1256346558.0,
|
|
"step": 12560
|
|
},
|
|
{
|
|
"entropy": 0.864453125,
|
|
"epoch": 1.7179662406888538,
|
|
"grad_norm": 0.2855284274626001,
|
|
"learning_rate": 7.27420036635198e-07,
|
|
"loss": 0.8687,
|
|
"mean_token_accuracy": 0.7755021721124649,
|
|
"num_tokens": 1257320925.0,
|
|
"step": 12570
|
|
},
|
|
{
|
|
"entropy": 0.8681640625,
|
|
"epoch": 1.7193330144194627,
|
|
"grad_norm": 0.28304033174830534,
|
|
"learning_rate": 7.238974214456813e-07,
|
|
"loss": 0.8658,
|
|
"mean_token_accuracy": 0.7737716287374496,
|
|
"num_tokens": 1258356911.0,
|
|
"step": 12580
|
|
},
|
|
{
|
|
"entropy": 0.8669921875,
|
|
"epoch": 1.7206997881500716,
|
|
"grad_norm": 0.2828765580568121,
|
|
"learning_rate": 7.203748062561647e-07,
|
|
"loss": 0.8603,
|
|
"mean_token_accuracy": 0.7777844905853272,
|
|
"num_tokens": 1259426716.0,
|
|
"step": 12590
|
|
},
|
|
{
|
|
"entropy": 0.8591796875,
|
|
"epoch": 1.7220665618806805,
|
|
"grad_norm": 0.2913553303327493,
|
|
"learning_rate": 7.168521910666479e-07,
|
|
"loss": 0.8559,
|
|
"mean_token_accuracy": 0.7787601619958877,
|
|
"num_tokens": 1260419988.0,
|
|
"step": 12600
|
|
},
|
|
{
|
|
"entropy": 0.8619140625,
|
|
"epoch": 1.7234333356112894,
|
|
"grad_norm": 0.2739682612018677,
|
|
"learning_rate": 7.133295758771312e-07,
|
|
"loss": 0.8679,
|
|
"mean_token_accuracy": 0.7749360471963882,
|
|
"num_tokens": 1261416792.0,
|
|
"step": 12610
|
|
},
|
|
{
|
|
"entropy": 0.8416015625,
|
|
"epoch": 1.7248001093418983,
|
|
"grad_norm": 0.28972988678663913,
|
|
"learning_rate": 7.098069606876146e-07,
|
|
"loss": 0.8345,
|
|
"mean_token_accuracy": 0.7823008924722672,
|
|
"num_tokens": 1262413344.0,
|
|
"step": 12620
|
|
},
|
|
{
|
|
"entropy": 0.8431640625,
|
|
"epoch": 1.7261668830725072,
|
|
"grad_norm": 0.3077027009502952,
|
|
"learning_rate": 7.062843454980979e-07,
|
|
"loss": 0.8443,
|
|
"mean_token_accuracy": 0.782577633857727,
|
|
"num_tokens": 1263384718.0,
|
|
"step": 12630
|
|
},
|
|
{
|
|
"entropy": 0.85078125,
|
|
"epoch": 1.727533656803116,
|
|
"grad_norm": 0.27883356346738186,
|
|
"learning_rate": 7.027617303085811e-07,
|
|
"loss": 0.8476,
|
|
"mean_token_accuracy": 0.7806531667709351,
|
|
"num_tokens": 1264399034.0,
|
|
"step": 12640
|
|
},
|
|
{
|
|
"entropy": 0.8513671875,
|
|
"epoch": 1.728900430533725,
|
|
"grad_norm": 0.33715132936136316,
|
|
"learning_rate": 6.992391151190645e-07,
|
|
"loss": 0.8597,
|
|
"mean_token_accuracy": 0.7778229743242264,
|
|
"num_tokens": 1265410032.0,
|
|
"step": 12650
|
|
},
|
|
{
|
|
"entropy": 0.8861328125,
|
|
"epoch": 1.730267204264334,
|
|
"grad_norm": 0.310940345799163,
|
|
"learning_rate": 6.957164999295478e-07,
|
|
"loss": 0.8807,
|
|
"mean_token_accuracy": 0.7710263967514038,
|
|
"num_tokens": 1266366798.0,
|
|
"step": 12660
|
|
},
|
|
{
|
|
"entropy": 0.849609375,
|
|
"epoch": 1.7316339779949428,
|
|
"grad_norm": 0.2795783444901613,
|
|
"learning_rate": 6.921938847400309e-07,
|
|
"loss": 0.8424,
|
|
"mean_token_accuracy": 0.7776149988174439,
|
|
"num_tokens": 1267326616.0,
|
|
"step": 12670
|
|
},
|
|
{
|
|
"entropy": 0.8603515625,
|
|
"epoch": 1.7330007517255517,
|
|
"grad_norm": 0.2556123708527629,
|
|
"learning_rate": 6.886712695505144e-07,
|
|
"loss": 0.8597,
|
|
"mean_token_accuracy": 0.777879199385643,
|
|
"num_tokens": 1268360970.0,
|
|
"step": 12680
|
|
},
|
|
{
|
|
"entropy": 0.903515625,
|
|
"epoch": 1.7343675254561606,
|
|
"grad_norm": 0.282713722028902,
|
|
"learning_rate": 6.851486543609977e-07,
|
|
"loss": 0.9075,
|
|
"mean_token_accuracy": 0.7659117221832276,
|
|
"num_tokens": 1269369794.0,
|
|
"step": 12690
|
|
},
|
|
{
|
|
"entropy": 0.883984375,
|
|
"epoch": 1.7357342991867695,
|
|
"grad_norm": 0.2779365562590532,
|
|
"learning_rate": 6.816260391714811e-07,
|
|
"loss": 0.8892,
|
|
"mean_token_accuracy": 0.7709726363420486,
|
|
"num_tokens": 1270357489.0,
|
|
"step": 12700
|
|
},
|
|
{
|
|
"entropy": 0.8748046875,
|
|
"epoch": 1.7371010729173784,
|
|
"grad_norm": 0.3474426894846395,
|
|
"learning_rate": 6.781034239819642e-07,
|
|
"loss": 0.8745,
|
|
"mean_token_accuracy": 0.7739379167556762,
|
|
"num_tokens": 1271356111.0,
|
|
"step": 12710
|
|
},
|
|
{
|
|
"entropy": 0.84765625,
|
|
"epoch": 1.7384678466479875,
|
|
"grad_norm": 0.2719322003092257,
|
|
"learning_rate": 6.745808087924475e-07,
|
|
"loss": 0.841,
|
|
"mean_token_accuracy": 0.7800267904996872,
|
|
"num_tokens": 1272347602.0,
|
|
"step": 12720
|
|
},
|
|
{
|
|
"entropy": 0.8744140625,
|
|
"epoch": 1.7398346203785964,
|
|
"grad_norm": 0.2999369416004266,
|
|
"learning_rate": 6.710581936029308e-07,
|
|
"loss": 0.881,
|
|
"mean_token_accuracy": 0.7730445951223374,
|
|
"num_tokens": 1273333765.0,
|
|
"step": 12730
|
|
},
|
|
{
|
|
"entropy": 0.831640625,
|
|
"epoch": 1.7412013941092053,
|
|
"grad_norm": 0.28969592353076096,
|
|
"learning_rate": 6.675355784134143e-07,
|
|
"loss": 0.8311,
|
|
"mean_token_accuracy": 0.7816889345645904,
|
|
"num_tokens": 1274373958.0,
|
|
"step": 12740
|
|
},
|
|
{
|
|
"entropy": 0.8740234375,
|
|
"epoch": 1.7425681678398142,
|
|
"grad_norm": 0.3036166107099827,
|
|
"learning_rate": 6.640129632238974e-07,
|
|
"loss": 0.8631,
|
|
"mean_token_accuracy": 0.7776484757661819,
|
|
"num_tokens": 1275372212.0,
|
|
"step": 12750
|
|
},
|
|
{
|
|
"entropy": 0.891015625,
|
|
"epoch": 1.743934941570423,
|
|
"grad_norm": 0.31821314259825895,
|
|
"learning_rate": 6.604903480343807e-07,
|
|
"loss": 0.8885,
|
|
"mean_token_accuracy": 0.7691589057445526,
|
|
"num_tokens": 1276394337.0,
|
|
"step": 12760
|
|
},
|
|
{
|
|
"entropy": 0.841015625,
|
|
"epoch": 1.745301715301032,
|
|
"grad_norm": 0.3113033713309314,
|
|
"learning_rate": 6.569677328448641e-07,
|
|
"loss": 0.8348,
|
|
"mean_token_accuracy": 0.7830172151327133,
|
|
"num_tokens": 1277360803.0,
|
|
"step": 12770
|
|
},
|
|
{
|
|
"entropy": 0.8365234375,
|
|
"epoch": 1.7466684890316408,
|
|
"grad_norm": 0.2911154761626767,
|
|
"learning_rate": 6.534451176553473e-07,
|
|
"loss": 0.8252,
|
|
"mean_token_accuracy": 0.7856366157531738,
|
|
"num_tokens": 1278352607.0,
|
|
"step": 12780
|
|
},
|
|
{
|
|
"entropy": 0.8564453125,
|
|
"epoch": 1.7480352627622497,
|
|
"grad_norm": 0.3209613412314652,
|
|
"learning_rate": 6.499225024658306e-07,
|
|
"loss": 0.8527,
|
|
"mean_token_accuracy": 0.7766816645860672,
|
|
"num_tokens": 1279376044.0,
|
|
"step": 12790
|
|
},
|
|
{
|
|
"entropy": 0.851171875,
|
|
"epoch": 1.7494020364928586,
|
|
"grad_norm": 0.27460030604211205,
|
|
"learning_rate": 6.46399887276314e-07,
|
|
"loss": 0.8445,
|
|
"mean_token_accuracy": 0.7813642024993896,
|
|
"num_tokens": 1280327985.0,
|
|
"step": 12800
|
|
},
|
|
{
|
|
"entropy": 0.89375,
|
|
"epoch": 1.7507688102234675,
|
|
"grad_norm": 0.3253076844943484,
|
|
"learning_rate": 6.428772720867973e-07,
|
|
"loss": 0.8838,
|
|
"mean_token_accuracy": 0.7711131900548935,
|
|
"num_tokens": 1281302003.0,
|
|
"step": 12810
|
|
},
|
|
{
|
|
"entropy": 0.8662109375,
|
|
"epoch": 1.7521355839540764,
|
|
"grad_norm": 0.2760234919530227,
|
|
"learning_rate": 6.393546568972805e-07,
|
|
"loss": 0.8602,
|
|
"mean_token_accuracy": 0.7755971848964691,
|
|
"num_tokens": 1282270663.0,
|
|
"step": 12820
|
|
},
|
|
{
|
|
"entropy": 0.8765625,
|
|
"epoch": 1.7535023576846853,
|
|
"grad_norm": 0.2931790089560915,
|
|
"learning_rate": 6.358320417077639e-07,
|
|
"loss": 0.8708,
|
|
"mean_token_accuracy": 0.7747057318687439,
|
|
"num_tokens": 1283257236.0,
|
|
"step": 12830
|
|
},
|
|
{
|
|
"entropy": 0.8501953125,
|
|
"epoch": 1.7548691314152942,
|
|
"grad_norm": 0.28414654157834,
|
|
"learning_rate": 6.323094265182472e-07,
|
|
"loss": 0.8515,
|
|
"mean_token_accuracy": 0.7770916193723678,
|
|
"num_tokens": 1284214291.0,
|
|
"step": 12840
|
|
},
|
|
{
|
|
"entropy": 0.825,
|
|
"epoch": 1.756235905145903,
|
|
"grad_norm": 0.2686629591569446,
|
|
"learning_rate": 6.287868113287306e-07,
|
|
"loss": 0.8115,
|
|
"mean_token_accuracy": 0.78643339574337,
|
|
"num_tokens": 1285254842.0,
|
|
"step": 12850
|
|
},
|
|
{
|
|
"entropy": 0.851171875,
|
|
"epoch": 1.757602678876512,
|
|
"grad_norm": 0.3036890408755304,
|
|
"learning_rate": 6.252641961392138e-07,
|
|
"loss": 0.8509,
|
|
"mean_token_accuracy": 0.7789995819330215,
|
|
"num_tokens": 1286277014.0,
|
|
"step": 12860
|
|
},
|
|
{
|
|
"entropy": 0.81484375,
|
|
"epoch": 1.758969452607121,
|
|
"grad_norm": 0.2649895374010035,
|
|
"learning_rate": 6.217415809496971e-07,
|
|
"loss": 0.8204,
|
|
"mean_token_accuracy": 0.78486949801445,
|
|
"num_tokens": 1287287798.0,
|
|
"step": 12870
|
|
},
|
|
{
|
|
"entropy": 0.877734375,
|
|
"epoch": 1.7603362263377298,
|
|
"grad_norm": 0.2884238491267175,
|
|
"learning_rate": 6.182189657601804e-07,
|
|
"loss": 0.8866,
|
|
"mean_token_accuracy": 0.7730829536914825,
|
|
"num_tokens": 1288316440.0,
|
|
"step": 12880
|
|
},
|
|
{
|
|
"entropy": 0.858984375,
|
|
"epoch": 1.7617030000683387,
|
|
"grad_norm": 0.31350662086582887,
|
|
"learning_rate": 6.146963505706637e-07,
|
|
"loss": 0.857,
|
|
"mean_token_accuracy": 0.7772395074367523,
|
|
"num_tokens": 1289303570.0,
|
|
"step": 12890
|
|
},
|
|
{
|
|
"entropy": 0.8708984375,
|
|
"epoch": 1.7630697737989476,
|
|
"grad_norm": 0.2851016189187536,
|
|
"learning_rate": 6.11173735381147e-07,
|
|
"loss": 0.8632,
|
|
"mean_token_accuracy": 0.7777277797460556,
|
|
"num_tokens": 1290267351.0,
|
|
"step": 12900
|
|
},
|
|
{
|
|
"entropy": 0.8689453125,
|
|
"epoch": 1.7644365475295565,
|
|
"grad_norm": 0.26537577145443136,
|
|
"learning_rate": 6.076511201916304e-07,
|
|
"loss": 0.8801,
|
|
"mean_token_accuracy": 0.7741623729467392,
|
|
"num_tokens": 1291253162.0,
|
|
"step": 12910
|
|
},
|
|
{
|
|
"entropy": 0.85,
|
|
"epoch": 1.7658033212601654,
|
|
"grad_norm": 0.32421593447456387,
|
|
"learning_rate": 6.041285050021136e-07,
|
|
"loss": 0.8436,
|
|
"mean_token_accuracy": 0.777398219704628,
|
|
"num_tokens": 1292188353.0,
|
|
"step": 12920
|
|
},
|
|
{
|
|
"entropy": 0.87578125,
|
|
"epoch": 1.7671700949907743,
|
|
"grad_norm": 0.27263706899419843,
|
|
"learning_rate": 6.006058898125969e-07,
|
|
"loss": 0.8677,
|
|
"mean_token_accuracy": 0.7756602883338928,
|
|
"num_tokens": 1293232443.0,
|
|
"step": 12930
|
|
},
|
|
{
|
|
"entropy": 0.8951171875,
|
|
"epoch": 1.7685368687213832,
|
|
"grad_norm": 0.29709894711557905,
|
|
"learning_rate": 5.970832746230803e-07,
|
|
"loss": 0.9004,
|
|
"mean_token_accuracy": 0.7689566016197205,
|
|
"num_tokens": 1294305727.0,
|
|
"step": 12940
|
|
},
|
|
{
|
|
"entropy": 0.8787109375,
|
|
"epoch": 1.769903642451992,
|
|
"grad_norm": 0.3092671264978073,
|
|
"learning_rate": 5.935606594335636e-07,
|
|
"loss": 0.8684,
|
|
"mean_token_accuracy": 0.7747088223695755,
|
|
"num_tokens": 1295353743.0,
|
|
"step": 12950
|
|
},
|
|
{
|
|
"entropy": 0.871484375,
|
|
"epoch": 1.771270416182601,
|
|
"grad_norm": 0.27704445707431075,
|
|
"learning_rate": 5.900380442440468e-07,
|
|
"loss": 0.8715,
|
|
"mean_token_accuracy": 0.7746877431869507,
|
|
"num_tokens": 1296349968.0,
|
|
"step": 12960
|
|
},
|
|
{
|
|
"entropy": 0.9171875,
|
|
"epoch": 1.7726371899132098,
|
|
"grad_norm": 0.270933271429711,
|
|
"learning_rate": 5.8651542905453e-07,
|
|
"loss": 0.9106,
|
|
"mean_token_accuracy": 0.7664764702320099,
|
|
"num_tokens": 1297364866.0,
|
|
"step": 12970
|
|
},
|
|
{
|
|
"entropy": 0.8421875,
|
|
"epoch": 1.7740039636438187,
|
|
"grad_norm": 0.3055707913035734,
|
|
"learning_rate": 5.829928138650135e-07,
|
|
"loss": 0.842,
|
|
"mean_token_accuracy": 0.781931248307228,
|
|
"num_tokens": 1298351650.0,
|
|
"step": 12980
|
|
},
|
|
{
|
|
"entropy": 0.8548828125,
|
|
"epoch": 1.7753707373744276,
|
|
"grad_norm": 0.2730320528301002,
|
|
"learning_rate": 5.794701986754967e-07,
|
|
"loss": 0.8532,
|
|
"mean_token_accuracy": 0.7777145713567734,
|
|
"num_tokens": 1299341127.0,
|
|
"step": 12990
|
|
},
|
|
{
|
|
"entropy": 0.86640625,
|
|
"epoch": 1.7767375111050365,
|
|
"grad_norm": 0.2876266985118012,
|
|
"learning_rate": 5.759475834859801e-07,
|
|
"loss": 0.866,
|
|
"mean_token_accuracy": 0.7759812861680985,
|
|
"num_tokens": 1300341259.0,
|
|
"step": 13000
|
|
},
|
|
{
|
|
"entropy": 0.8955078125,
|
|
"epoch": 1.7781042848356454,
|
|
"grad_norm": 0.33542372698752926,
|
|
"learning_rate": 5.724249682964633e-07,
|
|
"loss": 0.8948,
|
|
"mean_token_accuracy": 0.7696751356124878,
|
|
"num_tokens": 1301303846.0,
|
|
"step": 13010
|
|
},
|
|
{
|
|
"entropy": 0.8724609375,
|
|
"epoch": 1.7794710585662543,
|
|
"grad_norm": 0.30215901202157264,
|
|
"learning_rate": 5.689023531069466e-07,
|
|
"loss": 0.8688,
|
|
"mean_token_accuracy": 0.7737321138381958,
|
|
"num_tokens": 1302348199.0,
|
|
"step": 13020
|
|
},
|
|
{
|
|
"entropy": 0.8744140625,
|
|
"epoch": 1.7808378322968632,
|
|
"grad_norm": 0.2713449079689415,
|
|
"learning_rate": 5.6537973791743e-07,
|
|
"loss": 0.884,
|
|
"mean_token_accuracy": 0.7727026909589767,
|
|
"num_tokens": 1303327894.0,
|
|
"step": 13030
|
|
},
|
|
{
|
|
"entropy": 0.833984375,
|
|
"epoch": 1.782204606027472,
|
|
"grad_norm": 0.27279675050971286,
|
|
"learning_rate": 5.618571227279133e-07,
|
|
"loss": 0.8342,
|
|
"mean_token_accuracy": 0.7837386667728424,
|
|
"num_tokens": 1304363815.0,
|
|
"step": 13040
|
|
},
|
|
{
|
|
"entropy": 0.840625,
|
|
"epoch": 1.783571379758081,
|
|
"grad_norm": 0.2890326490647689,
|
|
"learning_rate": 5.583345075383965e-07,
|
|
"loss": 0.8333,
|
|
"mean_token_accuracy": 0.7818005710840226,
|
|
"num_tokens": 1305398954.0,
|
|
"step": 13050
|
|
},
|
|
{
|
|
"entropy": 0.82734375,
|
|
"epoch": 1.78493815348869,
|
|
"grad_norm": 0.29301380848148606,
|
|
"learning_rate": 5.548118923488799e-07,
|
|
"loss": 0.8267,
|
|
"mean_token_accuracy": 0.7829346209764481,
|
|
"num_tokens": 1306358108.0,
|
|
"step": 13060
|
|
},
|
|
{
|
|
"entropy": 0.8478515625,
|
|
"epoch": 1.786304927219299,
|
|
"grad_norm": 0.27386416308434386,
|
|
"learning_rate": 5.512892771593632e-07,
|
|
"loss": 0.8363,
|
|
"mean_token_accuracy": 0.780574893951416,
|
|
"num_tokens": 1307372279.0,
|
|
"step": 13070
|
|
},
|
|
{
|
|
"entropy": 0.8923828125,
|
|
"epoch": 1.787671700949908,
|
|
"grad_norm": 0.2594162777391185,
|
|
"learning_rate": 5.477666619698464e-07,
|
|
"loss": 0.8996,
|
|
"mean_token_accuracy": 0.7693529993295669,
|
|
"num_tokens": 1308383483.0,
|
|
"step": 13080
|
|
},
|
|
{
|
|
"entropy": 0.8583984375,
|
|
"epoch": 1.7890384746805168,
|
|
"grad_norm": 0.27366989739173797,
|
|
"learning_rate": 5.442440467803298e-07,
|
|
"loss": 0.8534,
|
|
"mean_token_accuracy": 0.7779334396123886,
|
|
"num_tokens": 1309378075.0,
|
|
"step": 13090
|
|
},
|
|
{
|
|
"entropy": 0.8794921875,
|
|
"epoch": 1.7904052484111257,
|
|
"grad_norm": 0.3363945744665371,
|
|
"learning_rate": 5.407214315908131e-07,
|
|
"loss": 0.8649,
|
|
"mean_token_accuracy": 0.7766058683395386,
|
|
"num_tokens": 1310398593.0,
|
|
"step": 13100
|
|
},
|
|
{
|
|
"entropy": 0.865625,
|
|
"epoch": 1.7917720221417346,
|
|
"grad_norm": 0.2644088832930569,
|
|
"learning_rate": 5.371988164012964e-07,
|
|
"loss": 0.8601,
|
|
"mean_token_accuracy": 0.7754588216543198,
|
|
"num_tokens": 1311444376.0,
|
|
"step": 13110
|
|
},
|
|
{
|
|
"entropy": 0.83125,
|
|
"epoch": 1.7931387958723435,
|
|
"grad_norm": 0.2563477217035111,
|
|
"learning_rate": 5.336762012117797e-07,
|
|
"loss": 0.8301,
|
|
"mean_token_accuracy": 0.7829560220241547,
|
|
"num_tokens": 1312465136.0,
|
|
"step": 13120
|
|
},
|
|
{
|
|
"entropy": 0.8662109375,
|
|
"epoch": 1.7945055696029524,
|
|
"grad_norm": 0.2933991675041587,
|
|
"learning_rate": 5.30153586022263e-07,
|
|
"loss": 0.8633,
|
|
"mean_token_accuracy": 0.7762812435626983,
|
|
"num_tokens": 1313480371.0,
|
|
"step": 13130
|
|
},
|
|
{
|
|
"entropy": 0.8646484375,
|
|
"epoch": 1.7958723433335613,
|
|
"grad_norm": 0.2566817042644541,
|
|
"learning_rate": 5.266309708327462e-07,
|
|
"loss": 0.8535,
|
|
"mean_token_accuracy": 0.7786486774682999,
|
|
"num_tokens": 1314476274.0,
|
|
"step": 13140
|
|
},
|
|
{
|
|
"entropy": 0.8380859375,
|
|
"epoch": 1.7972391170641702,
|
|
"grad_norm": 0.2886422330951591,
|
|
"learning_rate": 5.231083556432296e-07,
|
|
"loss": 0.8418,
|
|
"mean_token_accuracy": 0.7817468285560608,
|
|
"num_tokens": 1315455711.0,
|
|
"step": 13150
|
|
},
|
|
{
|
|
"entropy": 0.901171875,
|
|
"epoch": 1.798605890794779,
|
|
"grad_norm": 0.2969638425487085,
|
|
"learning_rate": 5.195857404537129e-07,
|
|
"loss": 0.9123,
|
|
"mean_token_accuracy": 0.7690717726945877,
|
|
"num_tokens": 1316446453.0,
|
|
"step": 13160
|
|
},
|
|
{
|
|
"entropy": 0.865625,
|
|
"epoch": 1.799972664525388,
|
|
"grad_norm": 0.3260131004673831,
|
|
"learning_rate": 5.160631252641961e-07,
|
|
"loss": 0.8548,
|
|
"mean_token_accuracy": 0.779237762093544,
|
|
"num_tokens": 1317416832.0,
|
|
"step": 13170
|
|
},
|
|
{
|
|
"entropy": 0.8501953125,
|
|
"epoch": 1.8013394382559968,
|
|
"grad_norm": 0.31709132030511616,
|
|
"learning_rate": 5.125405100746795e-07,
|
|
"loss": 0.8484,
|
|
"mean_token_accuracy": 0.780627766251564,
|
|
"num_tokens": 1318453000.0,
|
|
"step": 13180
|
|
},
|
|
{
|
|
"entropy": 0.8720703125,
|
|
"epoch": 1.8027062119866057,
|
|
"grad_norm": 0.29221717719491497,
|
|
"learning_rate": 5.090178948851628e-07,
|
|
"loss": 0.8777,
|
|
"mean_token_accuracy": 0.7724189698696137,
|
|
"num_tokens": 1319483648.0,
|
|
"step": 13190
|
|
},
|
|
{
|
|
"entropy": 0.9005859375,
|
|
"epoch": 1.8040729857172146,
|
|
"grad_norm": 0.3121181676245004,
|
|
"learning_rate": 5.054952796956461e-07,
|
|
"loss": 0.8962,
|
|
"mean_token_accuracy": 0.7695844650268555,
|
|
"num_tokens": 1320459238.0,
|
|
"step": 13200
|
|
},
|
|
{
|
|
"entropy": 0.828125,
|
|
"epoch": 1.8054397594478235,
|
|
"grad_norm": 0.2743650116421169,
|
|
"learning_rate": 5.019726645061294e-07,
|
|
"loss": 0.8245,
|
|
"mean_token_accuracy": 0.7849496513605118,
|
|
"num_tokens": 1321481916.0,
|
|
"step": 13210
|
|
},
|
|
{
|
|
"entropy": 0.909375,
|
|
"epoch": 1.8068065331784324,
|
|
"grad_norm": 0.32099416656757784,
|
|
"learning_rate": 4.984500493166127e-07,
|
|
"loss": 0.9122,
|
|
"mean_token_accuracy": 0.7689237892627716,
|
|
"num_tokens": 1322494318.0,
|
|
"step": 13220
|
|
},
|
|
{
|
|
"entropy": 0.8125,
|
|
"epoch": 1.8081733069090413,
|
|
"grad_norm": 0.47767939414232874,
|
|
"learning_rate": 4.94927434127096e-07,
|
|
"loss": 0.807,
|
|
"mean_token_accuracy": 0.7866127014160156,
|
|
"num_tokens": 1323477552.0,
|
|
"step": 13230
|
|
},
|
|
{
|
|
"entropy": 0.8548828125,
|
|
"epoch": 1.8095400806396502,
|
|
"grad_norm": 0.31866756823898607,
|
|
"learning_rate": 4.914048189375794e-07,
|
|
"loss": 0.8517,
|
|
"mean_token_accuracy": 0.7794121891260147,
|
|
"num_tokens": 1324450561.0,
|
|
"step": 13240
|
|
},
|
|
{
|
|
"entropy": 0.859765625,
|
|
"epoch": 1.8109068543702591,
|
|
"grad_norm": 0.3262247412438452,
|
|
"learning_rate": 4.878822037480626e-07,
|
|
"loss": 0.8502,
|
|
"mean_token_accuracy": 0.7781051218509674,
|
|
"num_tokens": 1325368306.0,
|
|
"step": 13250
|
|
},
|
|
{
|
|
"entropy": 0.8560546875,
|
|
"epoch": 1.812273628100868,
|
|
"grad_norm": 0.2530347072351271,
|
|
"learning_rate": 4.843595885585459e-07,
|
|
"loss": 0.8543,
|
|
"mean_token_accuracy": 0.7774137705564499,
|
|
"num_tokens": 1326335671.0,
|
|
"step": 13260
|
|
},
|
|
{
|
|
"entropy": 0.85546875,
|
|
"epoch": 1.813640401831477,
|
|
"grad_norm": 0.34384650325364663,
|
|
"learning_rate": 4.808369733690292e-07,
|
|
"loss": 0.856,
|
|
"mean_token_accuracy": 0.7767894804477692,
|
|
"num_tokens": 1327363211.0,
|
|
"step": 13270
|
|
},
|
|
{
|
|
"entropy": 0.90703125,
|
|
"epoch": 1.8150071755620858,
|
|
"grad_norm": 0.28318908035943524,
|
|
"learning_rate": 4.773143581795125e-07,
|
|
"loss": 0.9062,
|
|
"mean_token_accuracy": 0.7692195862531662,
|
|
"num_tokens": 1328406573.0,
|
|
"step": 13280
|
|
},
|
|
{
|
|
"entropy": 0.819921875,
|
|
"epoch": 1.8163739492926947,
|
|
"grad_norm": 0.2928752786888582,
|
|
"learning_rate": 4.7379174298999583e-07,
|
|
"loss": 0.8088,
|
|
"mean_token_accuracy": 0.7881260365247726,
|
|
"num_tokens": 1329389875.0,
|
|
"step": 13290
|
|
},
|
|
{
|
|
"entropy": 0.858984375,
|
|
"epoch": 1.8177407230233036,
|
|
"grad_norm": 0.33004053579604953,
|
|
"learning_rate": 4.702691278004791e-07,
|
|
"loss": 0.8559,
|
|
"mean_token_accuracy": 0.7788754492998123,
|
|
"num_tokens": 1330433660.0,
|
|
"step": 13300
|
|
},
|
|
{
|
|
"entropy": 0.8533203125,
|
|
"epoch": 1.8191074967539125,
|
|
"grad_norm": 0.3364923763976067,
|
|
"learning_rate": 4.6674651261096245e-07,
|
|
"loss": 0.8537,
|
|
"mean_token_accuracy": 0.7789005935192108,
|
|
"num_tokens": 1331440962.0,
|
|
"step": 13310
|
|
},
|
|
{
|
|
"entropy": 0.8259765625,
|
|
"epoch": 1.8204742704845214,
|
|
"grad_norm": 0.28254081530837755,
|
|
"learning_rate": 4.632238974214457e-07,
|
|
"loss": 0.8157,
|
|
"mean_token_accuracy": 0.7867335259914399,
|
|
"num_tokens": 1332454433.0,
|
|
"step": 13320
|
|
},
|
|
{
|
|
"entropy": 0.83984375,
|
|
"epoch": 1.8218410442151303,
|
|
"grad_norm": 0.2603532243282532,
|
|
"learning_rate": 4.5970128223192907e-07,
|
|
"loss": 0.8334,
|
|
"mean_token_accuracy": 0.7833128809928894,
|
|
"num_tokens": 1333452099.0,
|
|
"step": 13330
|
|
},
|
|
{
|
|
"entropy": 0.82578125,
|
|
"epoch": 1.8232078179457392,
|
|
"grad_norm": 0.27838510367366603,
|
|
"learning_rate": 4.561786670424123e-07,
|
|
"loss": 0.825,
|
|
"mean_token_accuracy": 0.7844924300909042,
|
|
"num_tokens": 1334427965.0,
|
|
"step": 13340
|
|
},
|
|
{
|
|
"entropy": 0.85859375,
|
|
"epoch": 1.824574591676348,
|
|
"grad_norm": 0.2903237303790456,
|
|
"learning_rate": 4.5265605185289563e-07,
|
|
"loss": 0.8672,
|
|
"mean_token_accuracy": 0.7765445828437805,
|
|
"num_tokens": 1335454235.0,
|
|
"step": 13350
|
|
},
|
|
{
|
|
"entropy": 0.84609375,
|
|
"epoch": 1.825941365406957,
|
|
"grad_norm": 0.2558873660238092,
|
|
"learning_rate": 4.491334366633789e-07,
|
|
"loss": 0.8343,
|
|
"mean_token_accuracy": 0.781817901134491,
|
|
"num_tokens": 1336477671.0,
|
|
"step": 13360
|
|
},
|
|
{
|
|
"entropy": 0.8830078125,
|
|
"epoch": 1.8273081391375658,
|
|
"grad_norm": 0.28315018655925006,
|
|
"learning_rate": 4.456108214738622e-07,
|
|
"loss": 0.8827,
|
|
"mean_token_accuracy": 0.7727438420057297,
|
|
"num_tokens": 1337476094.0,
|
|
"step": 13370
|
|
},
|
|
{
|
|
"entropy": 0.8650390625,
|
|
"epoch": 1.8286749128681747,
|
|
"grad_norm": 0.30660725233660585,
|
|
"learning_rate": 4.4208820628434553e-07,
|
|
"loss": 0.8572,
|
|
"mean_token_accuracy": 0.7781480848789215,
|
|
"num_tokens": 1338474602.0,
|
|
"step": 13380
|
|
},
|
|
{
|
|
"entropy": 0.8822265625,
|
|
"epoch": 1.8300416865987836,
|
|
"grad_norm": 0.2955845936864148,
|
|
"learning_rate": 4.385655910948288e-07,
|
|
"loss": 0.8713,
|
|
"mean_token_accuracy": 0.7725894689559937,
|
|
"num_tokens": 1339479929.0,
|
|
"step": 13390
|
|
},
|
|
{
|
|
"entropy": 0.88828125,
|
|
"epoch": 1.8314084603293925,
|
|
"grad_norm": 0.2770919664717596,
|
|
"learning_rate": 4.3504297590531215e-07,
|
|
"loss": 0.8921,
|
|
"mean_token_accuracy": 0.7699492424726486,
|
|
"num_tokens": 1340488570.0,
|
|
"step": 13400
|
|
},
|
|
{
|
|
"entropy": 0.83671875,
|
|
"epoch": 1.8327752340600014,
|
|
"grad_norm": 0.27443845563758895,
|
|
"learning_rate": 4.3152036071579543e-07,
|
|
"loss": 0.8508,
|
|
"mean_token_accuracy": 0.7824245005846023,
|
|
"num_tokens": 1341511821.0,
|
|
"step": 13410
|
|
},
|
|
{
|
|
"entropy": 0.8427734375,
|
|
"epoch": 1.8341420077906103,
|
|
"grad_norm": 0.2850086956375871,
|
|
"learning_rate": 4.2799774552627877e-07,
|
|
"loss": 0.8469,
|
|
"mean_token_accuracy": 0.7799729228019714,
|
|
"num_tokens": 1342536338.0,
|
|
"step": 13420
|
|
},
|
|
{
|
|
"entropy": 0.8515625,
|
|
"epoch": 1.8355087815212192,
|
|
"grad_norm": 0.3214839495190864,
|
|
"learning_rate": 4.2447513033676205e-07,
|
|
"loss": 0.8493,
|
|
"mean_token_accuracy": 0.7796382635831833,
|
|
"num_tokens": 1343526789.0,
|
|
"step": 13430
|
|
},
|
|
{
|
|
"entropy": 0.8525390625,
|
|
"epoch": 1.836875555251828,
|
|
"grad_norm": 0.30426974719448235,
|
|
"learning_rate": 4.209525151472454e-07,
|
|
"loss": 0.8386,
|
|
"mean_token_accuracy": 0.7817652255296708,
|
|
"num_tokens": 1344504256.0,
|
|
"step": 13440
|
|
},
|
|
{
|
|
"entropy": 0.8658203125,
|
|
"epoch": 1.838242328982437,
|
|
"grad_norm": 0.286818517987764,
|
|
"learning_rate": 4.1742989995772867e-07,
|
|
"loss": 0.8627,
|
|
"mean_token_accuracy": 0.7758882611989975,
|
|
"num_tokens": 1345538290.0,
|
|
"step": 13450
|
|
},
|
|
{
|
|
"entropy": 0.89140625,
|
|
"epoch": 1.839609102713046,
|
|
"grad_norm": 0.26944210406159175,
|
|
"learning_rate": 4.13907284768212e-07,
|
|
"loss": 0.8991,
|
|
"mean_token_accuracy": 0.7706076174974441,
|
|
"num_tokens": 1346503361.0,
|
|
"step": 13460
|
|
},
|
|
{
|
|
"entropy": 0.87890625,
|
|
"epoch": 1.8409758764436548,
|
|
"grad_norm": 0.27236063435659064,
|
|
"learning_rate": 4.1038466957869523e-07,
|
|
"loss": 0.8752,
|
|
"mean_token_accuracy": 0.7745659410953522,
|
|
"num_tokens": 1347471362.0,
|
|
"step": 13470
|
|
},
|
|
{
|
|
"entropy": 0.8525390625,
|
|
"epoch": 1.8423426501742637,
|
|
"grad_norm": 0.2877085034257825,
|
|
"learning_rate": 4.068620543891785e-07,
|
|
"loss": 0.8489,
|
|
"mean_token_accuracy": 0.7794237107038497,
|
|
"num_tokens": 1348543692.0,
|
|
"step": 13480
|
|
},
|
|
{
|
|
"entropy": 0.85546875,
|
|
"epoch": 1.8437094239048726,
|
|
"grad_norm": 0.26703740484725147,
|
|
"learning_rate": 4.0333943919966185e-07,
|
|
"loss": 0.8553,
|
|
"mean_token_accuracy": 0.7776215672492981,
|
|
"num_tokens": 1349597476.0,
|
|
"step": 13490
|
|
},
|
|
{
|
|
"entropy": 0.83671875,
|
|
"epoch": 1.8450761976354815,
|
|
"grad_norm": 0.2756659366676005,
|
|
"learning_rate": 3.9981682401014513e-07,
|
|
"loss": 0.8297,
|
|
"mean_token_accuracy": 0.7835846364498138,
|
|
"num_tokens": 1350649328.0,
|
|
"step": 13500
|
|
},
|
|
{
|
|
"entropy": 0.8375,
|
|
"epoch": 1.8464429713660904,
|
|
"grad_norm": 0.271663263841052,
|
|
"learning_rate": 3.9629420882062847e-07,
|
|
"loss": 0.8332,
|
|
"mean_token_accuracy": 0.7818010121583938,
|
|
"num_tokens": 1351647445.0,
|
|
"step": 13510
|
|
},
|
|
{
|
|
"entropy": 0.88125,
|
|
"epoch": 1.8478097450966993,
|
|
"grad_norm": 0.27381589707690607,
|
|
"learning_rate": 3.9277159363111175e-07,
|
|
"loss": 0.8665,
|
|
"mean_token_accuracy": 0.7759990721940995,
|
|
"num_tokens": 1352646669.0,
|
|
"step": 13520
|
|
},
|
|
{
|
|
"entropy": 0.8693359375,
|
|
"epoch": 1.8491765188273082,
|
|
"grad_norm": 0.29997698805820283,
|
|
"learning_rate": 3.892489784415951e-07,
|
|
"loss": 0.8688,
|
|
"mean_token_accuracy": 0.7727596789598465,
|
|
"num_tokens": 1353638717.0,
|
|
"step": 13530
|
|
},
|
|
{
|
|
"entropy": 0.898828125,
|
|
"epoch": 1.850543292557917,
|
|
"grad_norm": 0.3649741290643643,
|
|
"learning_rate": 3.8572636325207837e-07,
|
|
"loss": 0.9023,
|
|
"mean_token_accuracy": 0.7690045237541199,
|
|
"num_tokens": 1354556958.0,
|
|
"step": 13540
|
|
},
|
|
{
|
|
"entropy": 0.8890625,
|
|
"epoch": 1.851910066288526,
|
|
"grad_norm": 0.29189766917599996,
|
|
"learning_rate": 3.822037480625617e-07,
|
|
"loss": 0.8792,
|
|
"mean_token_accuracy": 0.7724256306886673,
|
|
"num_tokens": 1355561263.0,
|
|
"step": 13550
|
|
},
|
|
{
|
|
"entropy": 0.82734375,
|
|
"epoch": 1.8532768400191348,
|
|
"grad_norm": 0.26807459665462013,
|
|
"learning_rate": 3.78681132873045e-07,
|
|
"loss": 0.8162,
|
|
"mean_token_accuracy": 0.7863162994384766,
|
|
"num_tokens": 1356631450.0,
|
|
"step": 13560
|
|
},
|
|
{
|
|
"entropy": 0.8751953125,
|
|
"epoch": 1.8546436137497437,
|
|
"grad_norm": 0.25752251068903925,
|
|
"learning_rate": 3.7515851768352827e-07,
|
|
"loss": 0.8584,
|
|
"mean_token_accuracy": 0.7764760017395019,
|
|
"num_tokens": 1357622176.0,
|
|
"step": 13570
|
|
},
|
|
{
|
|
"entropy": 0.88828125,
|
|
"epoch": 1.8560103874803526,
|
|
"grad_norm": 0.32212801540527053,
|
|
"learning_rate": 3.716359024940116e-07,
|
|
"loss": 0.8913,
|
|
"mean_token_accuracy": 0.7693298906087875,
|
|
"num_tokens": 1358578034.0,
|
|
"step": 13580
|
|
},
|
|
{
|
|
"entropy": 0.877734375,
|
|
"epoch": 1.8573771612109615,
|
|
"grad_norm": 0.2968474200797984,
|
|
"learning_rate": 3.6811328730449484e-07,
|
|
"loss": 0.879,
|
|
"mean_token_accuracy": 0.7723047316074372,
|
|
"num_tokens": 1359604164.0,
|
|
"step": 13590
|
|
},
|
|
{
|
|
"entropy": 0.858203125,
|
|
"epoch": 1.8587439349415704,
|
|
"grad_norm": 0.2812281616117325,
|
|
"learning_rate": 3.645906721149782e-07,
|
|
"loss": 0.8566,
|
|
"mean_token_accuracy": 0.7776282280683517,
|
|
"num_tokens": 1360550654.0,
|
|
"step": 13600
|
|
},
|
|
{
|
|
"entropy": 0.863671875,
|
|
"epoch": 1.8601107086721793,
|
|
"grad_norm": 0.3048218906588462,
|
|
"learning_rate": 3.6106805692546145e-07,
|
|
"loss": 0.8666,
|
|
"mean_token_accuracy": 0.7764086872339249,
|
|
"num_tokens": 1361594445.0,
|
|
"step": 13610
|
|
},
|
|
{
|
|
"entropy": 0.8712890625,
|
|
"epoch": 1.8614774824027882,
|
|
"grad_norm": 0.2785937478899551,
|
|
"learning_rate": 3.575454417359448e-07,
|
|
"loss": 0.8667,
|
|
"mean_token_accuracy": 0.7769855231046676,
|
|
"num_tokens": 1362588624.0,
|
|
"step": 13620
|
|
},
|
|
{
|
|
"entropy": 0.8591796875,
|
|
"epoch": 1.862844256133397,
|
|
"grad_norm": 0.2967541060712264,
|
|
"learning_rate": 3.5402282654642807e-07,
|
|
"loss": 0.8492,
|
|
"mean_token_accuracy": 0.7799870878458023,
|
|
"num_tokens": 1363559822.0,
|
|
"step": 13630
|
|
},
|
|
{
|
|
"entropy": 0.8818359375,
|
|
"epoch": 1.864211029864006,
|
|
"grad_norm": 0.28871369946277553,
|
|
"learning_rate": 3.505002113569114e-07,
|
|
"loss": 0.8834,
|
|
"mean_token_accuracy": 0.7718955785036087,
|
|
"num_tokens": 1364584957.0,
|
|
"step": 13640
|
|
},
|
|
{
|
|
"entropy": 0.8603515625,
|
|
"epoch": 1.865577803594615,
|
|
"grad_norm": 0.33666881356308,
|
|
"learning_rate": 3.469775961673947e-07,
|
|
"loss": 0.8582,
|
|
"mean_token_accuracy": 0.7770739734172821,
|
|
"num_tokens": 1365552739.0,
|
|
"step": 13650
|
|
},
|
|
{
|
|
"entropy": 0.8455078125,
|
|
"epoch": 1.8669445773252238,
|
|
"grad_norm": 0.27582434100711434,
|
|
"learning_rate": 3.43454980977878e-07,
|
|
"loss": 0.8406,
|
|
"mean_token_accuracy": 0.7834489852190017,
|
|
"num_tokens": 1366523444.0,
|
|
"step": 13660
|
|
},
|
|
{
|
|
"entropy": 0.8765625,
|
|
"epoch": 1.8683113510558327,
|
|
"grad_norm": 0.28174427809233826,
|
|
"learning_rate": 3.399323657883613e-07,
|
|
"loss": 0.867,
|
|
"mean_token_accuracy": 0.7742910206317901,
|
|
"num_tokens": 1367508825.0,
|
|
"step": 13670
|
|
},
|
|
{
|
|
"entropy": 0.8794921875,
|
|
"epoch": 1.8696781247864416,
|
|
"grad_norm": 0.27931161010491135,
|
|
"learning_rate": 3.364097505988446e-07,
|
|
"loss": 0.8791,
|
|
"mean_token_accuracy": 0.7734599173069,
|
|
"num_tokens": 1368544707.0,
|
|
"step": 13680
|
|
},
|
|
{
|
|
"entropy": 0.86328125,
|
|
"epoch": 1.8710448985170505,
|
|
"grad_norm": 0.30273581988275333,
|
|
"learning_rate": 3.328871354093279e-07,
|
|
"loss": 0.8612,
|
|
"mean_token_accuracy": 0.775649219751358,
|
|
"num_tokens": 1369580430.0,
|
|
"step": 13690
|
|
},
|
|
{
|
|
"entropy": 0.8845703125,
|
|
"epoch": 1.8724116722476594,
|
|
"grad_norm": 0.285445363683713,
|
|
"learning_rate": 3.293645202198112e-07,
|
|
"loss": 0.8875,
|
|
"mean_token_accuracy": 0.7723607778549194,
|
|
"num_tokens": 1370581412.0,
|
|
"step": 13700
|
|
},
|
|
{
|
|
"entropy": 0.8890625,
|
|
"epoch": 1.8737784459782683,
|
|
"grad_norm": 0.2586850519384564,
|
|
"learning_rate": 3.2584190503029454e-07,
|
|
"loss": 0.8851,
|
|
"mean_token_accuracy": 0.7705006241798401,
|
|
"num_tokens": 1371543600.0,
|
|
"step": 13710
|
|
},
|
|
{
|
|
"entropy": 0.849609375,
|
|
"epoch": 1.8751452197088772,
|
|
"grad_norm": 0.2910474011537201,
|
|
"learning_rate": 3.223192898407778e-07,
|
|
"loss": 0.8499,
|
|
"mean_token_accuracy": 0.7813055396080018,
|
|
"num_tokens": 1372552489.0,
|
|
"step": 13720
|
|
},
|
|
{
|
|
"entropy": 0.8677734375,
|
|
"epoch": 1.876511993439486,
|
|
"grad_norm": 0.29298167221527316,
|
|
"learning_rate": 3.1879667465126116e-07,
|
|
"loss": 0.8615,
|
|
"mean_token_accuracy": 0.7784838020801544,
|
|
"num_tokens": 1373567220.0,
|
|
"step": 13730
|
|
},
|
|
{
|
|
"entropy": 0.844140625,
|
|
"epoch": 1.877878767170095,
|
|
"grad_norm": 0.25343987187681555,
|
|
"learning_rate": 3.152740594617444e-07,
|
|
"loss": 0.8423,
|
|
"mean_token_accuracy": 0.782135272026062,
|
|
"num_tokens": 1374630605.0,
|
|
"step": 13740
|
|
},
|
|
{
|
|
"entropy": 0.874609375,
|
|
"epoch": 1.8792455409007038,
|
|
"grad_norm": 0.2649718203179233,
|
|
"learning_rate": 3.117514442722277e-07,
|
|
"loss": 0.8657,
|
|
"mean_token_accuracy": 0.7756816476583481,
|
|
"num_tokens": 1375662535.0,
|
|
"step": 13750
|
|
},
|
|
{
|
|
"entropy": 0.7912109375,
|
|
"epoch": 1.8806123146313127,
|
|
"grad_norm": 0.26152560890344884,
|
|
"learning_rate": 3.08228829082711e-07,
|
|
"loss": 0.7916,
|
|
"mean_token_accuracy": 0.7913625746965408,
|
|
"num_tokens": 1376719597.0,
|
|
"step": 13760
|
|
},
|
|
{
|
|
"entropy": 0.866796875,
|
|
"epoch": 1.8819790883619216,
|
|
"grad_norm": 0.2986463850686021,
|
|
"learning_rate": 3.0470621389319434e-07,
|
|
"loss": 0.8602,
|
|
"mean_token_accuracy": 0.7770053416490554,
|
|
"num_tokens": 1377692863.0,
|
|
"step": 13770
|
|
},
|
|
{
|
|
"entropy": 0.85546875,
|
|
"epoch": 1.8833458620925305,
|
|
"grad_norm": 0.2586623704712727,
|
|
"learning_rate": 3.0118359870367763e-07,
|
|
"loss": 0.8595,
|
|
"mean_token_accuracy": 0.7793031066656113,
|
|
"num_tokens": 1378637635.0,
|
|
"step": 13780
|
|
},
|
|
{
|
|
"entropy": 0.827734375,
|
|
"epoch": 1.8847126358231394,
|
|
"grad_norm": 0.27449227498049544,
|
|
"learning_rate": 2.9766098351416096e-07,
|
|
"loss": 0.8224,
|
|
"mean_token_accuracy": 0.7851754695177078,
|
|
"num_tokens": 1379655490.0,
|
|
"step": 13790
|
|
},
|
|
{
|
|
"entropy": 0.8634765625,
|
|
"epoch": 1.8860794095537483,
|
|
"grad_norm": 0.25673246892748114,
|
|
"learning_rate": 2.9413836832464424e-07,
|
|
"loss": 0.8582,
|
|
"mean_token_accuracy": 0.7766181319952011,
|
|
"num_tokens": 1380686194.0,
|
|
"step": 13800
|
|
},
|
|
{
|
|
"entropy": 0.9001953125,
|
|
"epoch": 1.8874461832843572,
|
|
"grad_norm": 0.2979819862640686,
|
|
"learning_rate": 2.906157531351276e-07,
|
|
"loss": 0.9078,
|
|
"mean_token_accuracy": 0.7699101775884628,
|
|
"num_tokens": 1381695289.0,
|
|
"step": 13810
|
|
},
|
|
{
|
|
"entropy": 0.8552734375,
|
|
"epoch": 1.888812957014966,
|
|
"grad_norm": 0.3091107342333432,
|
|
"learning_rate": 2.8709313794561086e-07,
|
|
"loss": 0.8465,
|
|
"mean_token_accuracy": 0.7797375500202179,
|
|
"num_tokens": 1382721614.0,
|
|
"step": 13820
|
|
},
|
|
{
|
|
"entropy": 0.8740234375,
|
|
"epoch": 1.890179730745575,
|
|
"grad_norm": 0.3129035305283891,
|
|
"learning_rate": 2.8357052275609415e-07,
|
|
"loss": 0.8652,
|
|
"mean_token_accuracy": 0.7737299710512161,
|
|
"num_tokens": 1383676808.0,
|
|
"step": 13830
|
|
},
|
|
{
|
|
"entropy": 0.9060546875,
|
|
"epoch": 1.8915465044761839,
|
|
"grad_norm": 0.30510742805098945,
|
|
"learning_rate": 2.8004790756657743e-07,
|
|
"loss": 0.9055,
|
|
"mean_token_accuracy": 0.770076510310173,
|
|
"num_tokens": 1384669165.0,
|
|
"step": 13840
|
|
},
|
|
{
|
|
"entropy": 0.8611328125,
|
|
"epoch": 1.8929132782067928,
|
|
"grad_norm": 0.24965460207993626,
|
|
"learning_rate": 2.7652529237706076e-07,
|
|
"loss": 0.837,
|
|
"mean_token_accuracy": 0.7807908833026886,
|
|
"num_tokens": 1385658422.0,
|
|
"step": 13850
|
|
},
|
|
{
|
|
"entropy": 0.83828125,
|
|
"epoch": 1.8942800519374017,
|
|
"grad_norm": 0.27226271277007336,
|
|
"learning_rate": 2.7300267718754405e-07,
|
|
"loss": 0.8373,
|
|
"mean_token_accuracy": 0.782850256562233,
|
|
"num_tokens": 1386627672.0,
|
|
"step": 13860
|
|
},
|
|
{
|
|
"entropy": 0.7998046875,
|
|
"epoch": 1.8956468256680106,
|
|
"grad_norm": 0.27827213939883066,
|
|
"learning_rate": 2.694800619980274e-07,
|
|
"loss": 0.7988,
|
|
"mean_token_accuracy": 0.7894185036420822,
|
|
"num_tokens": 1387628052.0,
|
|
"step": 13870
|
|
},
|
|
{
|
|
"entropy": 0.8466796875,
|
|
"epoch": 1.8970135993986195,
|
|
"grad_norm": 0.2927269726929299,
|
|
"learning_rate": 2.6595744680851066e-07,
|
|
"loss": 0.8407,
|
|
"mean_token_accuracy": 0.7809716582298278,
|
|
"num_tokens": 1388669454.0,
|
|
"step": 13880
|
|
},
|
|
{
|
|
"entropy": 0.8763671875,
|
|
"epoch": 1.8983803731292284,
|
|
"grad_norm": 0.2935418986535845,
|
|
"learning_rate": 2.6243483161899395e-07,
|
|
"loss": 0.882,
|
|
"mean_token_accuracy": 0.7722764641046524,
|
|
"num_tokens": 1389648608.0,
|
|
"step": 13890
|
|
},
|
|
{
|
|
"entropy": 0.8587890625,
|
|
"epoch": 1.8997471468598373,
|
|
"grad_norm": 0.2737697997227319,
|
|
"learning_rate": 2.589122164294773e-07,
|
|
"loss": 0.8545,
|
|
"mean_token_accuracy": 0.7776701003313065,
|
|
"num_tokens": 1390701471.0,
|
|
"step": 13900
|
|
},
|
|
{
|
|
"entropy": 0.851171875,
|
|
"epoch": 1.9011139205904461,
|
|
"grad_norm": 0.3789819329897561,
|
|
"learning_rate": 2.5538960123996056e-07,
|
|
"loss": 0.8553,
|
|
"mean_token_accuracy": 0.7789779424667358,
|
|
"num_tokens": 1391701108.0,
|
|
"step": 13910
|
|
},
|
|
{
|
|
"entropy": 0.8609375,
|
|
"epoch": 1.902480694321055,
|
|
"grad_norm": 0.2935279692869397,
|
|
"learning_rate": 2.518669860504439e-07,
|
|
"loss": 0.8604,
|
|
"mean_token_accuracy": 0.7763250380754471,
|
|
"num_tokens": 1392659076.0,
|
|
"step": 13920
|
|
},
|
|
{
|
|
"entropy": 0.8638671875,
|
|
"epoch": 1.903847468051664,
|
|
"grad_norm": 0.30750348817473255,
|
|
"learning_rate": 2.483443708609272e-07,
|
|
"loss": 0.8631,
|
|
"mean_token_accuracy": 0.7758825182914734,
|
|
"num_tokens": 1393642094.0,
|
|
"step": 13930
|
|
},
|
|
{
|
|
"entropy": 0.8375,
|
|
"epoch": 1.9052142417822728,
|
|
"grad_norm": 0.32917005791143467,
|
|
"learning_rate": 2.4482175567141046e-07,
|
|
"loss": 0.8416,
|
|
"mean_token_accuracy": 0.7809051424264908,
|
|
"num_tokens": 1394661359.0,
|
|
"step": 13940
|
|
},
|
|
{
|
|
"entropy": 0.8765625,
|
|
"epoch": 1.9065810155128817,
|
|
"grad_norm": 0.24501099595160908,
|
|
"learning_rate": 2.4129914048189375e-07,
|
|
"loss": 0.8717,
|
|
"mean_token_accuracy": 0.7747946888208389,
|
|
"num_tokens": 1395713458.0,
|
|
"step": 13950
|
|
},
|
|
{
|
|
"entropy": 0.87890625,
|
|
"epoch": 1.9079477892434906,
|
|
"grad_norm": 0.2624777211035248,
|
|
"learning_rate": 2.3777652529237708e-07,
|
|
"loss": 0.8817,
|
|
"mean_token_accuracy": 0.7745908945798874,
|
|
"num_tokens": 1396733047.0,
|
|
"step": 13960
|
|
},
|
|
{
|
|
"entropy": 0.8916015625,
|
|
"epoch": 1.9093145629740995,
|
|
"grad_norm": 0.28284536880427164,
|
|
"learning_rate": 2.342539101028604e-07,
|
|
"loss": 0.8947,
|
|
"mean_token_accuracy": 0.7701242476701736,
|
|
"num_tokens": 1397783144.0,
|
|
"step": 13970
|
|
},
|
|
{
|
|
"entropy": 0.8916015625,
|
|
"epoch": 1.9106813367047084,
|
|
"grad_norm": 0.30979883712977546,
|
|
"learning_rate": 2.307312949133437e-07,
|
|
"loss": 0.8969,
|
|
"mean_token_accuracy": 0.7690603703260421,
|
|
"num_tokens": 1398770137.0,
|
|
"step": 13980
|
|
},
|
|
{
|
|
"entropy": 0.8435546875,
|
|
"epoch": 1.9120481104353173,
|
|
"grad_norm": 0.27572866198234586,
|
|
"learning_rate": 2.2720867972382698e-07,
|
|
"loss": 0.8481,
|
|
"mean_token_accuracy": 0.7783573657274246,
|
|
"num_tokens": 1399778288.0,
|
|
"step": 13990
|
|
},
|
|
{
|
|
"entropy": 0.8875,
|
|
"epoch": 1.9134148841659262,
|
|
"grad_norm": 0.27582099230687407,
|
|
"learning_rate": 2.236860645343103e-07,
|
|
"loss": 0.8764,
|
|
"mean_token_accuracy": 0.7734238117933273,
|
|
"num_tokens": 1400753880.0,
|
|
"step": 14000
|
|
},
|
|
{
|
|
"entropy": 0.831640625,
|
|
"epoch": 1.914781657896535,
|
|
"grad_norm": 0.33949731455812043,
|
|
"learning_rate": 2.201634493447936e-07,
|
|
"loss": 0.8164,
|
|
"mean_token_accuracy": 0.7854899495840073,
|
|
"num_tokens": 1401736123.0,
|
|
"step": 14010
|
|
},
|
|
{
|
|
"entropy": 0.8541015625,
|
|
"epoch": 1.916148431627144,
|
|
"grad_norm": 0.2802589557948454,
|
|
"learning_rate": 2.166408341552769e-07,
|
|
"loss": 0.8541,
|
|
"mean_token_accuracy": 0.7768935203552246,
|
|
"num_tokens": 1402717620.0,
|
|
"step": 14020
|
|
},
|
|
{
|
|
"entropy": 0.8462890625,
|
|
"epoch": 1.9175152053577529,
|
|
"grad_norm": 0.2556949067326232,
|
|
"learning_rate": 2.131182189657602e-07,
|
|
"loss": 0.8349,
|
|
"mean_token_accuracy": 0.7815765708684921,
|
|
"num_tokens": 1403701055.0,
|
|
"step": 14030
|
|
},
|
|
{
|
|
"entropy": 0.85546875,
|
|
"epoch": 1.9188819790883618,
|
|
"grad_norm": 0.29325473473971303,
|
|
"learning_rate": 2.095956037762435e-07,
|
|
"loss": 0.8557,
|
|
"mean_token_accuracy": 0.7772633314132691,
|
|
"num_tokens": 1404698236.0,
|
|
"step": 14040
|
|
},
|
|
{
|
|
"entropy": 0.8720703125,
|
|
"epoch": 1.9202487528189707,
|
|
"grad_norm": 0.26916948252092865,
|
|
"learning_rate": 2.0607298858672678e-07,
|
|
"loss": 0.8729,
|
|
"mean_token_accuracy": 0.7766596019268036,
|
|
"num_tokens": 1405695907.0,
|
|
"step": 14050
|
|
},
|
|
{
|
|
"entropy": 0.8529296875,
|
|
"epoch": 1.9216155265495796,
|
|
"grad_norm": 0.31554098961075805,
|
|
"learning_rate": 2.025503733972101e-07,
|
|
"loss": 0.858,
|
|
"mean_token_accuracy": 0.7783398002386093,
|
|
"num_tokens": 1406712450.0,
|
|
"step": 14060
|
|
},
|
|
{
|
|
"entropy": 0.89296875,
|
|
"epoch": 1.9229823002801885,
|
|
"grad_norm": 0.30413111711886287,
|
|
"learning_rate": 1.990277582076934e-07,
|
|
"loss": 0.8932,
|
|
"mean_token_accuracy": 0.7710342288017273,
|
|
"num_tokens": 1407746334.0,
|
|
"step": 14070
|
|
},
|
|
{
|
|
"entropy": 0.865234375,
|
|
"epoch": 1.9243490740107974,
|
|
"grad_norm": 0.2920469338430756,
|
|
"learning_rate": 1.955051430181767e-07,
|
|
"loss": 0.8491,
|
|
"mean_token_accuracy": 0.7775528639554977,
|
|
"num_tokens": 1408679942.0,
|
|
"step": 14080
|
|
},
|
|
{
|
|
"entropy": 0.85078125,
|
|
"epoch": 1.9257158477414063,
|
|
"grad_norm": 0.3036132544490115,
|
|
"learning_rate": 1.9198252782866002e-07,
|
|
"loss": 0.8449,
|
|
"mean_token_accuracy": 0.7816273897886277,
|
|
"num_tokens": 1409668536.0,
|
|
"step": 14090
|
|
},
|
|
{
|
|
"entropy": 0.87421875,
|
|
"epoch": 1.9270826214720151,
|
|
"grad_norm": 0.2752528618298979,
|
|
"learning_rate": 1.8845991263914333e-07,
|
|
"loss": 0.8694,
|
|
"mean_token_accuracy": 0.7767285943031311,
|
|
"num_tokens": 1410663341.0,
|
|
"step": 14100
|
|
},
|
|
{
|
|
"entropy": 0.883984375,
|
|
"epoch": 1.928449395202624,
|
|
"grad_norm": 0.3028286086842123,
|
|
"learning_rate": 1.8493729744962664e-07,
|
|
"loss": 0.8766,
|
|
"mean_token_accuracy": 0.7731164425611496,
|
|
"num_tokens": 1411582681.0,
|
|
"step": 14110
|
|
},
|
|
{
|
|
"entropy": 0.915625,
|
|
"epoch": 1.9298161689332332,
|
|
"grad_norm": 0.2719279482806134,
|
|
"learning_rate": 1.8141468226010995e-07,
|
|
"loss": 0.9159,
|
|
"mean_token_accuracy": 0.7667460680007935,
|
|
"num_tokens": 1412599119.0,
|
|
"step": 14120
|
|
},
|
|
{
|
|
"entropy": 0.8376953125,
|
|
"epoch": 1.931182942663842,
|
|
"grad_norm": 0.299043158108045,
|
|
"learning_rate": 1.778920670705932e-07,
|
|
"loss": 0.8302,
|
|
"mean_token_accuracy": 0.7837609499692917,
|
|
"num_tokens": 1413601008.0,
|
|
"step": 14130
|
|
},
|
|
{
|
|
"entropy": 0.8505859375,
|
|
"epoch": 1.932549716394451,
|
|
"grad_norm": 0.2560298138420745,
|
|
"learning_rate": 1.743694518810765e-07,
|
|
"loss": 0.8447,
|
|
"mean_token_accuracy": 0.779724907875061,
|
|
"num_tokens": 1414627353.0,
|
|
"step": 14140
|
|
},
|
|
{
|
|
"entropy": 0.8427734375,
|
|
"epoch": 1.9339164901250598,
|
|
"grad_norm": 0.2713282226817263,
|
|
"learning_rate": 1.7084683669155982e-07,
|
|
"loss": 0.8277,
|
|
"mean_token_accuracy": 0.783468097448349,
|
|
"num_tokens": 1415631253.0,
|
|
"step": 14150
|
|
},
|
|
{
|
|
"entropy": 0.8853515625,
|
|
"epoch": 1.9352832638556687,
|
|
"grad_norm": 0.29640679702742617,
|
|
"learning_rate": 1.6732422150204313e-07,
|
|
"loss": 0.8798,
|
|
"mean_token_accuracy": 0.772931033372879,
|
|
"num_tokens": 1416603580.0,
|
|
"step": 14160
|
|
},
|
|
{
|
|
"entropy": 0.832421875,
|
|
"epoch": 1.9366500375862776,
|
|
"grad_norm": 0.25840088391663996,
|
|
"learning_rate": 1.6380160631252644e-07,
|
|
"loss": 0.8349,
|
|
"mean_token_accuracy": 0.7833683133125305,
|
|
"num_tokens": 1417661996.0,
|
|
"step": 14170
|
|
},
|
|
{
|
|
"entropy": 0.872265625,
|
|
"epoch": 1.9380168113168865,
|
|
"grad_norm": 0.2684568590232537,
|
|
"learning_rate": 1.6027899112300975e-07,
|
|
"loss": 0.8609,
|
|
"mean_token_accuracy": 0.7771875262260437,
|
|
"num_tokens": 1418652273.0,
|
|
"step": 14180
|
|
},
|
|
{
|
|
"entropy": 0.8388671875,
|
|
"epoch": 1.9393835850474954,
|
|
"grad_norm": 0.2734987998235929,
|
|
"learning_rate": 1.5675637593349303e-07,
|
|
"loss": 0.8272,
|
|
"mean_token_accuracy": 0.7844835430383682,
|
|
"num_tokens": 1419678719.0,
|
|
"step": 14190
|
|
},
|
|
{
|
|
"entropy": 0.8546875,
|
|
"epoch": 1.9407503587781043,
|
|
"grad_norm": 0.3369513568568491,
|
|
"learning_rate": 1.5323376074397634e-07,
|
|
"loss": 0.8412,
|
|
"mean_token_accuracy": 0.778885343670845,
|
|
"num_tokens": 1420698946.0,
|
|
"step": 14200
|
|
},
|
|
{
|
|
"entropy": 0.9291015625,
|
|
"epoch": 1.9421171325087132,
|
|
"grad_norm": 0.32420933792731205,
|
|
"learning_rate": 1.4971114555445965e-07,
|
|
"loss": 0.9308,
|
|
"mean_token_accuracy": 0.7626196801662445,
|
|
"num_tokens": 1421712641.0,
|
|
"step": 14210
|
|
},
|
|
{
|
|
"entropy": 0.8853515625,
|
|
"epoch": 1.943483906239322,
|
|
"grad_norm": 0.28103797307788775,
|
|
"learning_rate": 1.4618853036494293e-07,
|
|
"loss": 0.8921,
|
|
"mean_token_accuracy": 0.770270150899887,
|
|
"num_tokens": 1422695791.0,
|
|
"step": 14220
|
|
},
|
|
{
|
|
"entropy": 0.86953125,
|
|
"epoch": 1.944850679969931,
|
|
"grad_norm": 0.27932825916633824,
|
|
"learning_rate": 1.4266591517542624e-07,
|
|
"loss": 0.8731,
|
|
"mean_token_accuracy": 0.773793363571167,
|
|
"num_tokens": 1423674664.0,
|
|
"step": 14230
|
|
},
|
|
{
|
|
"entropy": 0.85234375,
|
|
"epoch": 1.94621745370054,
|
|
"grad_norm": 0.2604217316751798,
|
|
"learning_rate": 1.3914329998590955e-07,
|
|
"loss": 0.8442,
|
|
"mean_token_accuracy": 0.7810238510370254,
|
|
"num_tokens": 1424607105.0,
|
|
"step": 14240
|
|
},
|
|
{
|
|
"entropy": 0.86640625,
|
|
"epoch": 1.9475842274311488,
|
|
"grad_norm": 0.2781507219968767,
|
|
"learning_rate": 1.3562068479639286e-07,
|
|
"loss": 0.8557,
|
|
"mean_token_accuracy": 0.7777597039937973,
|
|
"num_tokens": 1425649722.0,
|
|
"step": 14250
|
|
},
|
|
{
|
|
"entropy": 0.8296875,
|
|
"epoch": 1.9489510011617577,
|
|
"grad_norm": 0.27443768972567517,
|
|
"learning_rate": 1.3209806960687614e-07,
|
|
"loss": 0.8221,
|
|
"mean_token_accuracy": 0.7850887954235077,
|
|
"num_tokens": 1426614286.0,
|
|
"step": 14260
|
|
},
|
|
{
|
|
"entropy": 0.8544921875,
|
|
"epoch": 1.9503177748923666,
|
|
"grad_norm": 0.301876947164387,
|
|
"learning_rate": 1.2857545441735945e-07,
|
|
"loss": 0.8559,
|
|
"mean_token_accuracy": 0.7781076908111573,
|
|
"num_tokens": 1427602537.0,
|
|
"step": 14270
|
|
},
|
|
{
|
|
"entropy": 0.878125,
|
|
"epoch": 1.9516845486229755,
|
|
"grad_norm": 0.3032512296453283,
|
|
"learning_rate": 1.2505283922784276e-07,
|
|
"loss": 0.8722,
|
|
"mean_token_accuracy": 0.7755087643861771,
|
|
"num_tokens": 1428633543.0,
|
|
"step": 14280
|
|
},
|
|
{
|
|
"entropy": 0.844140625,
|
|
"epoch": 1.9530513223535844,
|
|
"grad_norm": 0.2657424357752277,
|
|
"learning_rate": 1.2153022403832607e-07,
|
|
"loss": 0.8425,
|
|
"mean_token_accuracy": 0.780802184343338,
|
|
"num_tokens": 1429646323.0,
|
|
"step": 14290
|
|
},
|
|
{
|
|
"entropy": 0.8619140625,
|
|
"epoch": 1.9544180960841933,
|
|
"grad_norm": 0.30855393299013917,
|
|
"learning_rate": 1.1800760884880937e-07,
|
|
"loss": 0.8634,
|
|
"mean_token_accuracy": 0.7747360229492187,
|
|
"num_tokens": 1430690918.0,
|
|
"step": 14300
|
|
},
|
|
{
|
|
"entropy": 0.868359375,
|
|
"epoch": 1.9557848698148022,
|
|
"grad_norm": 0.276524262039025,
|
|
"learning_rate": 1.1448499365929266e-07,
|
|
"loss": 0.8657,
|
|
"mean_token_accuracy": 0.775782236456871,
|
|
"num_tokens": 1431701637.0,
|
|
"step": 14310
|
|
},
|
|
{
|
|
"entropy": 0.862109375,
|
|
"epoch": 1.957151643545411,
|
|
"grad_norm": 0.28678999706934727,
|
|
"learning_rate": 1.1096237846977597e-07,
|
|
"loss": 0.8524,
|
|
"mean_token_accuracy": 0.7790747135877609,
|
|
"num_tokens": 1432695345.0,
|
|
"step": 14320
|
|
},
|
|
{
|
|
"entropy": 0.86328125,
|
|
"epoch": 1.95851841727602,
|
|
"grad_norm": 0.292283065691729,
|
|
"learning_rate": 1.0743976328025928e-07,
|
|
"loss": 0.8501,
|
|
"mean_token_accuracy": 0.7765097469091415,
|
|
"num_tokens": 1433721683.0,
|
|
"step": 14330
|
|
},
|
|
{
|
|
"entropy": 0.8181640625,
|
|
"epoch": 1.9598851910066288,
|
|
"grad_norm": 0.3053010805458119,
|
|
"learning_rate": 1.0391714809074258e-07,
|
|
"loss": 0.8162,
|
|
"mean_token_accuracy": 0.7840531408786774,
|
|
"num_tokens": 1434741958.0,
|
|
"step": 14340
|
|
},
|
|
{
|
|
"entropy": 0.8564453125,
|
|
"epoch": 1.9612519647372377,
|
|
"grad_norm": 0.2821108113694897,
|
|
"learning_rate": 1.0039453290122588e-07,
|
|
"loss": 0.8589,
|
|
"mean_token_accuracy": 0.7749951153993606,
|
|
"num_tokens": 1435738881.0,
|
|
"step": 14350
|
|
},
|
|
{
|
|
"entropy": 0.83125,
|
|
"epoch": 1.9626187384678466,
|
|
"grad_norm": 0.29913055894050045,
|
|
"learning_rate": 9.687191771170918e-08,
|
|
"loss": 0.8271,
|
|
"mean_token_accuracy": 0.784439179301262,
|
|
"num_tokens": 1436722777.0,
|
|
"step": 14360
|
|
},
|
|
{
|
|
"entropy": 0.890234375,
|
|
"epoch": 1.9639855121984555,
|
|
"grad_norm": 0.29284318068569765,
|
|
"learning_rate": 9.334930252219248e-08,
|
|
"loss": 0.8928,
|
|
"mean_token_accuracy": 0.7714187949895859,
|
|
"num_tokens": 1437762126.0,
|
|
"step": 14370
|
|
},
|
|
{
|
|
"entropy": 0.858203125,
|
|
"epoch": 1.9653522859290644,
|
|
"grad_norm": 0.26362163463214766,
|
|
"learning_rate": 8.982668733267578e-08,
|
|
"loss": 0.8615,
|
|
"mean_token_accuracy": 0.7767615616321564,
|
|
"num_tokens": 1438775874.0,
|
|
"step": 14380
|
|
},
|
|
{
|
|
"entropy": 0.81796875,
|
|
"epoch": 1.9667190596596733,
|
|
"grad_norm": 0.2692424926255788,
|
|
"learning_rate": 8.630407214315909e-08,
|
|
"loss": 0.8165,
|
|
"mean_token_accuracy": 0.785688915848732,
|
|
"num_tokens": 1439781057.0,
|
|
"step": 14390
|
|
},
|
|
{
|
|
"entropy": 0.8765625,
|
|
"epoch": 1.9680858333902822,
|
|
"grad_norm": 0.28107771243176083,
|
|
"learning_rate": 8.27814569536424e-08,
|
|
"loss": 0.8706,
|
|
"mean_token_accuracy": 0.7760464668273925,
|
|
"num_tokens": 1440796500.0,
|
|
"step": 14400
|
|
},
|
|
{
|
|
"entropy": 0.8865234375,
|
|
"epoch": 1.969452607120891,
|
|
"grad_norm": 0.31779291406620247,
|
|
"learning_rate": 7.925884176412568e-08,
|
|
"loss": 0.8966,
|
|
"mean_token_accuracy": 0.7697353184223175,
|
|
"num_tokens": 1441773687.0,
|
|
"step": 14410
|
|
},
|
|
{
|
|
"entropy": 0.8857421875,
|
|
"epoch": 1.9708193808515,
|
|
"grad_norm": 0.30583139681323396,
|
|
"learning_rate": 7.573622657460899e-08,
|
|
"loss": 0.8912,
|
|
"mean_token_accuracy": 0.771193528175354,
|
|
"num_tokens": 1442759617.0,
|
|
"step": 14420
|
|
},
|
|
{
|
|
"entropy": 0.8431640625,
|
|
"epoch": 1.9721861545821089,
|
|
"grad_norm": 0.3032057014510316,
|
|
"learning_rate": 7.22136113850923e-08,
|
|
"loss": 0.8365,
|
|
"mean_token_accuracy": 0.7830202430486679,
|
|
"num_tokens": 1443765717.0,
|
|
"step": 14430
|
|
},
|
|
{
|
|
"entropy": 0.841796875,
|
|
"epoch": 1.9735529283127178,
|
|
"grad_norm": 0.2855501669733693,
|
|
"learning_rate": 6.869099619557561e-08,
|
|
"loss": 0.8348,
|
|
"mean_token_accuracy": 0.781060916185379,
|
|
"num_tokens": 1444773623.0,
|
|
"step": 14440
|
|
},
|
|
{
|
|
"entropy": 0.847265625,
|
|
"epoch": 1.9749197020433267,
|
|
"grad_norm": 0.3569142570669626,
|
|
"learning_rate": 6.51683810060589e-08,
|
|
"loss": 0.8515,
|
|
"mean_token_accuracy": 0.7792166978120804,
|
|
"num_tokens": 1445817445.0,
|
|
"step": 14450
|
|
},
|
|
{
|
|
"entropy": 0.877734375,
|
|
"epoch": 1.9762864757739358,
|
|
"grad_norm": 0.2953749969912388,
|
|
"learning_rate": 6.164576581654221e-08,
|
|
"loss": 0.8723,
|
|
"mean_token_accuracy": 0.774108162522316,
|
|
"num_tokens": 1446775538.0,
|
|
"step": 14460
|
|
},
|
|
{
|
|
"entropy": 0.8513671875,
|
|
"epoch": 1.9776532495045447,
|
|
"grad_norm": 0.2938371755139225,
|
|
"learning_rate": 5.8123150627025515e-08,
|
|
"loss": 0.8434,
|
|
"mean_token_accuracy": 0.7798854798078537,
|
|
"num_tokens": 1447746801.0,
|
|
"step": 14470
|
|
},
|
|
{
|
|
"entropy": 0.85234375,
|
|
"epoch": 1.9790200232351536,
|
|
"grad_norm": 0.28707899413589677,
|
|
"learning_rate": 5.460053543750881e-08,
|
|
"loss": 0.8513,
|
|
"mean_token_accuracy": 0.778857734799385,
|
|
"num_tokens": 1448668656.0,
|
|
"step": 14480
|
|
},
|
|
{
|
|
"entropy": 0.89140625,
|
|
"epoch": 1.9803867969657625,
|
|
"grad_norm": 0.2758184455670959,
|
|
"learning_rate": 5.107792024799211e-08,
|
|
"loss": 0.8947,
|
|
"mean_token_accuracy": 0.7700428247451783,
|
|
"num_tokens": 1449668808.0,
|
|
"step": 14490
|
|
},
|
|
{
|
|
"entropy": 0.90234375,
|
|
"epoch": 1.9817535706963714,
|
|
"grad_norm": 0.4285785263184179,
|
|
"learning_rate": 4.7555305058475415e-08,
|
|
"loss": 0.9012,
|
|
"mean_token_accuracy": 0.7675414949655532,
|
|
"num_tokens": 1450628279.0,
|
|
"step": 14500
|
|
},
|
|
{
|
|
"entropy": 0.8732421875,
|
|
"epoch": 1.9831203444269803,
|
|
"grad_norm": 0.3134855643748628,
|
|
"learning_rate": 4.403268986895872e-08,
|
|
"loss": 0.8868,
|
|
"mean_token_accuracy": 0.7709311813116073,
|
|
"num_tokens": 1451678331.0,
|
|
"step": 14510
|
|
},
|
|
{
|
|
"entropy": 0.880859375,
|
|
"epoch": 1.9844871181575892,
|
|
"grad_norm": 0.28060483180354906,
|
|
"learning_rate": 4.0510074679442026e-08,
|
|
"loss": 0.8786,
|
|
"mean_token_accuracy": 0.7716478049755097,
|
|
"num_tokens": 1452675629.0,
|
|
"step": 14520
|
|
},
|
|
{
|
|
"entropy": 0.859765625,
|
|
"epoch": 1.985853891888198,
|
|
"grad_norm": 0.2999776695274254,
|
|
"learning_rate": 3.698745948992532e-08,
|
|
"loss": 0.8539,
|
|
"mean_token_accuracy": 0.7791115492582321,
|
|
"num_tokens": 1453616235.0,
|
|
"step": 14530
|
|
},
|
|
{
|
|
"entropy": 0.8494140625,
|
|
"epoch": 1.987220665618807,
|
|
"grad_norm": 0.29202162191149333,
|
|
"learning_rate": 3.3464844300408624e-08,
|
|
"loss": 0.8436,
|
|
"mean_token_accuracy": 0.7793088674545288,
|
|
"num_tokens": 1454562236.0,
|
|
"step": 14540
|
|
},
|
|
{
|
|
"entropy": 0.8435546875,
|
|
"epoch": 1.9885874393494158,
|
|
"grad_norm": 0.2763912018809915,
|
|
"learning_rate": 2.994222911089193e-08,
|
|
"loss": 0.8323,
|
|
"mean_token_accuracy": 0.7838230967521668,
|
|
"num_tokens": 1455572631.0,
|
|
"step": 14550
|
|
},
|
|
{
|
|
"entropy": 0.8896484375,
|
|
"epoch": 1.9899542130800247,
|
|
"grad_norm": 0.28260075774062365,
|
|
"learning_rate": 2.6419613921375232e-08,
|
|
"loss": 0.8857,
|
|
"mean_token_accuracy": 0.7731207460165024,
|
|
"num_tokens": 1456593277.0,
|
|
"step": 14560
|
|
},
|
|
{
|
|
"entropy": 0.8314453125,
|
|
"epoch": 1.9913209868106336,
|
|
"grad_norm": 0.2897930830461894,
|
|
"learning_rate": 2.289699873185853e-08,
|
|
"loss": 0.8325,
|
|
"mean_token_accuracy": 0.7812548220157624,
|
|
"num_tokens": 1457604095.0,
|
|
"step": 14570
|
|
},
|
|
{
|
|
"entropy": 0.9197265625,
|
|
"epoch": 1.9926877605412425,
|
|
"grad_norm": 0.2715687465719341,
|
|
"learning_rate": 1.9374383542341837e-08,
|
|
"loss": 0.9167,
|
|
"mean_token_accuracy": 0.7634588927030563,
|
|
"num_tokens": 1458553896.0,
|
|
"step": 14580
|
|
},
|
|
{
|
|
"entropy": 0.8306640625,
|
|
"epoch": 1.9940545342718514,
|
|
"grad_norm": 0.28012593762376287,
|
|
"learning_rate": 1.585176835282514e-08,
|
|
"loss": 0.8152,
|
|
"mean_token_accuracy": 0.7851119339466095,
|
|
"num_tokens": 1459560082.0,
|
|
"step": 14590
|
|
},
|
|
{
|
|
"entropy": 0.898046875,
|
|
"epoch": 1.9954213080024603,
|
|
"grad_norm": 0.28535789744149487,
|
|
"learning_rate": 1.2329153163308442e-08,
|
|
"loss": 0.8908,
|
|
"mean_token_accuracy": 0.7706924676895142,
|
|
"num_tokens": 1460582129.0,
|
|
"step": 14600
|
|
},
|
|
{
|
|
"entropy": 0.8669921875,
|
|
"epoch": 1.9967880817330692,
|
|
"grad_norm": 0.28902737455998245,
|
|
"learning_rate": 8.806537973791744e-09,
|
|
"loss": 0.8604,
|
|
"mean_token_accuracy": 0.7766288936138153,
|
|
"num_tokens": 1461619051.0,
|
|
"step": 14610
|
|
},
|
|
{
|
|
"entropy": 0.843359375,
|
|
"epoch": 1.998154855463678,
|
|
"grad_norm": 0.2644613918371533,
|
|
"learning_rate": 5.2839227842750465e-09,
|
|
"loss": 0.8453,
|
|
"mean_token_accuracy": 0.7792757660150528,
|
|
"num_tokens": 1462651274.0,
|
|
"step": 14620
|
|
},
|
|
{
|
|
"entropy": 0.87734375,
|
|
"epoch": 1.999521629194287,
|
|
"grad_norm": 0.2880336244952379,
|
|
"learning_rate": 1.7613075947583486e-09,
|
|
"loss": 0.8665,
|
|
"mean_token_accuracy": 0.7765883415937423,
|
|
"num_tokens": 1463658762.0,
|
|
"step": 14630
|
|
},
|
|
{
|
|
"entropy": 0.8805803571428571,
|
|
"epoch": 2.0,
|
|
"mean_token_accuracy": 0.7757753474371774,
|
|
"num_tokens": 1464020160.0,
|
|
"step": 14634,
|
|
"total_flos": 3199233565392896.0,
|
|
"train_loss": 0.8991405903616938,
|
|
"train_runtime": 65922.1312,
|
|
"train_samples_per_second": 28.413,
|
|
"train_steps_per_second": 0.222
|
|
}
|
|
],
|
|
"logging_steps": 10,
|
|
"max_steps": 14634,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 2,
|
|
"save_steps": 200,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": true
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 3199233565392896.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|