024d74a8bd
Model: fpadovani/ita-latn-10mb-after-ppt-Dp-10mb-ckpt500_seed10 Source: Original Platform
9134 lines
249 KiB
JSON
9134 lines
249 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 4.536560766515381,
|
|
"eval_steps": 500,
|
|
"global_step": 4500,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 7.2617237091064455,
|
|
"epoch": 0.005042864346949067,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 2e-06,
|
|
"loss": 6.8056,
|
|
"mean_token_accuracy": 0.09737122654914857,
|
|
"num_tokens": 11122.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 7.295074510574341,
|
|
"epoch": 0.010085728693898134,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 6.8855,
|
|
"mean_token_accuracy": 0.09183728471398353,
|
|
"num_tokens": 20499.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 7.276497220993042,
|
|
"epoch": 0.015128593040847202,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 7e-06,
|
|
"loss": 6.8595,
|
|
"mean_token_accuracy": 0.08771465197205544,
|
|
"num_tokens": 30664.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 7.273712778091431,
|
|
"epoch": 0.020171457387796268,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 6.7871,
|
|
"mean_token_accuracy": 0.0909557193517685,
|
|
"num_tokens": 41767.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 7.286615514755249,
|
|
"epoch": 0.025214321734745335,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 6.8418,
|
|
"mean_token_accuracy": 0.08894069865345955,
|
|
"num_tokens": 53233.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 7.305130863189698,
|
|
"epoch": 0.030257186081694403,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 6.8041,
|
|
"mean_token_accuracy": 0.08889350816607475,
|
|
"num_tokens": 63591.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 7.283368015289307,
|
|
"epoch": 0.03530005042864347,
|
|
"grad_norm": 0.80078125,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 6.9145,
|
|
"mean_token_accuracy": 0.08502179905772209,
|
|
"num_tokens": 75211.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 7.304086351394654,
|
|
"epoch": 0.040342914775592535,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 6.8684,
|
|
"mean_token_accuracy": 0.09216773957014084,
|
|
"num_tokens": 86163.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 7.274821615219116,
|
|
"epoch": 0.0453857791225416,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 6.8388,
|
|
"mean_token_accuracy": 0.09460289925336837,
|
|
"num_tokens": 96812.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 7.264426898956299,
|
|
"epoch": 0.05042864346949067,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 6.7243,
|
|
"mean_token_accuracy": 0.09600351974368096,
|
|
"num_tokens": 107246.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 7.250004816055298,
|
|
"epoch": 0.05547150781643974,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 6.811,
|
|
"mean_token_accuracy": 0.0851331278681755,
|
|
"num_tokens": 118169.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 7.238719892501831,
|
|
"epoch": 0.060514372163388806,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 6.729,
|
|
"mean_token_accuracy": 0.09378238096833229,
|
|
"num_tokens": 129744.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 7.2794413566589355,
|
|
"epoch": 0.06555723651033787,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 6.7855,
|
|
"mean_token_accuracy": 0.0973732315003872,
|
|
"num_tokens": 139826.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 7.233024740219117,
|
|
"epoch": 0.07060010085728693,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 6.7528,
|
|
"mean_token_accuracy": 0.09823397472500801,
|
|
"num_tokens": 149767.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 7.2112311840057375,
|
|
"epoch": 0.07564296520423601,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 6.8292,
|
|
"mean_token_accuracy": 0.09117417111992836,
|
|
"num_tokens": 160895.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 7.262843465805053,
|
|
"epoch": 0.08068582955118507,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 6.8576,
|
|
"mean_token_accuracy": 0.09194095991551876,
|
|
"num_tokens": 172080.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 7.305284118652343,
|
|
"epoch": 0.08572869389813415,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 6.8127,
|
|
"mean_token_accuracy": 0.0868966355919838,
|
|
"num_tokens": 182836.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 7.297293424606323,
|
|
"epoch": 0.0907715582450832,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 6.8288,
|
|
"mean_token_accuracy": 0.09184827134013177,
|
|
"num_tokens": 194781.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 7.244852924346924,
|
|
"epoch": 0.09581442259203228,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 6.8081,
|
|
"mean_token_accuracy": 0.09626058414578438,
|
|
"num_tokens": 206722.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 7.217989826202393,
|
|
"epoch": 0.10085728693898134,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 6.7712,
|
|
"mean_token_accuracy": 0.08988409265875816,
|
|
"num_tokens": 217167.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 7.27686243057251,
|
|
"epoch": 0.1059001512859304,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 6.8198,
|
|
"mean_token_accuracy": 0.09096184596419335,
|
|
"num_tokens": 228427.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 7.3203953266143795,
|
|
"epoch": 0.11094301563287948,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 6.8077,
|
|
"mean_token_accuracy": 0.09291791915893555,
|
|
"num_tokens": 239169.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 7.299695920944214,
|
|
"epoch": 0.11598587997982854,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 6.7301,
|
|
"mean_token_accuracy": 0.08924292176961898,
|
|
"num_tokens": 250393.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 7.252042818069458,
|
|
"epoch": 0.12102874432677761,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 6.7156,
|
|
"mean_token_accuracy": 0.09393385872244835,
|
|
"num_tokens": 260573.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 7.229413604736328,
|
|
"epoch": 0.1260716086737267,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 6.7662,
|
|
"mean_token_accuracy": 0.09847120642662048,
|
|
"num_tokens": 271537.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 7.2568823337554935,
|
|
"epoch": 0.13111447302067575,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 6.8175,
|
|
"mean_token_accuracy": 0.09340149909257889,
|
|
"num_tokens": 282080.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 7.2490527629852295,
|
|
"epoch": 0.1361573373676248,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 6.7345,
|
|
"mean_token_accuracy": 0.09461229220032692,
|
|
"num_tokens": 292509.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 7.2526007175445555,
|
|
"epoch": 0.14120020171457387,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 6.7403,
|
|
"mean_token_accuracy": 0.10240771397948265,
|
|
"num_tokens": 303156.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 7.221787309646606,
|
|
"epoch": 0.14624306606152296,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 6.7642,
|
|
"mean_token_accuracy": 0.10082685723900794,
|
|
"num_tokens": 314398.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 7.230781364440918,
|
|
"epoch": 0.15128593040847202,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 6.8726,
|
|
"mean_token_accuracy": 0.08736162185668946,
|
|
"num_tokens": 324774.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 7.317606258392334,
|
|
"epoch": 0.15632879475542108,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 6.7056,
|
|
"mean_token_accuracy": 0.1041252300143242,
|
|
"num_tokens": 335154.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 7.228124189376831,
|
|
"epoch": 0.16137165910237014,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 6.8042,
|
|
"mean_token_accuracy": 0.09531342014670371,
|
|
"num_tokens": 346180.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 7.3271020412445065,
|
|
"epoch": 0.1664145234493192,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 6.8265,
|
|
"mean_token_accuracy": 0.09442770406603813,
|
|
"num_tokens": 356722.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 7.268499708175659,
|
|
"epoch": 0.1714573877962683,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 6.7918,
|
|
"mean_token_accuracy": 0.0951433815062046,
|
|
"num_tokens": 367245.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 7.274964809417725,
|
|
"epoch": 0.17650025214321735,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 6.7936,
|
|
"mean_token_accuracy": 0.09519812390208245,
|
|
"num_tokens": 377434.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 7.194955635070801,
|
|
"epoch": 0.1815431164901664,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 6.7597,
|
|
"mean_token_accuracy": 0.09381223171949386,
|
|
"num_tokens": 388685.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 7.225965404510498,
|
|
"epoch": 0.18658598083711547,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 6.7328,
|
|
"mean_token_accuracy": 0.09787951037287712,
|
|
"num_tokens": 400559.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 7.295945835113526,
|
|
"epoch": 0.19162884518406456,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 6.7887,
|
|
"mean_token_accuracy": 0.09319317564368249,
|
|
"num_tokens": 413059.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 7.232351922988892,
|
|
"epoch": 0.19667170953101362,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 6.7232,
|
|
"mean_token_accuracy": 0.10411638170480728,
|
|
"num_tokens": 424073.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 7.209564685821533,
|
|
"epoch": 0.20171457387796268,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 6.6992,
|
|
"mean_token_accuracy": 0.0943477101624012,
|
|
"num_tokens": 435960.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 7.195583391189575,
|
|
"epoch": 0.20675743822491174,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000102,
|
|
"loss": 6.6617,
|
|
"mean_token_accuracy": 0.0991024374961853,
|
|
"num_tokens": 445458.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 7.186323165893555,
|
|
"epoch": 0.2118003025718608,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 6.716,
|
|
"mean_token_accuracy": 0.09350097700953483,
|
|
"num_tokens": 456237.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 7.203773832321167,
|
|
"epoch": 0.2168431669188099,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000107,
|
|
"loss": 6.7379,
|
|
"mean_token_accuracy": 0.09597255811095237,
|
|
"num_tokens": 467210.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 7.248147916793823,
|
|
"epoch": 0.22188603126575895,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 6.76,
|
|
"mean_token_accuracy": 0.09765960536897182,
|
|
"num_tokens": 478447.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 7.2771947383880615,
|
|
"epoch": 0.22692889561270801,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000112,
|
|
"loss": 6.7595,
|
|
"mean_token_accuracy": 0.09517326056957245,
|
|
"num_tokens": 489108.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 7.191942548751831,
|
|
"epoch": 0.23197175995965708,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 6.8357,
|
|
"mean_token_accuracy": 0.09494637697935104,
|
|
"num_tokens": 500889.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 7.266203784942627,
|
|
"epoch": 0.23701462430660616,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 6.6802,
|
|
"mean_token_accuracy": 0.09685032814741135,
|
|
"num_tokens": 511038.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 7.235995197296143,
|
|
"epoch": 0.24205748865355523,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 6.7868,
|
|
"mean_token_accuracy": 0.09551029652357101,
|
|
"num_tokens": 522801.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.309071063995361,
|
|
"epoch": 0.24710035300050429,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.000122,
|
|
"loss": 6.8082,
|
|
"mean_token_accuracy": 0.09518910646438598,
|
|
"num_tokens": 534795.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.229827165603638,
|
|
"epoch": 0.2521432173474534,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 6.8318,
|
|
"mean_token_accuracy": 0.09635223001241684,
|
|
"num_tokens": 546656.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 7.282848215103149,
|
|
"epoch": 0.25718608169440244,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000127,
|
|
"loss": 6.6941,
|
|
"mean_token_accuracy": 0.10145046636462211,
|
|
"num_tokens": 556673.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.170023965835571,
|
|
"epoch": 0.2622289460413515,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 6.7434,
|
|
"mean_token_accuracy": 0.09910880774259567,
|
|
"num_tokens": 567044.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.191144895553589,
|
|
"epoch": 0.26727181038830056,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000132,
|
|
"loss": 6.7009,
|
|
"mean_token_accuracy": 0.1039057917892933,
|
|
"num_tokens": 577305.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.184087181091309,
|
|
"epoch": 0.2723146747352496,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 6.7991,
|
|
"mean_token_accuracy": 0.10096023976802826,
|
|
"num_tokens": 589285.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.228927707672119,
|
|
"epoch": 0.2773575390821987,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 6.6089,
|
|
"mean_token_accuracy": 0.09944374114274979,
|
|
"num_tokens": 598609.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.188283395767212,
|
|
"epoch": 0.28240040342914774,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 6.7164,
|
|
"mean_token_accuracy": 0.10120134726166725,
|
|
"num_tokens": 609872.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.215302181243897,
|
|
"epoch": 0.2874432677760968,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 6.7278,
|
|
"mean_token_accuracy": 0.10587062165141106,
|
|
"num_tokens": 622583.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.145442628860474,
|
|
"epoch": 0.2924861321230459,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 6.713,
|
|
"mean_token_accuracy": 0.10004803910851479,
|
|
"num_tokens": 633592.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.203916120529175,
|
|
"epoch": 0.297528996469995,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000147,
|
|
"loss": 6.6463,
|
|
"mean_token_accuracy": 0.10566810145974159,
|
|
"num_tokens": 644858.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.087139463424682,
|
|
"epoch": 0.30257186081694404,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 6.7014,
|
|
"mean_token_accuracy": 0.09902146607637405,
|
|
"num_tokens": 656438.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.318219518661499,
|
|
"epoch": 0.3076147251638931,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000152,
|
|
"loss": 6.723,
|
|
"mean_token_accuracy": 0.10366858541965485,
|
|
"num_tokens": 667942.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.006646156311035,
|
|
"epoch": 0.31265758951084216,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 6.6336,
|
|
"mean_token_accuracy": 0.10130216181278229,
|
|
"num_tokens": 678596.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.340433883666992,
|
|
"epoch": 0.3177004538577912,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000157,
|
|
"loss": 6.7974,
|
|
"mean_token_accuracy": 0.09621977061033249,
|
|
"num_tokens": 691014.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.1193399906158445,
|
|
"epoch": 0.3227433182047403,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 6.7128,
|
|
"mean_token_accuracy": 0.10053488761186599,
|
|
"num_tokens": 701849.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.1668201923370365,
|
|
"epoch": 0.32778618255168934,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000162,
|
|
"loss": 6.6173,
|
|
"mean_token_accuracy": 0.10294296890497208,
|
|
"num_tokens": 712652.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.140855503082276,
|
|
"epoch": 0.3328290468986384,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 6.6802,
|
|
"mean_token_accuracy": 0.10038527771830559,
|
|
"num_tokens": 723704.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.10222053527832,
|
|
"epoch": 0.3378719112455875,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 6.6324,
|
|
"mean_token_accuracy": 0.10234513729810715,
|
|
"num_tokens": 734715.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.173121929168701,
|
|
"epoch": 0.3429147755925366,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 6.6738,
|
|
"mean_token_accuracy": 0.09645056575536728,
|
|
"num_tokens": 746772.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.129774475097657,
|
|
"epoch": 0.34795763993948564,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 6.6369,
|
|
"mean_token_accuracy": 0.09956872910261154,
|
|
"num_tokens": 757356.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.149800109863281,
|
|
"epoch": 0.3530005042864347,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 6.65,
|
|
"mean_token_accuracy": 0.10092864036560059,
|
|
"num_tokens": 768269.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.10404052734375,
|
|
"epoch": 0.35804336863338376,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.000177,
|
|
"loss": 6.598,
|
|
"mean_token_accuracy": 0.10126638934016227,
|
|
"num_tokens": 779553.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.155447053909302,
|
|
"epoch": 0.3630862329803328,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 6.7056,
|
|
"mean_token_accuracy": 0.10217617824673653,
|
|
"num_tokens": 791466.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.085558032989502,
|
|
"epoch": 0.3681290973272819,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000182,
|
|
"loss": 6.6028,
|
|
"mean_token_accuracy": 0.0978229895234108,
|
|
"num_tokens": 802064.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.105999708175659,
|
|
"epoch": 0.37317196167423095,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 6.6462,
|
|
"mean_token_accuracy": 0.09985020831227302,
|
|
"num_tokens": 813534.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.152196645736694,
|
|
"epoch": 0.37821482602118,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000187,
|
|
"loss": 6.6466,
|
|
"mean_token_accuracy": 0.09975219443440438,
|
|
"num_tokens": 824783.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.07678337097168,
|
|
"epoch": 0.3832576903681291,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 6.6484,
|
|
"mean_token_accuracy": 0.10236879363656044,
|
|
"num_tokens": 835019.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.1131932735443115,
|
|
"epoch": 0.3883005547150782,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000192,
|
|
"loss": 6.5494,
|
|
"mean_token_accuracy": 0.10095319002866746,
|
|
"num_tokens": 845954.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.060244226455689,
|
|
"epoch": 0.39334341906202724,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 6.5885,
|
|
"mean_token_accuracy": 0.10450648739933968,
|
|
"num_tokens": 856393.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.0785552024841305,
|
|
"epoch": 0.3983862834089763,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 6.6235,
|
|
"mean_token_accuracy": 0.10468716621398926,
|
|
"num_tokens": 868079.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.139645624160766,
|
|
"epoch": 0.40342914775592537,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 6.6665,
|
|
"mean_token_accuracy": 0.10834810510277748,
|
|
"num_tokens": 878995.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.10420970916748,
|
|
"epoch": 0.4084720121028744,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000202,
|
|
"loss": 6.6001,
|
|
"mean_token_accuracy": 0.0996965579688549,
|
|
"num_tokens": 889159.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.117736911773681,
|
|
"epoch": 0.4135148764498235,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 6.7006,
|
|
"mean_token_accuracy": 0.0964696764945984,
|
|
"num_tokens": 900186.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.083884239196777,
|
|
"epoch": 0.41855774079677255,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000207,
|
|
"loss": 6.6357,
|
|
"mean_token_accuracy": 0.09679703637957573,
|
|
"num_tokens": 910713.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.165265321731567,
|
|
"epoch": 0.4236006051437216,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 6.6897,
|
|
"mean_token_accuracy": 0.09690401703119278,
|
|
"num_tokens": 922020.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.095804452896118,
|
|
"epoch": 0.4286434694906707,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000212,
|
|
"loss": 6.664,
|
|
"mean_token_accuracy": 0.0992746777832508,
|
|
"num_tokens": 933033.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.1213442325592045,
|
|
"epoch": 0.4336863338376198,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 6.6859,
|
|
"mean_token_accuracy": 0.1018094040453434,
|
|
"num_tokens": 944394.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.0874436378479,
|
|
"epoch": 0.43872919818456885,
|
|
"grad_norm": 1.9140625,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 6.6229,
|
|
"mean_token_accuracy": 0.1008293315768242,
|
|
"num_tokens": 955337.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.019024896621704,
|
|
"epoch": 0.4437720625315179,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 6.5898,
|
|
"mean_token_accuracy": 0.10415408983826638,
|
|
"num_tokens": 967020.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.075509738922119,
|
|
"epoch": 0.44881492687846697,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000222,
|
|
"loss": 6.6064,
|
|
"mean_token_accuracy": 0.1006775140762329,
|
|
"num_tokens": 978223.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.066374015808106,
|
|
"epoch": 0.45385779122541603,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 6.6365,
|
|
"mean_token_accuracy": 0.10147074833512307,
|
|
"num_tokens": 990239.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.090767812728882,
|
|
"epoch": 0.4589006555723651,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 6.5587,
|
|
"mean_token_accuracy": 0.10007018893957138,
|
|
"num_tokens": 1001834.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 6.929092979431152,
|
|
"epoch": 0.46394351991931415,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 6.5505,
|
|
"mean_token_accuracy": 0.10508496388792991,
|
|
"num_tokens": 1012321.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.125833368301391,
|
|
"epoch": 0.4689863842662632,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 6.6506,
|
|
"mean_token_accuracy": 0.10367532521486282,
|
|
"num_tokens": 1024500.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.003234386444092,
|
|
"epoch": 0.47402924861321233,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.5685,
|
|
"mean_token_accuracy": 0.10941206514835358,
|
|
"num_tokens": 1034825.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 6.989887046813965,
|
|
"epoch": 0.4790721129601614,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.5501,
|
|
"mean_token_accuracy": 0.10572357177734375,
|
|
"num_tokens": 1046174.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 6.959958076477051,
|
|
"epoch": 0.48411497730711045,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.4638,
|
|
"mean_token_accuracy": 0.11223497986793518,
|
|
"num_tokens": 1055930.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.015145301818848,
|
|
"epoch": 0.4891578416540595,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.5128,
|
|
"mean_token_accuracy": 0.11089911982417107,
|
|
"num_tokens": 1066433.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.067583465576172,
|
|
"epoch": 0.49420070600100857,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.6169,
|
|
"mean_token_accuracy": 0.09941509291529656,
|
|
"num_tokens": 1077303.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 6.997565412521363,
|
|
"epoch": 0.49924357034795763,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.5535,
|
|
"mean_token_accuracy": 0.10777900144457817,
|
|
"num_tokens": 1088053.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 6.9472572803497314,
|
|
"epoch": 0.5042864346949067,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 6.5527,
|
|
"mean_token_accuracy": 0.10729503110051156,
|
|
"num_tokens": 1098518.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 0.5042864346949067,
|
|
"eval_entropy": 7.060033433006617,
|
|
"eval_loss": 6.604281902313232,
|
|
"eval_mean_token_accuracy": 0.10525795909322552,
|
|
"eval_num_tokens": 1098518.0,
|
|
"eval_runtime": 2.0309,
|
|
"eval_samples_per_second": 1545.14,
|
|
"eval_steps_per_second": 193.512,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.069717311859131,
|
|
"epoch": 0.5093292990418558,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.5707,
|
|
"mean_token_accuracy": 0.10687356069684029,
|
|
"num_tokens": 1109768.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 6.963026094436645,
|
|
"epoch": 0.5143721633888049,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.6064,
|
|
"mean_token_accuracy": 0.10696747824549675,
|
|
"num_tokens": 1121969.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 6.951304149627686,
|
|
"epoch": 0.5194150277357539,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.4483,
|
|
"mean_token_accuracy": 0.10768067240715026,
|
|
"num_tokens": 1132405.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.03754153251648,
|
|
"epoch": 0.524457892082703,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.5904,
|
|
"mean_token_accuracy": 0.09844426289200783,
|
|
"num_tokens": 1143705.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.015472745895385,
|
|
"epoch": 0.529500756429652,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.5613,
|
|
"mean_token_accuracy": 0.10252415537834167,
|
|
"num_tokens": 1154445.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 6.97110652923584,
|
|
"epoch": 0.5345436207766011,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.6562,
|
|
"mean_token_accuracy": 0.10317649990320206,
|
|
"num_tokens": 1166895.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.0424469947814945,
|
|
"epoch": 0.5395864851235502,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.5539,
|
|
"mean_token_accuracy": 0.10833557322621346,
|
|
"num_tokens": 1176744.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 6.992588472366333,
|
|
"epoch": 0.5446293494704992,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.5471,
|
|
"mean_token_accuracy": 0.1062086969614029,
|
|
"num_tokens": 1187024.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 6.923277378082275,
|
|
"epoch": 0.5496722138174484,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.5361,
|
|
"mean_token_accuracy": 0.1097439169883728,
|
|
"num_tokens": 1198799.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 6.881106615066528,
|
|
"epoch": 0.5547150781643974,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.4971,
|
|
"mean_token_accuracy": 0.10624601915478707,
|
|
"num_tokens": 1208585.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 6.963460206985474,
|
|
"epoch": 0.5597579425113465,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.5382,
|
|
"mean_token_accuracy": 0.11043122410774231,
|
|
"num_tokens": 1218268.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 6.98740587234497,
|
|
"epoch": 0.5648008068582955,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.5585,
|
|
"mean_token_accuracy": 0.1052480012178421,
|
|
"num_tokens": 1229106.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 6.9895421981811525,
|
|
"epoch": 0.5698436712052446,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.4845,
|
|
"mean_token_accuracy": 0.10795176848769188,
|
|
"num_tokens": 1239741.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 6.880759191513062,
|
|
"epoch": 0.5748865355521936,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.4538,
|
|
"mean_token_accuracy": 0.10221756920218468,
|
|
"num_tokens": 1249658.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 6.9524817943573,
|
|
"epoch": 0.5799293998991427,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.5063,
|
|
"mean_token_accuracy": 0.11247168630361556,
|
|
"num_tokens": 1259731.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 6.972530508041382,
|
|
"epoch": 0.5849722642460918,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.513,
|
|
"mean_token_accuracy": 0.1096158690750599,
|
|
"num_tokens": 1270087.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 6.778511095046997,
|
|
"epoch": 0.5900151285930408,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.4893,
|
|
"mean_token_accuracy": 0.11202885210514069,
|
|
"num_tokens": 1282567.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 6.9974161148071286,
|
|
"epoch": 0.59505799293999,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.5577,
|
|
"mean_token_accuracy": 0.10413449108600617,
|
|
"num_tokens": 1294133.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 6.887672233581543,
|
|
"epoch": 0.600100857286939,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.415,
|
|
"mean_token_accuracy": 0.11203444376587868,
|
|
"num_tokens": 1305306.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 6.8654016017913815,
|
|
"epoch": 0.6051437216338881,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.4948,
|
|
"mean_token_accuracy": 0.11017663776874542,
|
|
"num_tokens": 1316379.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 6.896556615829468,
|
|
"epoch": 0.6101865859808371,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.4725,
|
|
"mean_token_accuracy": 0.106760623306036,
|
|
"num_tokens": 1328121.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 6.924191999435425,
|
|
"epoch": 0.6152294503277862,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.5234,
|
|
"mean_token_accuracy": 0.10283615961670875,
|
|
"num_tokens": 1338728.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 6.927851343154908,
|
|
"epoch": 0.6202723146747352,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.4897,
|
|
"mean_token_accuracy": 0.10542894005775452,
|
|
"num_tokens": 1349126.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 6.790197515487671,
|
|
"epoch": 0.6253151790216843,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.3832,
|
|
"mean_token_accuracy": 0.1110648825764656,
|
|
"num_tokens": 1359813.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 6.906334066390992,
|
|
"epoch": 0.6303580433686334,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.5601,
|
|
"mean_token_accuracy": 0.09851819723844528,
|
|
"num_tokens": 1371872.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 6.909338903427124,
|
|
"epoch": 0.6354009077155824,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.4156,
|
|
"mean_token_accuracy": 0.10582672283053399,
|
|
"num_tokens": 1382789.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 6.804861688613892,
|
|
"epoch": 0.6404437720625316,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.5106,
|
|
"mean_token_accuracy": 0.10281929969787598,
|
|
"num_tokens": 1393419.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 6.8114887237548825,
|
|
"epoch": 0.6454866364094806,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.3617,
|
|
"mean_token_accuracy": 0.11038425639271736,
|
|
"num_tokens": 1404674.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 7.008165168762207,
|
|
"epoch": 0.6505295007564297,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.5633,
|
|
"mean_token_accuracy": 0.10471803769469261,
|
|
"num_tokens": 1416252.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 6.761869144439697,
|
|
"epoch": 0.6555723651033787,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.403,
|
|
"mean_token_accuracy": 0.10741439834237099,
|
|
"num_tokens": 1426681.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 6.852089309692383,
|
|
"epoch": 0.6606152294503278,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.385,
|
|
"mean_token_accuracy": 0.11156737729907036,
|
|
"num_tokens": 1436640.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 6.795392417907715,
|
|
"epoch": 0.6656580937972768,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.4142,
|
|
"mean_token_accuracy": 0.10801487416028976,
|
|
"num_tokens": 1447183.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 6.8026244163513185,
|
|
"epoch": 0.6707009581442259,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.4642,
|
|
"mean_token_accuracy": 0.10689442455768586,
|
|
"num_tokens": 1458351.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 6.816894626617431,
|
|
"epoch": 0.675743822491175,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.4583,
|
|
"mean_token_accuracy": 0.10456106811761856,
|
|
"num_tokens": 1469522.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 6.791521787643433,
|
|
"epoch": 0.680786686838124,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.3776,
|
|
"mean_token_accuracy": 0.1091451145708561,
|
|
"num_tokens": 1479947.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 6.8350683689117435,
|
|
"epoch": 0.6858295511850732,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.4012,
|
|
"mean_token_accuracy": 0.10581399723887444,
|
|
"num_tokens": 1490374.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 6.767127323150635,
|
|
"epoch": 0.6908724155320222,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.3965,
|
|
"mean_token_accuracy": 0.11490654349327087,
|
|
"num_tokens": 1500939.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 6.8393982410430905,
|
|
"epoch": 0.6959152798789713,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.5141,
|
|
"mean_token_accuracy": 0.10641767829656601,
|
|
"num_tokens": 1512803.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 6.697325801849365,
|
|
"epoch": 0.7009581442259203,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.3355,
|
|
"mean_token_accuracy": 0.11386772841215134,
|
|
"num_tokens": 1523679.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 6.699429321289062,
|
|
"epoch": 0.7060010085728694,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.3581,
|
|
"mean_token_accuracy": 0.11364351511001587,
|
|
"num_tokens": 1533892.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 6.7454273223876955,
|
|
"epoch": 0.7110438729198184,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.3217,
|
|
"mean_token_accuracy": 0.11361813247203827,
|
|
"num_tokens": 1543935.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 6.703502035140991,
|
|
"epoch": 0.7160867372667675,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.4326,
|
|
"mean_token_accuracy": 0.10583555996418,
|
|
"num_tokens": 1555989.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 6.864142894744873,
|
|
"epoch": 0.7211296016137166,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.3888,
|
|
"mean_token_accuracy": 0.10969913080334663,
|
|
"num_tokens": 1566355.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 6.730572843551636,
|
|
"epoch": 0.7261724659606656,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.4177,
|
|
"mean_token_accuracy": 0.11031778380274773,
|
|
"num_tokens": 1578403.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 6.761675405502319,
|
|
"epoch": 0.7312153303076148,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.4644,
|
|
"mean_token_accuracy": 0.10985069870948791,
|
|
"num_tokens": 1590072.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 6.820707082748413,
|
|
"epoch": 0.7362581946545638,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.4126,
|
|
"mean_token_accuracy": 0.11028422862291336,
|
|
"num_tokens": 1602051.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 6.766252660751343,
|
|
"epoch": 0.7413010590015129,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.4978,
|
|
"mean_token_accuracy": 0.11021630316972733,
|
|
"num_tokens": 1615741.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.66566915512085,
|
|
"epoch": 0.7463439233484619,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.3608,
|
|
"mean_token_accuracy": 0.11155292764306068,
|
|
"num_tokens": 1627042.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 6.806642866134643,
|
|
"epoch": 0.751386787695411,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.4065,
|
|
"mean_token_accuracy": 0.11185121461749077,
|
|
"num_tokens": 1638777.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.65570330619812,
|
|
"epoch": 0.75642965204236,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.316,
|
|
"mean_token_accuracy": 0.11404741406440735,
|
|
"num_tokens": 1650266.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 6.707036685943604,
|
|
"epoch": 0.7614725163893091,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.3594,
|
|
"mean_token_accuracy": 0.11311964690685272,
|
|
"num_tokens": 1660500.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 6.65473165512085,
|
|
"epoch": 0.7665153807362582,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.3177,
|
|
"mean_token_accuracy": 0.11318458691239357,
|
|
"num_tokens": 1671685.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.785394287109375,
|
|
"epoch": 0.7715582450832073,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.3959,
|
|
"mean_token_accuracy": 0.10736957937479019,
|
|
"num_tokens": 1681954.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.691954660415649,
|
|
"epoch": 0.7766011094301564,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.3974,
|
|
"mean_token_accuracy": 0.10903321579098701,
|
|
"num_tokens": 1691761.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.773788595199585,
|
|
"epoch": 0.7816439737771054,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.331,
|
|
"mean_token_accuracy": 0.11704913899302483,
|
|
"num_tokens": 1701715.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.5382654666900635,
|
|
"epoch": 0.7866868381240545,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.2663,
|
|
"mean_token_accuracy": 0.1199510358273983,
|
|
"num_tokens": 1714530.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.750433683395386,
|
|
"epoch": 0.7917297024710035,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.3839,
|
|
"mean_token_accuracy": 0.11454055681824685,
|
|
"num_tokens": 1725766.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.619108819961548,
|
|
"epoch": 0.7967725668179526,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.287,
|
|
"mean_token_accuracy": 0.12154799103736877,
|
|
"num_tokens": 1738307.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.705838060379028,
|
|
"epoch": 0.8018154311649016,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.3568,
|
|
"mean_token_accuracy": 0.11317641213536263,
|
|
"num_tokens": 1749129.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.698246812820434,
|
|
"epoch": 0.8068582955118507,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.3766,
|
|
"mean_token_accuracy": 0.10854917168617248,
|
|
"num_tokens": 1759426.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.596355867385864,
|
|
"epoch": 0.8119011598587998,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.2227,
|
|
"mean_token_accuracy": 0.11731593608856201,
|
|
"num_tokens": 1770906.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.6554864883422855,
|
|
"epoch": 0.8169440242057489,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.3713,
|
|
"mean_token_accuracy": 0.11492997258901597,
|
|
"num_tokens": 1782424.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.676445198059082,
|
|
"epoch": 0.821986888552698,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.3618,
|
|
"mean_token_accuracy": 0.11061819568276406,
|
|
"num_tokens": 1794331.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.6911256313323975,
|
|
"epoch": 0.827029752899647,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.2919,
|
|
"mean_token_accuracy": 0.11114619076251983,
|
|
"num_tokens": 1805395.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.559292125701904,
|
|
"epoch": 0.8320726172465961,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.2624,
|
|
"mean_token_accuracy": 0.1154046356678009,
|
|
"num_tokens": 1815820.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.627557945251465,
|
|
"epoch": 0.8371154815935451,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.2833,
|
|
"mean_token_accuracy": 0.11381199508905411,
|
|
"num_tokens": 1825754.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.595713376998901,
|
|
"epoch": 0.8421583459404942,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.3297,
|
|
"mean_token_accuracy": 0.11465614810585975,
|
|
"num_tokens": 1837899.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.735434579849243,
|
|
"epoch": 0.8472012102874432,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.2849,
|
|
"mean_token_accuracy": 0.11436659395694733,
|
|
"num_tokens": 1847659.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.551667785644531,
|
|
"epoch": 0.8522440746343923,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.2459,
|
|
"mean_token_accuracy": 0.11650302037596702,
|
|
"num_tokens": 1860126.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.459526205062867,
|
|
"epoch": 0.8572869389813415,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.2456,
|
|
"mean_token_accuracy": 0.11517760902643204,
|
|
"num_tokens": 1870567.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.737062883377075,
|
|
"epoch": 0.8623298033282905,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.3663,
|
|
"mean_token_accuracy": 0.11021938621997833,
|
|
"num_tokens": 1880274.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.584101152420044,
|
|
"epoch": 0.8673726676752396,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.3503,
|
|
"mean_token_accuracy": 0.11465656012296677,
|
|
"num_tokens": 1891122.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.649456930160523,
|
|
"epoch": 0.8724155320221886,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.3542,
|
|
"mean_token_accuracy": 0.1074532501399517,
|
|
"num_tokens": 1902629.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.633570432662964,
|
|
"epoch": 0.8774583963691377,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.2811,
|
|
"mean_token_accuracy": 0.11369187757372856,
|
|
"num_tokens": 1913808.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.550907325744629,
|
|
"epoch": 0.8825012607160867,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.2978,
|
|
"mean_token_accuracy": 0.1129491537809372,
|
|
"num_tokens": 1925973.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.628742361068726,
|
|
"epoch": 0.8875441250630358,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.2835,
|
|
"mean_token_accuracy": 0.12150776162743568,
|
|
"num_tokens": 1937808.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.597337293624878,
|
|
"epoch": 0.8925869894099848,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.2729,
|
|
"mean_token_accuracy": 0.11988255828619003,
|
|
"num_tokens": 1948581.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.559816074371338,
|
|
"epoch": 0.8976298537569339,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.3105,
|
|
"mean_token_accuracy": 0.11762073561549187,
|
|
"num_tokens": 1959977.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.6265698909759525,
|
|
"epoch": 0.902672718103883,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.3523,
|
|
"mean_token_accuracy": 0.1119124561548233,
|
|
"num_tokens": 1971328.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.592280387878418,
|
|
"epoch": 0.9077155824508321,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.4507,
|
|
"mean_token_accuracy": 0.11198776960372925,
|
|
"num_tokens": 1983805.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.724513721466065,
|
|
"epoch": 0.9127584467977812,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.3949,
|
|
"mean_token_accuracy": 0.10829810798168182,
|
|
"num_tokens": 1994773.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.579326343536377,
|
|
"epoch": 0.9178013111447302,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.26,
|
|
"mean_token_accuracy": 0.11540416032075881,
|
|
"num_tokens": 2005600.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.476587581634521,
|
|
"epoch": 0.9228441754916793,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.2251,
|
|
"mean_token_accuracy": 0.11662456914782524,
|
|
"num_tokens": 2016580.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.60697374343872,
|
|
"epoch": 0.9278870398386283,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.3115,
|
|
"mean_token_accuracy": 0.11826610565185547,
|
|
"num_tokens": 2030318.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.533180665969849,
|
|
"epoch": 0.9329299041855774,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.2687,
|
|
"mean_token_accuracy": 0.11941404268145561,
|
|
"num_tokens": 2041744.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.507296323776245,
|
|
"epoch": 0.9379727685325264,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.3127,
|
|
"mean_token_accuracy": 0.11259680539369583,
|
|
"num_tokens": 2053628.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.574466943740845,
|
|
"epoch": 0.9430156328794755,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.2969,
|
|
"mean_token_accuracy": 0.11736747696995735,
|
|
"num_tokens": 2065069.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.638606882095337,
|
|
"epoch": 0.9480584972264247,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.3201,
|
|
"mean_token_accuracy": 0.1115170620381832,
|
|
"num_tokens": 2076792.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.460126352310181,
|
|
"epoch": 0.9531013615733737,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.1793,
|
|
"mean_token_accuracy": 0.11806143298745156,
|
|
"num_tokens": 2087373.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.524953126907349,
|
|
"epoch": 0.9581442259203228,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.2281,
|
|
"mean_token_accuracy": 0.11285180300474167,
|
|
"num_tokens": 2100846.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.457489919662476,
|
|
"epoch": 0.9631870902672718,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.2246,
|
|
"mean_token_accuracy": 0.11971727535128593,
|
|
"num_tokens": 2110668.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.50002498626709,
|
|
"epoch": 0.9682299546142209,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.2273,
|
|
"mean_token_accuracy": 0.1177629716694355,
|
|
"num_tokens": 2120917.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.47516942024231,
|
|
"epoch": 0.9732728189611699,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.2124,
|
|
"mean_token_accuracy": 0.11174920573830605,
|
|
"num_tokens": 2132929.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.457158470153809,
|
|
"epoch": 0.978315683308119,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.1831,
|
|
"mean_token_accuracy": 0.1205831728875637,
|
|
"num_tokens": 2145251.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.493334341049194,
|
|
"epoch": 0.983358547655068,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.2186,
|
|
"mean_token_accuracy": 0.12307626530528068,
|
|
"num_tokens": 2154702.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.501614618301391,
|
|
"epoch": 0.9884014120020171,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.2312,
|
|
"mean_token_accuracy": 0.12058174088597298,
|
|
"num_tokens": 2165924.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.444549608230591,
|
|
"epoch": 0.9934442763489663,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.2136,
|
|
"mean_token_accuracy": 0.11923774480819702,
|
|
"num_tokens": 2176933.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.511642837524414,
|
|
"epoch": 0.9984871406959153,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.3003,
|
|
"mean_token_accuracy": 0.11393278166651725,
|
|
"num_tokens": 2188812.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.463429715898302,
|
|
"epoch": 1.0030257186081695,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.027,
|
|
"mean_token_accuracy": 0.11740631692939335,
|
|
"num_tokens": 2199238.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.489912509918213,
|
|
"epoch": 1.0080685829551186,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.0516,
|
|
"mean_token_accuracy": 0.1249640092253685,
|
|
"num_tokens": 2210530.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 1.0080685829551186,
|
|
"eval_entropy": 6.2810746780181965,
|
|
"eval_loss": 6.284154891967773,
|
|
"eval_mean_token_accuracy": 0.12013455548586736,
|
|
"eval_num_tokens": 2210530.0,
|
|
"eval_runtime": 2.0114,
|
|
"eval_samples_per_second": 1560.098,
|
|
"eval_steps_per_second": 195.385,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.447393751144409,
|
|
"epoch": 1.0131114473020675,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004999997762223182,
|
|
"loss": 6.07,
|
|
"mean_token_accuracy": 0.1205926515161991,
|
|
"num_tokens": 2221721.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.407036399841308,
|
|
"epoch": 1.0181543116490166,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004999988671262489,
|
|
"loss": 6.1107,
|
|
"mean_token_accuracy": 0.12448062747716904,
|
|
"num_tokens": 2233571.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.4423901557922365,
|
|
"epoch": 1.0231971759959657,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004999972587285105,
|
|
"loss": 6.1121,
|
|
"mean_token_accuracy": 0.1273207150399685,
|
|
"num_tokens": 2243419.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.499564075469971,
|
|
"epoch": 1.0282400403429148,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999949510341015,
|
|
"loss": 6.0819,
|
|
"mean_token_accuracy": 0.12823617607355117,
|
|
"num_tokens": 2254461.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.352968502044678,
|
|
"epoch": 1.033282904689864,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999919440501947,
|
|
"loss": 6.0502,
|
|
"mean_token_accuracy": 0.11969454362988471,
|
|
"num_tokens": 2266107.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.434625720977783,
|
|
"epoch": 1.0383257690368128,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004999882377861355,
|
|
"loss": 6.0985,
|
|
"mean_token_accuracy": 0.12077055871486664,
|
|
"num_tokens": 2276084.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.439082241058349,
|
|
"epoch": 1.043368633383762,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004999838322534433,
|
|
"loss": 6.1894,
|
|
"mean_token_accuracy": 0.11787507310509682,
|
|
"num_tokens": 2287364.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.478655195236206,
|
|
"epoch": 1.048411497730711,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999787274658107,
|
|
"loss": 6.0605,
|
|
"mean_token_accuracy": 0.12589174658060073,
|
|
"num_tokens": 2298313.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.380098056793213,
|
|
"epoch": 1.0534543620776602,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004999729234391032,
|
|
"loss": 6.061,
|
|
"mean_token_accuracy": 0.12527281194925308,
|
|
"num_tokens": 2310126.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.350380373001099,
|
|
"epoch": 1.058497226424609,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00049996642019136,
|
|
"loss": 6.0008,
|
|
"mean_token_accuracy": 0.12471744045615196,
|
|
"num_tokens": 2320830.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.332125091552735,
|
|
"epoch": 1.0635400907715582,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.0004999592177427933,
|
|
"loss": 6.0628,
|
|
"mean_token_accuracy": 0.12736885622143745,
|
|
"num_tokens": 2332027.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.423359632492065,
|
|
"epoch": 1.0685829551185073,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999513161157885,
|
|
"loss": 6.0875,
|
|
"mean_token_accuracy": 0.12787878662347793,
|
|
"num_tokens": 2342989.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.357008934020996,
|
|
"epoch": 1.0736258194654564,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999427153349041,
|
|
"loss": 6.0381,
|
|
"mean_token_accuracy": 0.12251123264431954,
|
|
"num_tokens": 2353321.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.386510133743286,
|
|
"epoch": 1.0786686838124055,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999334154268714,
|
|
"loss": 6.0701,
|
|
"mean_token_accuracy": 0.12497062757611274,
|
|
"num_tokens": 2363910.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.377871799468994,
|
|
"epoch": 1.0837115481593544,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999234164205949,
|
|
"loss": 6.0782,
|
|
"mean_token_accuracy": 0.12658404186367989,
|
|
"num_tokens": 2374589.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.461424016952515,
|
|
"epoch": 1.0887544125063036,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004999127183471517,
|
|
"loss": 6.1491,
|
|
"mean_token_accuracy": 0.12108687609434128,
|
|
"num_tokens": 2384684.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.421203565597534,
|
|
"epoch": 1.0937972768532527,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004999013212397915,
|
|
"loss": 6.1419,
|
|
"mean_token_accuracy": 0.12452849447727203,
|
|
"num_tokens": 2396649.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.405708503723145,
|
|
"epoch": 1.0988401412002018,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004998892251339368,
|
|
"loss": 6.0285,
|
|
"mean_token_accuracy": 0.1342999219894409,
|
|
"num_tokens": 2407696.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.423946237564087,
|
|
"epoch": 1.1038830055471507,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004998764300671827,
|
|
"loss": 6.1091,
|
|
"mean_token_accuracy": 0.12314606979489326,
|
|
"num_tokens": 2418333.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.297891998291016,
|
|
"epoch": 1.1089258698940998,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004998629360792965,
|
|
"loss": 6.1175,
|
|
"mean_token_accuracy": 0.121670650690794,
|
|
"num_tokens": 2431081.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.463546562194824,
|
|
"epoch": 1.113968734241049,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004998487432122179,
|
|
"loss": 6.0779,
|
|
"mean_token_accuracy": 0.1292563296854496,
|
|
"num_tokens": 2441055.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.340994119644165,
|
|
"epoch": 1.119011598587998,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004998338515100585,
|
|
"loss": 6.099,
|
|
"mean_token_accuracy": 0.12495715767145157,
|
|
"num_tokens": 2451933.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.426595640182495,
|
|
"epoch": 1.1240544629349472,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004998182610191022,
|
|
"loss": 6.1062,
|
|
"mean_token_accuracy": 0.12290222644805908,
|
|
"num_tokens": 2463062.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.373396110534668,
|
|
"epoch": 1.129097327281896,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004998019717878047,
|
|
"loss": 6.1153,
|
|
"mean_token_accuracy": 0.11965839490294457,
|
|
"num_tokens": 2473650.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.376273536682129,
|
|
"epoch": 1.1341401916288452,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000499784983866793,
|
|
"loss": 6.1622,
|
|
"mean_token_accuracy": 0.12190458104014397,
|
|
"num_tokens": 2484510.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.404547119140625,
|
|
"epoch": 1.1391830559757943,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004997672973088661,
|
|
"loss": 6.0546,
|
|
"mean_token_accuracy": 0.12516579329967498,
|
|
"num_tokens": 2495881.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.320344066619873,
|
|
"epoch": 1.1442259203227434,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004997489121689944,
|
|
"loss": 6.0118,
|
|
"mean_token_accuracy": 0.12706558033823967,
|
|
"num_tokens": 2506657.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.276113080978393,
|
|
"epoch": 1.1492687846696923,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004997298285043189,
|
|
"loss": 6.0261,
|
|
"mean_token_accuracy": 0.1287092536687851,
|
|
"num_tokens": 2518082.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.282301521301269,
|
|
"epoch": 1.1543116490166414,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004997100463741527,
|
|
"loss": 5.9992,
|
|
"mean_token_accuracy": 0.13235862776637078,
|
|
"num_tokens": 2528722.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.4319908142089846,
|
|
"epoch": 1.1593545133635905,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004996895658399785,
|
|
"loss": 5.989,
|
|
"mean_token_accuracy": 0.1265767492353916,
|
|
"num_tokens": 2539008.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.202495479583741,
|
|
"epoch": 1.1643973777105396,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004996683869654507,
|
|
"loss": 6.0081,
|
|
"mean_token_accuracy": 0.12378924116492271,
|
|
"num_tokens": 2549445.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.262302541732788,
|
|
"epoch": 1.1694402420574885,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004996465098163934,
|
|
"loss": 6.0604,
|
|
"mean_token_accuracy": 0.12376785799860954,
|
|
"num_tokens": 2561079.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.3276917934417725,
|
|
"epoch": 1.1744831064044376,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004996239344608015,
|
|
"loss": 6.0052,
|
|
"mean_token_accuracy": 0.12625558227300643,
|
|
"num_tokens": 2571540.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.2939674854278564,
|
|
"epoch": 1.1795259707513868,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004996006609688398,
|
|
"loss": 5.9617,
|
|
"mean_token_accuracy": 0.1271153785288334,
|
|
"num_tokens": 2582184.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.216596078872681,
|
|
"epoch": 1.1845688350983359,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004995766894128426,
|
|
"loss": 6.0812,
|
|
"mean_token_accuracy": 0.1252618446946144,
|
|
"num_tokens": 2592771.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.355675554275512,
|
|
"epoch": 1.189611699445285,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004995520198673143,
|
|
"loss": 5.972,
|
|
"mean_token_accuracy": 0.12432529479265213,
|
|
"num_tokens": 2602974.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.265661001205444,
|
|
"epoch": 1.194654563792234,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004995266524089283,
|
|
"loss": 5.9881,
|
|
"mean_token_accuracy": 0.12971638441085814,
|
|
"num_tokens": 2613915.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.377945756912231,
|
|
"epoch": 1.199697428139183,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004995005871165271,
|
|
"loss": 6.06,
|
|
"mean_token_accuracy": 0.12619838640093803,
|
|
"num_tokens": 2626038.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.240468740463257,
|
|
"epoch": 1.2047402924861321,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004994738240711225,
|
|
"loss": 6.0132,
|
|
"mean_token_accuracy": 0.12550414875149726,
|
|
"num_tokens": 2636448.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.406741380691528,
|
|
"epoch": 1.2097831568330812,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004994463633558945,
|
|
"loss": 6.1128,
|
|
"mean_token_accuracy": 0.1222912460565567,
|
|
"num_tokens": 2648803.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.3317461013793945,
|
|
"epoch": 1.2148260211800301,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004994182050561917,
|
|
"loss": 6.0209,
|
|
"mean_token_accuracy": 0.13138329982757568,
|
|
"num_tokens": 2659981.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.1513348579406735,
|
|
"epoch": 1.2198688855269793,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004993893492595306,
|
|
"loss": 5.9957,
|
|
"mean_token_accuracy": 0.12883782237768174,
|
|
"num_tokens": 2670340.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.35710973739624,
|
|
"epoch": 1.2249117498739284,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004993597960555959,
|
|
"loss": 6.0213,
|
|
"mean_token_accuracy": 0.12856497168540953,
|
|
"num_tokens": 2681147.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.3188924312591555,
|
|
"epoch": 1.2299546142208775,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004993295455362395,
|
|
"loss": 5.9561,
|
|
"mean_token_accuracy": 0.13173700943589212,
|
|
"num_tokens": 2693877.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.207800817489624,
|
|
"epoch": 1.2349974785678266,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004992985977954808,
|
|
"loss": 5.9825,
|
|
"mean_token_accuracy": 0.12401358932256698,
|
|
"num_tokens": 2704371.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.372862815856934,
|
|
"epoch": 1.2400403429147757,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.000499266952929506,
|
|
"loss": 6.0539,
|
|
"mean_token_accuracy": 0.12760528549551964,
|
|
"num_tokens": 2714549.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.172968769073487,
|
|
"epoch": 1.2450832072617246,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004992346110366682,
|
|
"loss": 5.9879,
|
|
"mean_token_accuracy": 0.12741455659270287,
|
|
"num_tokens": 2723750.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.330499887466431,
|
|
"epoch": 1.2501260716086737,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004992015722174865,
|
|
"loss": 6.0366,
|
|
"mean_token_accuracy": 0.12289296686649323,
|
|
"num_tokens": 2734557.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.2792950630187985,
|
|
"epoch": 1.2551689359556228,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004991678365746466,
|
|
"loss": 6.038,
|
|
"mean_token_accuracy": 0.12846242338418962,
|
|
"num_tokens": 2746521.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.2055260181427006,
|
|
"epoch": 1.2602118003025717,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004991334042129996,
|
|
"loss": 6.0438,
|
|
"mean_token_accuracy": 0.1301764003932476,
|
|
"num_tokens": 2758074.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.263936710357666,
|
|
"epoch": 1.2652546646495209,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004990982752395618,
|
|
"loss": 5.9069,
|
|
"mean_token_accuracy": 0.13039981722831726,
|
|
"num_tokens": 2770059.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.219188785552978,
|
|
"epoch": 1.27029752899647,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004990624497635152,
|
|
"loss": 5.9737,
|
|
"mean_token_accuracy": 0.12883303165435792,
|
|
"num_tokens": 2780802.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.340838193893433,
|
|
"epoch": 1.275340393343419,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000499025927896206,
|
|
"loss": 6.0658,
|
|
"mean_token_accuracy": 0.12543840259313582,
|
|
"num_tokens": 2793693.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.29937858581543,
|
|
"epoch": 1.2803832576903682,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004989887097511451,
|
|
"loss": 6.0798,
|
|
"mean_token_accuracy": 0.12394527494907379,
|
|
"num_tokens": 2805547.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.219983339309692,
|
|
"epoch": 1.2854261220373173,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004989507954440074,
|
|
"loss": 5.9966,
|
|
"mean_token_accuracy": 0.12440208271145821,
|
|
"num_tokens": 2816207.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.305551385879516,
|
|
"epoch": 1.2904689863842662,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004989121850926313,
|
|
"loss": 6.0032,
|
|
"mean_token_accuracy": 0.12171063870191574,
|
|
"num_tokens": 2827272.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.167698240280151,
|
|
"epoch": 1.2955118507312153,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004988728788170188,
|
|
"loss": 5.953,
|
|
"mean_token_accuracy": 0.1269591346383095,
|
|
"num_tokens": 2839814.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.367572212219239,
|
|
"epoch": 1.3005547150781644,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004988328767393346,
|
|
"loss": 6.1097,
|
|
"mean_token_accuracy": 0.12364127933979034,
|
|
"num_tokens": 2851629.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.212838983535766,
|
|
"epoch": 1.3055975794251133,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004987921789839061,
|
|
"loss": 6.0245,
|
|
"mean_token_accuracy": 0.13126721307635308,
|
|
"num_tokens": 2863518.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.211556339263916,
|
|
"epoch": 1.3106404437720625,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004987507856772229,
|
|
"loss": 5.9859,
|
|
"mean_token_accuracy": 0.12847694158554077,
|
|
"num_tokens": 2874803.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.286357021331787,
|
|
"epoch": 1.3156833081190116,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004987086969479364,
|
|
"loss": 5.9707,
|
|
"mean_token_accuracy": 0.1285534307360649,
|
|
"num_tokens": 2886052.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.237084150314331,
|
|
"epoch": 1.3207261724659607,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004986659129268592,
|
|
"loss": 5.9869,
|
|
"mean_token_accuracy": 0.12723974362015725,
|
|
"num_tokens": 2897797.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.174638843536377,
|
|
"epoch": 1.3257690368129098,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004986224337469652,
|
|
"loss": 5.9547,
|
|
"mean_token_accuracy": 0.12878516539931298,
|
|
"num_tokens": 2908752.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.231057977676391,
|
|
"epoch": 1.330811901159859,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004985782595433886,
|
|
"loss": 6.0251,
|
|
"mean_token_accuracy": 0.12619352489709854,
|
|
"num_tokens": 2919351.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.283967208862305,
|
|
"epoch": 1.3358547655068078,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004985333904534237,
|
|
"loss": 5.9879,
|
|
"mean_token_accuracy": 0.12927695512771606,
|
|
"num_tokens": 2930198.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.181272029876709,
|
|
"epoch": 1.340897629853757,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.000498487826616525,
|
|
"loss": 5.9854,
|
|
"mean_token_accuracy": 0.13312289491295815,
|
|
"num_tokens": 2941349.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.184599733352661,
|
|
"epoch": 1.345940494200706,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004984415681743056,
|
|
"loss": 5.9022,
|
|
"mean_token_accuracy": 0.12977838665246963,
|
|
"num_tokens": 2951740.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.227171850204468,
|
|
"epoch": 1.350983358547655,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004983946152705382,
|
|
"loss": 5.9575,
|
|
"mean_token_accuracy": 0.12978332713246346,
|
|
"num_tokens": 2963900.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.195958662033081,
|
|
"epoch": 1.356026222894604,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004983469680511531,
|
|
"loss": 5.9856,
|
|
"mean_token_accuracy": 0.12655954882502557,
|
|
"num_tokens": 2975385.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.2479315280914305,
|
|
"epoch": 1.3610690872415532,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004982986266642393,
|
|
"loss": 5.9775,
|
|
"mean_token_accuracy": 0.12489714547991752,
|
|
"num_tokens": 2986340.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.157347059249878,
|
|
"epoch": 1.3661119515885023,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004982495912600426,
|
|
"loss": 5.8547,
|
|
"mean_token_accuracy": 0.13616741970181465,
|
|
"num_tokens": 2996582.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.192052793502808,
|
|
"epoch": 1.3711548159354514,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004981998619909664,
|
|
"loss": 5.8838,
|
|
"mean_token_accuracy": 0.12813595458865165,
|
|
"num_tokens": 3007620.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.152505588531494,
|
|
"epoch": 1.3761976802824005,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004981494390115704,
|
|
"loss": 5.9991,
|
|
"mean_token_accuracy": 0.1324993073940277,
|
|
"num_tokens": 3019520.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.180707025527954,
|
|
"epoch": 1.3812405446293494,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004980983224785704,
|
|
"loss": 5.9832,
|
|
"mean_token_accuracy": 0.12408366501331329,
|
|
"num_tokens": 3030616.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.305917692184448,
|
|
"epoch": 1.3862834089762985,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004980465125508376,
|
|
"loss": 6.046,
|
|
"mean_token_accuracy": 0.1258719839155674,
|
|
"num_tokens": 3042383.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.218978786468506,
|
|
"epoch": 1.3913262733232477,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004979940093893986,
|
|
"loss": 5.9547,
|
|
"mean_token_accuracy": 0.12919819876551628,
|
|
"num_tokens": 3052833.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.144861125946045,
|
|
"epoch": 1.3963691376701965,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004979408131574345,
|
|
"loss": 5.9286,
|
|
"mean_token_accuracy": 0.13193050548434257,
|
|
"num_tokens": 3063800.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.260790491104126,
|
|
"epoch": 1.4014120020171457,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004978869240202802,
|
|
"loss": 6.0707,
|
|
"mean_token_accuracy": 0.1271720454096794,
|
|
"num_tokens": 3075243.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.18296537399292,
|
|
"epoch": 1.4064548663640948,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004978323421454246,
|
|
"loss": 5.9249,
|
|
"mean_token_accuracy": 0.13557921946048737,
|
|
"num_tokens": 3086590.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.140930032730102,
|
|
"epoch": 1.411497730711044,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004977770677025092,
|
|
"loss": 5.9767,
|
|
"mean_token_accuracy": 0.13090722784399986,
|
|
"num_tokens": 3097554.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.219397163391113,
|
|
"epoch": 1.416540595057993,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004977211008633284,
|
|
"loss": 5.9879,
|
|
"mean_token_accuracy": 0.1254073180258274,
|
|
"num_tokens": 3110051.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.1226543426513675,
|
|
"epoch": 1.4215834594049421,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004976644418018285,
|
|
"loss": 5.8721,
|
|
"mean_token_accuracy": 0.13179416358470916,
|
|
"num_tokens": 3120640.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.204487943649292,
|
|
"epoch": 1.426626323751891,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000497607090694107,
|
|
"loss": 5.896,
|
|
"mean_token_accuracy": 0.13633572533726693,
|
|
"num_tokens": 3131293.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.161991024017334,
|
|
"epoch": 1.4316691880988401,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004975490477184126,
|
|
"loss": 5.9284,
|
|
"mean_token_accuracy": 0.13297617733478545,
|
|
"num_tokens": 3141272.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.224547624588013,
|
|
"epoch": 1.4367120524457893,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000497490313055144,
|
|
"loss": 5.8888,
|
|
"mean_token_accuracy": 0.13297630921006204,
|
|
"num_tokens": 3151349.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.118403148651123,
|
|
"epoch": 1.4417549167927382,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004974308868868501,
|
|
"loss": 5.9594,
|
|
"mean_token_accuracy": 0.13158065676689149,
|
|
"num_tokens": 3162258.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.253059816360474,
|
|
"epoch": 1.4467977811396873,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004973707693982287,
|
|
"loss": 5.9934,
|
|
"mean_token_accuracy": 0.13116943165659906,
|
|
"num_tokens": 3172821.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.142082262039184,
|
|
"epoch": 1.4518406454866364,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004973099607761264,
|
|
"loss": 5.9322,
|
|
"mean_token_accuracy": 0.12523773387074472,
|
|
"num_tokens": 3183720.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.308740520477295,
|
|
"epoch": 1.4568835098335855,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004972484612095378,
|
|
"loss": 5.9893,
|
|
"mean_token_accuracy": 0.13426656946539878,
|
|
"num_tokens": 3192882.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.14431209564209,
|
|
"epoch": 1.4619263741805346,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000497186270889605,
|
|
"loss": 5.9204,
|
|
"mean_token_accuracy": 0.12773872464895247,
|
|
"num_tokens": 3203908.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.197584581375122,
|
|
"epoch": 1.4669692385274837,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004971233900096169,
|
|
"loss": 6.0159,
|
|
"mean_token_accuracy": 0.12962774708867073,
|
|
"num_tokens": 3215145.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.1720963478088375,
|
|
"epoch": 1.4720121028744326,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004970598187650085,
|
|
"loss": 5.8951,
|
|
"mean_token_accuracy": 0.13193247690796853,
|
|
"num_tokens": 3225714.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.174853086471558,
|
|
"epoch": 1.4770549672213817,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.000496995557353361,
|
|
"loss": 6.0193,
|
|
"mean_token_accuracy": 0.12787997275590895,
|
|
"num_tokens": 3238541.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.118457794189453,
|
|
"epoch": 1.4820978315683309,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004969306059744004,
|
|
"loss": 5.825,
|
|
"mean_token_accuracy": 0.13631526827812196,
|
|
"num_tokens": 3250219.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.043021106719971,
|
|
"epoch": 1.4871406959152798,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004968649648299968,
|
|
"loss": 5.884,
|
|
"mean_token_accuracy": 0.13325665295124053,
|
|
"num_tokens": 3261676.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.183950042724609,
|
|
"epoch": 1.4921835602622289,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004967986341241646,
|
|
"loss": 5.9726,
|
|
"mean_token_accuracy": 0.1213144265115261,
|
|
"num_tokens": 3272862.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.16568922996521,
|
|
"epoch": 1.497226424609178,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004967316140630611,
|
|
"loss": 5.8834,
|
|
"mean_token_accuracy": 0.13413841128349305,
|
|
"num_tokens": 3282891.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.0396716594696045,
|
|
"epoch": 1.502269288956127,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004966639048549862,
|
|
"loss": 5.8232,
|
|
"mean_token_accuracy": 0.13486799895763396,
|
|
"num_tokens": 3293105.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.129441499710083,
|
|
"epoch": 1.5073121533030762,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004965955067103816,
|
|
"loss": 5.8892,
|
|
"mean_token_accuracy": 0.13918987289071083,
|
|
"num_tokens": 3304431.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.214106845855713,
|
|
"epoch": 1.5123550176500253,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004965264198418305,
|
|
"loss": 5.9469,
|
|
"mean_token_accuracy": 0.1310076482594013,
|
|
"num_tokens": 3314391.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"epoch": 1.5123550176500253,
|
|
"eval_entropy": 6.069466057017863,
|
|
"eval_loss": 6.0718994140625,
|
|
"eval_mean_token_accuracy": 0.13017470201918188,
|
|
"eval_num_tokens": 3314391.0,
|
|
"eval_runtime": 2.003,
|
|
"eval_samples_per_second": 1566.682,
|
|
"eval_steps_per_second": 196.21,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 6.124775123596192,
|
|
"epoch": 1.5173978819969742,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004964566444640564,
|
|
"loss": 5.8422,
|
|
"mean_token_accuracy": 0.1313735343515873,
|
|
"num_tokens": 3324578.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 6.113707828521728,
|
|
"epoch": 1.5224407463439233,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004963861807939227,
|
|
"loss": 5.8864,
|
|
"mean_token_accuracy": 0.13799641504883767,
|
|
"num_tokens": 3334724.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 6.094220495223999,
|
|
"epoch": 1.5274836106908725,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004963150290504321,
|
|
"loss": 5.9887,
|
|
"mean_token_accuracy": 0.13121608048677444,
|
|
"num_tokens": 3345371.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 6.165015411376953,
|
|
"epoch": 1.5325264750378214,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004962431894547259,
|
|
"loss": 5.9314,
|
|
"mean_token_accuracy": 0.1312360055744648,
|
|
"num_tokens": 3357176.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 6.122510099411011,
|
|
"epoch": 1.5375693393847705,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004961706622300834,
|
|
"loss": 5.9476,
|
|
"mean_token_accuracy": 0.1257225900888443,
|
|
"num_tokens": 3368213.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 6.156673336029053,
|
|
"epoch": 1.5426122037317196,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004960974476019206,
|
|
"loss": 5.9041,
|
|
"mean_token_accuracy": 0.1333881177008152,
|
|
"num_tokens": 3379676.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 6.067805290222168,
|
|
"epoch": 1.5476550680786687,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004960235457977905,
|
|
"loss": 5.8493,
|
|
"mean_token_accuracy": 0.13325185924768448,
|
|
"num_tokens": 3390765.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 6.166102313995362,
|
|
"epoch": 1.5526979324256178,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004959489570473814,
|
|
"loss": 5.8747,
|
|
"mean_token_accuracy": 0.13599455803632737,
|
|
"num_tokens": 3401804.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 6.162479019165039,
|
|
"epoch": 1.557740796772567,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000495873681582517,
|
|
"loss": 6.0004,
|
|
"mean_token_accuracy": 0.13533953428268433,
|
|
"num_tokens": 3413509.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 6.145385789871216,
|
|
"epoch": 1.5627836611195158,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004957977196371552,
|
|
"loss": 5.8561,
|
|
"mean_token_accuracy": 0.13544412851333618,
|
|
"num_tokens": 3423807.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 6.006620979309082,
|
|
"epoch": 1.567826525466465,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004957210714473873,
|
|
"loss": 5.8197,
|
|
"mean_token_accuracy": 0.1363327421247959,
|
|
"num_tokens": 3434214.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 6.1910912036895756,
|
|
"epoch": 1.572869389813414,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004956437372514378,
|
|
"loss": 5.8765,
|
|
"mean_token_accuracy": 0.13662820681929588,
|
|
"num_tokens": 3445081.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 6.07164626121521,
|
|
"epoch": 1.577912254160363,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004955657172896631,
|
|
"loss": 5.861,
|
|
"mean_token_accuracy": 0.13575060442090034,
|
|
"num_tokens": 3454992.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 6.066172361373901,
|
|
"epoch": 1.582955118507312,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004954870118045511,
|
|
"loss": 5.9597,
|
|
"mean_token_accuracy": 0.12645145282149314,
|
|
"num_tokens": 3465635.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 6.161420822143555,
|
|
"epoch": 1.5879979828542612,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004954076210407203,
|
|
"loss": 5.8763,
|
|
"mean_token_accuracy": 0.13580406308174134,
|
|
"num_tokens": 3477746.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 6.062290048599243,
|
|
"epoch": 1.5930408472012103,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004953275452449188,
|
|
"loss": 5.9255,
|
|
"mean_token_accuracy": 0.1325630784034729,
|
|
"num_tokens": 3488915.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 6.116338920593262,
|
|
"epoch": 1.5980837115481594,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004952467846660243,
|
|
"loss": 5.8979,
|
|
"mean_token_accuracy": 0.13235626220703126,
|
|
"num_tokens": 3499096.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 6.011581039428711,
|
|
"epoch": 1.6031265758951085,
|
|
"grad_norm": 0.76171875,
|
|
"learning_rate": 0.0004951653395550423,
|
|
"loss": 5.7789,
|
|
"mean_token_accuracy": 0.12726815417408943,
|
|
"num_tokens": 3512544.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 6.102898740768433,
|
|
"epoch": 1.6081694402420574,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004950832101651062,
|
|
"loss": 5.911,
|
|
"mean_token_accuracy": 0.13158985376358032,
|
|
"num_tokens": 3522814.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 6.033138418197632,
|
|
"epoch": 1.6132123045890066,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004950003967514762,
|
|
"loss": 5.8334,
|
|
"mean_token_accuracy": 0.137307870388031,
|
|
"num_tokens": 3533724.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 6.14137282371521,
|
|
"epoch": 1.6182551689359557,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000494916899571538,
|
|
"loss": 5.8942,
|
|
"mean_token_accuracy": 0.13195543885231018,
|
|
"num_tokens": 3544777.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 6.140402793884277,
|
|
"epoch": 1.6232980332829046,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004948327188848029,
|
|
"loss": 5.9305,
|
|
"mean_token_accuracy": 0.13783448413014412,
|
|
"num_tokens": 3556397.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 6.045815944671631,
|
|
"epoch": 1.6283408976298537,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004947478549529065,
|
|
"loss": 5.8717,
|
|
"mean_token_accuracy": 0.13124166131019593,
|
|
"num_tokens": 3568320.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 6.119520902633667,
|
|
"epoch": 1.6333837619768028,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004946623080396077,
|
|
"loss": 5.8831,
|
|
"mean_token_accuracy": 0.13074934855103493,
|
|
"num_tokens": 3578289.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 6.113385438919067,
|
|
"epoch": 1.638426626323752,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004945760784107884,
|
|
"loss": 5.9308,
|
|
"mean_token_accuracy": 0.12974208667874337,
|
|
"num_tokens": 3589710.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 6.048600530624389,
|
|
"epoch": 1.643469490670701,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004944891663344522,
|
|
"loss": 5.8924,
|
|
"mean_token_accuracy": 0.13714981973171234,
|
|
"num_tokens": 3601618.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 6.163399028778076,
|
|
"epoch": 1.6485123550176501,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004944015720807239,
|
|
"loss": 5.9984,
|
|
"mean_token_accuracy": 0.13000239729881286,
|
|
"num_tokens": 3612329.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 6.17548017501831,
|
|
"epoch": 1.653555219364599,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004943132959218484,
|
|
"loss": 5.9568,
|
|
"mean_token_accuracy": 0.1287740148603916,
|
|
"num_tokens": 3623887.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 6.042219734191894,
|
|
"epoch": 1.6585980837115482,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00049422433813219,
|
|
"loss": 5.8914,
|
|
"mean_token_accuracy": 0.1308971956372261,
|
|
"num_tokens": 3635484.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 6.075435590744019,
|
|
"epoch": 1.6636409480584973,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004941346989882318,
|
|
"loss": 5.7768,
|
|
"mean_token_accuracy": 0.137286975979805,
|
|
"num_tokens": 3645802.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 6.030444717407226,
|
|
"epoch": 1.6686838124054462,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004940443787685739,
|
|
"loss": 5.8496,
|
|
"mean_token_accuracy": 0.13464259207248688,
|
|
"num_tokens": 3655599.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 6.12791600227356,
|
|
"epoch": 1.6737266767523953,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004939533777539341,
|
|
"loss": 5.8728,
|
|
"mean_token_accuracy": 0.13350022509694098,
|
|
"num_tokens": 3667019.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 5.980125093460083,
|
|
"epoch": 1.6787695410993444,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004938616962271453,
|
|
"loss": 5.75,
|
|
"mean_token_accuracy": 0.14068162590265273,
|
|
"num_tokens": 3678004.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 5.995948457717896,
|
|
"epoch": 1.6838124054462935,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004937693344731561,
|
|
"loss": 5.8314,
|
|
"mean_token_accuracy": 0.13376525416970253,
|
|
"num_tokens": 3688177.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 6.045566606521606,
|
|
"epoch": 1.6888552697932426,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004936762927790286,
|
|
"loss": 5.8514,
|
|
"mean_token_accuracy": 0.13292515948414801,
|
|
"num_tokens": 3700375.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 6.1141908168792725,
|
|
"epoch": 1.6938981341401917,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004935825714339389,
|
|
"loss": 5.9147,
|
|
"mean_token_accuracy": 0.13155888468027116,
|
|
"num_tokens": 3711629.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 6.076977109909057,
|
|
"epoch": 1.6989409984871406,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004934881707291748,
|
|
"loss": 5.8381,
|
|
"mean_token_accuracy": 0.1373666323721409,
|
|
"num_tokens": 3723570.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 6.011814069747925,
|
|
"epoch": 1.7039838628340898,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004933930909581362,
|
|
"loss": 5.7614,
|
|
"mean_token_accuracy": 0.1360690712928772,
|
|
"num_tokens": 3733687.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 6.038367223739624,
|
|
"epoch": 1.7090267271810389,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004932973324163328,
|
|
"loss": 5.7817,
|
|
"mean_token_accuracy": 0.13468129858374595,
|
|
"num_tokens": 3744455.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 6.023006772994995,
|
|
"epoch": 1.7140695915279878,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004932008954013846,
|
|
"loss": 5.7638,
|
|
"mean_token_accuracy": 0.13488564565777778,
|
|
"num_tokens": 3756079.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 5.9893406391143795,
|
|
"epoch": 1.7191124558749369,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004931037802130201,
|
|
"loss": 5.756,
|
|
"mean_token_accuracy": 0.13807947114109992,
|
|
"num_tokens": 3766548.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 6.13570351600647,
|
|
"epoch": 1.724155320221886,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004930059871530755,
|
|
"loss": 5.8541,
|
|
"mean_token_accuracy": 0.13406668081879616,
|
|
"num_tokens": 3776482.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 6.041903495788574,
|
|
"epoch": 1.7291981845688351,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004929075165254938,
|
|
"loss": 5.8412,
|
|
"mean_token_accuracy": 0.13717099651694298,
|
|
"num_tokens": 3787030.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 6.037131452560425,
|
|
"epoch": 1.7342410489157842,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004928083686363241,
|
|
"loss": 5.7882,
|
|
"mean_token_accuracy": 0.13730255141854286,
|
|
"num_tokens": 3797575.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 5.995001125335693,
|
|
"epoch": 1.7392839132627333,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004927085437937204,
|
|
"loss": 5.7656,
|
|
"mean_token_accuracy": 0.13784308210015297,
|
|
"num_tokens": 3807634.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 6.07080307006836,
|
|
"epoch": 1.7443267776096822,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004926080423079404,
|
|
"loss": 5.9462,
|
|
"mean_token_accuracy": 0.12904012203216553,
|
|
"num_tokens": 3820691.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 6.13170018196106,
|
|
"epoch": 1.7493696419566314,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000492506864491345,
|
|
"loss": 5.9675,
|
|
"mean_token_accuracy": 0.12840114980936052,
|
|
"num_tokens": 3832434.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 6.071109580993652,
|
|
"epoch": 1.7544125063035805,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004924050106583976,
|
|
"loss": 5.8643,
|
|
"mean_token_accuracy": 0.13895359113812447,
|
|
"num_tokens": 3842547.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 6.109825801849365,
|
|
"epoch": 1.7594553706505294,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004923024811256618,
|
|
"loss": 5.8591,
|
|
"mean_token_accuracy": 0.1310546763241291,
|
|
"num_tokens": 3852830.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 6.065856695175171,
|
|
"epoch": 1.7644982349974785,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000492199276211802,
|
|
"loss": 5.9149,
|
|
"mean_token_accuracy": 0.12785259932279586,
|
|
"num_tokens": 3863396.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 5.958159494400024,
|
|
"epoch": 1.7695410993444276,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004920953962375813,
|
|
"loss": 5.7969,
|
|
"mean_token_accuracy": 0.13590870574116706,
|
|
"num_tokens": 3874380.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 6.051842212677002,
|
|
"epoch": 1.7745839636913767,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004919908415258612,
|
|
"loss": 5.799,
|
|
"mean_token_accuracy": 0.1365792252123356,
|
|
"num_tokens": 3883454.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 5.945011806488037,
|
|
"epoch": 1.7796268280383258,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004918856124016002,
|
|
"loss": 5.8083,
|
|
"mean_token_accuracy": 0.13445548042654992,
|
|
"num_tokens": 3893419.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 5.993125104904175,
|
|
"epoch": 1.784669692385275,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004917797091918526,
|
|
"loss": 5.8438,
|
|
"mean_token_accuracy": 0.13580774739384652,
|
|
"num_tokens": 3903760.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 6.108698225021362,
|
|
"epoch": 1.789712556732224,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004916731322257683,
|
|
"loss": 5.8932,
|
|
"mean_token_accuracy": 0.134294281154871,
|
|
"num_tokens": 3914881.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 6.168590259552002,
|
|
"epoch": 1.794755421079173,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004915658818345908,
|
|
"loss": 5.9131,
|
|
"mean_token_accuracy": 0.13629055544734,
|
|
"num_tokens": 3925774.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 6.010029935836792,
|
|
"epoch": 1.799798285426122,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004914579583516568,
|
|
"loss": 5.8059,
|
|
"mean_token_accuracy": 0.1359529212117195,
|
|
"num_tokens": 3936357.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 6.051117753982544,
|
|
"epoch": 1.804841149773071,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000491349362112395,
|
|
"loss": 5.8743,
|
|
"mean_token_accuracy": 0.13820728287100792,
|
|
"num_tokens": 3946493.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 6.09987063407898,
|
|
"epoch": 1.80988401412002,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.000491240093454325,
|
|
"loss": 5.8574,
|
|
"mean_token_accuracy": 0.1364096872508526,
|
|
"num_tokens": 3958022.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 6.003648948669434,
|
|
"epoch": 1.8149268784669692,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004911301527170563,
|
|
"loss": 5.6941,
|
|
"mean_token_accuracy": 0.1455566868185997,
|
|
"num_tokens": 3968033.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 5.99794373512268,
|
|
"epoch": 1.8199697428139183,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004910195402422873,
|
|
"loss": 5.8445,
|
|
"mean_token_accuracy": 0.13037606030702592,
|
|
"num_tokens": 3978319.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 6.049502038955689,
|
|
"epoch": 1.8250126071608674,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004909082563738042,
|
|
"loss": 5.8065,
|
|
"mean_token_accuracy": 0.14026323035359384,
|
|
"num_tokens": 3989519.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 6.044651985168457,
|
|
"epoch": 1.8300554715078166,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004907963014574797,
|
|
"loss": 5.8798,
|
|
"mean_token_accuracy": 0.13415943160653115,
|
|
"num_tokens": 4000680.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 5.993740367889404,
|
|
"epoch": 1.8350983358547657,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004906836758412722,
|
|
"loss": 5.8182,
|
|
"mean_token_accuracy": 0.13594435751438141,
|
|
"num_tokens": 4012886.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 6.022174072265625,
|
|
"epoch": 1.8401412002017146,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004905703798752248,
|
|
"loss": 5.8551,
|
|
"mean_token_accuracy": 0.13361750766634942,
|
|
"num_tokens": 4023690.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 5.973624277114868,
|
|
"epoch": 1.8451840645486637,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.000490456413911464,
|
|
"loss": 5.711,
|
|
"mean_token_accuracy": 0.1417398914694786,
|
|
"num_tokens": 4033338.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 5.8953979969024655,
|
|
"epoch": 1.8502269288956126,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004903417783041987,
|
|
"loss": 5.7168,
|
|
"mean_token_accuracy": 0.14387569576501846,
|
|
"num_tokens": 4042873.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 6.057477140426636,
|
|
"epoch": 1.8552697932425617,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004902264734097188,
|
|
"loss": 5.8698,
|
|
"mean_token_accuracy": 0.13573912978172303,
|
|
"num_tokens": 4054049.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 6.087056970596313,
|
|
"epoch": 1.8603126575895108,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004901104995863948,
|
|
"loss": 5.8457,
|
|
"mean_token_accuracy": 0.13578775897622108,
|
|
"num_tokens": 4064254.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 6.011142873764038,
|
|
"epoch": 1.86535552193646,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004899938571946759,
|
|
"loss": 5.8505,
|
|
"mean_token_accuracy": 0.13446851521730424,
|
|
"num_tokens": 4075235.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 6.1420081615447994,
|
|
"epoch": 1.870398386283409,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004898765465970894,
|
|
"loss": 5.9698,
|
|
"mean_token_accuracy": 0.12983351200819016,
|
|
"num_tokens": 4087434.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 6.035971784591675,
|
|
"epoch": 1.8754412506303582,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004897585681582393,
|
|
"loss": 5.8005,
|
|
"mean_token_accuracy": 0.13715894967317582,
|
|
"num_tokens": 4099223.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 5.914529180526733,
|
|
"epoch": 1.8804841149773073,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004896399222448053,
|
|
"loss": 5.7706,
|
|
"mean_token_accuracy": 0.14449018537998198,
|
|
"num_tokens": 4110706.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 6.079052495956421,
|
|
"epoch": 1.8855269793242562,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004895206092255417,
|
|
"loss": 5.8654,
|
|
"mean_token_accuracy": 0.13343635126948356,
|
|
"num_tokens": 4122358.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 5.899663877487183,
|
|
"epoch": 1.8905698436712053,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004894006294712761,
|
|
"loss": 5.646,
|
|
"mean_token_accuracy": 0.1418477013707161,
|
|
"num_tokens": 4132603.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 5.909395980834961,
|
|
"epoch": 1.8956127080181542,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004892799833549083,
|
|
"loss": 5.784,
|
|
"mean_token_accuracy": 0.14420581832528115,
|
|
"num_tokens": 4144708.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 6.056397914886475,
|
|
"epoch": 1.9006555723651033,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004891586712514091,
|
|
"loss": 5.8264,
|
|
"mean_token_accuracy": 0.13727928027510644,
|
|
"num_tokens": 4155751.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 5.921746397018433,
|
|
"epoch": 1.9056984367120524,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004890366935378196,
|
|
"loss": 5.8011,
|
|
"mean_token_accuracy": 0.1392076902091503,
|
|
"num_tokens": 4167368.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 6.059268188476563,
|
|
"epoch": 1.9107413010590015,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004889140505932492,
|
|
"loss": 5.7462,
|
|
"mean_token_accuracy": 0.14092126935720445,
|
|
"num_tokens": 4178601.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 5.965820837020874,
|
|
"epoch": 1.9157841654059506,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004887907427988749,
|
|
"loss": 5.8034,
|
|
"mean_token_accuracy": 0.13304574340581893,
|
|
"num_tokens": 4190421.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 5.969606542587281,
|
|
"epoch": 1.9208270297528998,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004886667705379403,
|
|
"loss": 5.7763,
|
|
"mean_token_accuracy": 0.1364285759627819,
|
|
"num_tokens": 4201916.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 5.9489280700683596,
|
|
"epoch": 1.9258698940998489,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004885421341957541,
|
|
"loss": 5.7519,
|
|
"mean_token_accuracy": 0.13658896163105966,
|
|
"num_tokens": 4213742.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 6.012122106552124,
|
|
"epoch": 1.9309127584467978,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004884168341596888,
|
|
"loss": 5.9365,
|
|
"mean_token_accuracy": 0.13323641046881676,
|
|
"num_tokens": 4226743.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 6.0101381778717045,
|
|
"epoch": 1.9359556227937469,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004882908708191797,
|
|
"loss": 5.7409,
|
|
"mean_token_accuracy": 0.1390618160367012,
|
|
"num_tokens": 4237458.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 5.934157943725586,
|
|
"epoch": 1.9409984871406958,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048816424456572403,
|
|
"loss": 5.7983,
|
|
"mean_token_accuracy": 0.13807848915457727,
|
|
"num_tokens": 4248466.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 6.041328811645508,
|
|
"epoch": 1.946041351487645,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00048803695579287905,
|
|
"loss": 5.856,
|
|
"mean_token_accuracy": 0.13591841831803322,
|
|
"num_tokens": 4260186.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 6.065542125701905,
|
|
"epoch": 1.951084215834594,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00048790900489626104,
|
|
"loss": 5.8831,
|
|
"mean_token_accuracy": 0.13168502002954482,
|
|
"num_tokens": 4272357.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 6.0779314041137695,
|
|
"epoch": 1.9561270801815431,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.0004877803922735445,
|
|
"loss": 5.8879,
|
|
"mean_token_accuracy": 0.134299448877573,
|
|
"num_tokens": 4285583.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 5.964084768295288,
|
|
"epoch": 1.9611699445284922,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004876511183244603,
|
|
"loss": 5.7835,
|
|
"mean_token_accuracy": 0.14198529571294785,
|
|
"num_tokens": 4296396.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 5.921930980682373,
|
|
"epoch": 1.9662128088754414,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048752118345079496,
|
|
"loss": 5.7425,
|
|
"mean_token_accuracy": 0.1411866396665573,
|
|
"num_tokens": 4306630.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 6.058224439620972,
|
|
"epoch": 1.9712556732223905,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00048739058805638905,
|
|
"loss": 5.8634,
|
|
"mean_token_accuracy": 0.14103320091962815,
|
|
"num_tokens": 4318488.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 5.914703607559204,
|
|
"epoch": 1.9762985375693394,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000487259332547136,
|
|
"loss": 5.6999,
|
|
"mean_token_accuracy": 0.14214230626821517,
|
|
"num_tokens": 4329604.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 5.958194160461426,
|
|
"epoch": 1.9813414019162885,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.000487127417330981,
|
|
"loss": 5.6918,
|
|
"mean_token_accuracy": 0.14373154565691948,
|
|
"num_tokens": 4341325.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 5.932970142364502,
|
|
"epoch": 1.9863842662632374,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004869948428179196,
|
|
"loss": 5.7811,
|
|
"mean_token_accuracy": 0.13495632633566856,
|
|
"num_tokens": 4353308.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 5.997287845611572,
|
|
"epoch": 1.9914271306101865,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004868616094199964,
|
|
"loss": 5.7528,
|
|
"mean_token_accuracy": 0.13617552667856217,
|
|
"num_tokens": 4364937.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 5.856275367736816,
|
|
"epoch": 1.9964699949571356,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000486727717551304,
|
|
"loss": 5.7449,
|
|
"mean_token_accuracy": 0.1407320760190487,
|
|
"num_tokens": 4376477.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 6.1092311011420355,
|
|
"epoch": 2.00100857286939,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004865931676279813,
|
|
"loss": 6.0586,
|
|
"mean_token_accuracy": 0.12319794048865636,
|
|
"num_tokens": 4386192.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 6.017110157012939,
|
|
"epoch": 2.006051437216339,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004864579600682125,
|
|
"loss": 5.4975,
|
|
"mean_token_accuracy": 0.15035134851932525,
|
|
"num_tokens": 4397170.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 5.897778940200806,
|
|
"epoch": 2.011094301563288,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048632209529222607,
|
|
"loss": 5.5602,
|
|
"mean_token_accuracy": 0.14436283856630325,
|
|
"num_tokens": 4407523.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 5.997329187393189,
|
|
"epoch": 2.016137165910237,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00048618557372229253,
|
|
"loss": 5.553,
|
|
"mean_token_accuracy": 0.15389084070920944,
|
|
"num_tokens": 4418947.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 2.016137165910237,
|
|
"eval_entropy": 5.689014786679023,
|
|
"eval_loss": 5.9144511222839355,
|
|
"eval_mean_token_accuracy": 0.13692164085521044,
|
|
"eval_num_tokens": 4418947.0,
|
|
"eval_runtime": 2.0169,
|
|
"eval_samples_per_second": 1555.844,
|
|
"eval_steps_per_second": 194.852,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 5.817630338668823,
|
|
"epoch": 2.0211800302571863,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004860483957827243,
|
|
"loss": 5.4652,
|
|
"mean_token_accuracy": 0.1533464804291725,
|
|
"num_tokens": 4429923.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 5.853796815872192,
|
|
"epoch": 2.026222894604135,
|
|
"grad_norm": 0.80078125,
|
|
"learning_rate": 0.0004859105618998735,
|
|
"loss": 5.4471,
|
|
"mean_token_accuracy": 0.15438365787267685,
|
|
"num_tokens": 4441490.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 5.8975451469421385,
|
|
"epoch": 2.031265758951084,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004857720725021311,
|
|
"loss": 5.5128,
|
|
"mean_token_accuracy": 0.14797333329916001,
|
|
"num_tokens": 4453302.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 6.047691297531128,
|
|
"epoch": 2.036308623298033,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00048563292801992534,
|
|
"loss": 5.5043,
|
|
"mean_token_accuracy": 0.14959518909454345,
|
|
"num_tokens": 4465393.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 5.749587726593018,
|
|
"epoch": 2.0413514876449823,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004854931288857203,
|
|
"loss": 5.4063,
|
|
"mean_token_accuracy": 0.1620902106165886,
|
|
"num_tokens": 4475269.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 5.848663949966431,
|
|
"epoch": 2.0463943519919314,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00048535267553401505,
|
|
"loss": 5.411,
|
|
"mean_token_accuracy": 0.16191365271806718,
|
|
"num_tokens": 4486275.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 5.873682069778442,
|
|
"epoch": 2.0514372163388805,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00048521156840134193,
|
|
"loss": 5.4703,
|
|
"mean_token_accuracy": 0.15313748717308046,
|
|
"num_tokens": 4498684.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 5.93189172744751,
|
|
"epoch": 2.0564800806858297,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000485069807926265,
|
|
"loss": 5.4798,
|
|
"mean_token_accuracy": 0.1551349028944969,
|
|
"num_tokens": 4509482.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 5.891247749328613,
|
|
"epoch": 2.0615229450327788,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004849273945493792,
|
|
"loss": 5.5479,
|
|
"mean_token_accuracy": 0.1463974080979824,
|
|
"num_tokens": 4520638.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 5.885508680343628,
|
|
"epoch": 2.066565809379728,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004847843287133084,
|
|
"loss": 5.5682,
|
|
"mean_token_accuracy": 0.1476987987756729,
|
|
"num_tokens": 4531236.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 5.963169240951538,
|
|
"epoch": 2.0716086737266766,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004846406108627048,
|
|
"loss": 5.487,
|
|
"mean_token_accuracy": 0.15453453809022905,
|
|
"num_tokens": 4542590.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 5.762261486053466,
|
|
"epoch": 2.0766515380736257,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00048449624144424665,
|
|
"loss": 5.4497,
|
|
"mean_token_accuracy": 0.15495831221342088,
|
|
"num_tokens": 4553592.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 5.953275299072265,
|
|
"epoch": 2.081694402420575,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00048435122090663744,
|
|
"loss": 5.5483,
|
|
"mean_token_accuracy": 0.1519202321767807,
|
|
"num_tokens": 4564735.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 5.947865724563599,
|
|
"epoch": 2.086737266767524,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00048420554970060424,
|
|
"loss": 5.6062,
|
|
"mean_token_accuracy": 0.15552589893341065,
|
|
"num_tokens": 4576515.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 5.902169847488404,
|
|
"epoch": 2.091780131114473,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00048405922827889677,
|
|
"loss": 5.5382,
|
|
"mean_token_accuracy": 0.15266313701868056,
|
|
"num_tokens": 4588212.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 6.011742019653321,
|
|
"epoch": 2.096822995461422,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048391225709628515,
|
|
"loss": 5.637,
|
|
"mean_token_accuracy": 0.1484152540564537,
|
|
"num_tokens": 4599482.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 5.876632976531982,
|
|
"epoch": 2.1018658598083713,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004837646366095593,
|
|
"loss": 5.5198,
|
|
"mean_token_accuracy": 0.15388129502534867,
|
|
"num_tokens": 4609920.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 5.914855480194092,
|
|
"epoch": 2.1069087241553204,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.00048361636727752716,
|
|
"loss": 5.6533,
|
|
"mean_token_accuracy": 0.1419515274465084,
|
|
"num_tokens": 4621944.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 5.884690380096435,
|
|
"epoch": 2.1119515885022695,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004834674495610131,
|
|
"loss": 5.5349,
|
|
"mean_token_accuracy": 0.15173875391483307,
|
|
"num_tokens": 4634191.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 5.89428186416626,
|
|
"epoch": 2.116994452849218,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048331788392285693,
|
|
"loss": 5.484,
|
|
"mean_token_accuracy": 0.1486421748995781,
|
|
"num_tokens": 4645655.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 5.812280130386353,
|
|
"epoch": 2.1220373171961673,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00048316767082791205,
|
|
"loss": 5.5048,
|
|
"mean_token_accuracy": 0.14627429470419884,
|
|
"num_tokens": 4656922.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 5.8477333068847654,
|
|
"epoch": 2.1270801815431164,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00048301681074304427,
|
|
"loss": 5.5262,
|
|
"mean_token_accuracy": 0.1495222344994545,
|
|
"num_tokens": 4668647.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 5.884021377563476,
|
|
"epoch": 2.1321230458900655,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048286530413713016,
|
|
"loss": 5.4444,
|
|
"mean_token_accuracy": 0.15372890681028367,
|
|
"num_tokens": 4678708.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 5.846649503707885,
|
|
"epoch": 2.1371659102370146,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00048271315148105574,
|
|
"loss": 5.5498,
|
|
"mean_token_accuracy": 0.14921148419380187,
|
|
"num_tokens": 4690962.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 5.8664710998535154,
|
|
"epoch": 2.1422087745839637,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000482560353247715,
|
|
"loss": 5.5714,
|
|
"mean_token_accuracy": 0.14479363113641738,
|
|
"num_tokens": 4701833.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 5.818386459350586,
|
|
"epoch": 2.147251638930913,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004824069099120084,
|
|
"loss": 5.4388,
|
|
"mean_token_accuracy": 0.1544934794306755,
|
|
"num_tokens": 4712756.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 5.841196012496948,
|
|
"epoch": 2.152294503277862,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004822528219508413,
|
|
"loss": 5.4967,
|
|
"mean_token_accuracy": 0.1458820104598999,
|
|
"num_tokens": 4723226.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 5.872856712341308,
|
|
"epoch": 2.157337367624811,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004820980898431227,
|
|
"loss": 5.5238,
|
|
"mean_token_accuracy": 0.14719147384166717,
|
|
"num_tokens": 4734703.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 5.852089166641235,
|
|
"epoch": 2.1623802319717598,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004819427140697635,
|
|
"loss": 5.5187,
|
|
"mean_token_accuracy": 0.14853617995977403,
|
|
"num_tokens": 4745356.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 5.8280689239501955,
|
|
"epoch": 2.167423096318709,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004817866951136753,
|
|
"loss": 5.4844,
|
|
"mean_token_accuracy": 0.15176524072885514,
|
|
"num_tokens": 4756342.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 5.876274490356446,
|
|
"epoch": 2.172465960665658,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00048163003345976847,
|
|
"loss": 5.5735,
|
|
"mean_token_accuracy": 0.1529287338256836,
|
|
"num_tokens": 4768782.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 5.822108936309815,
|
|
"epoch": 2.177508825012607,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004814727295949512,
|
|
"loss": 5.485,
|
|
"mean_token_accuracy": 0.15548587143421172,
|
|
"num_tokens": 4779807.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 5.835571193695069,
|
|
"epoch": 2.1825516893595562,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00048131478400812735,
|
|
"loss": 5.5101,
|
|
"mean_token_accuracy": 0.15026734471321107,
|
|
"num_tokens": 4789924.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 5.919145202636718,
|
|
"epoch": 2.1875945537065054,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004811561971901956,
|
|
"loss": 5.5439,
|
|
"mean_token_accuracy": 0.15081341117620467,
|
|
"num_tokens": 4801112.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 5.792446088790894,
|
|
"epoch": 2.1926374180534545,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004809969696340474,
|
|
"loss": 5.4919,
|
|
"mean_token_accuracy": 0.14936817735433577,
|
|
"num_tokens": 4811777.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 5.776220083236694,
|
|
"epoch": 2.1976802824004036,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048083710183456554,
|
|
"loss": 5.4067,
|
|
"mean_token_accuracy": 0.15529889166355132,
|
|
"num_tokens": 4822666.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 5.791893243789673,
|
|
"epoch": 2.2027231467473527,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048067659428862306,
|
|
"loss": 5.4647,
|
|
"mean_token_accuracy": 0.1516457602381706,
|
|
"num_tokens": 4833315.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 5.886841583251953,
|
|
"epoch": 2.2077660110943014,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00048051544749508096,
|
|
"loss": 5.5199,
|
|
"mean_token_accuracy": 0.15019209831953048,
|
|
"num_tokens": 4844606.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 5.847221899032593,
|
|
"epoch": 2.2128088754412505,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004803536619547872,
|
|
"loss": 5.5872,
|
|
"mean_token_accuracy": 0.14769265949726104,
|
|
"num_tokens": 4857460.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 5.841250228881836,
|
|
"epoch": 2.2178517397881996,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000480191238170575,
|
|
"loss": 5.4963,
|
|
"mean_token_accuracy": 0.15116168111562728,
|
|
"num_tokens": 4869289.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 5.707509851455688,
|
|
"epoch": 2.2228946041351487,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004800281766472612,
|
|
"loss": 5.4637,
|
|
"mean_token_accuracy": 0.1537894457578659,
|
|
"num_tokens": 4878912.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 5.857531356811523,
|
|
"epoch": 2.227937468482098,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00047986447789164486,
|
|
"loss": 5.5377,
|
|
"mean_token_accuracy": 0.1474810093641281,
|
|
"num_tokens": 4891312.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 5.877925825119019,
|
|
"epoch": 2.232980332829047,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004797001424125054,
|
|
"loss": 5.5529,
|
|
"mean_token_accuracy": 0.14680176675319673,
|
|
"num_tokens": 4902877.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 5.8398065090179445,
|
|
"epoch": 2.238023197175996,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00047953517072060124,
|
|
"loss": 5.4594,
|
|
"mean_token_accuracy": 0.15368951857089996,
|
|
"num_tokens": 4914145.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 5.851288366317749,
|
|
"epoch": 2.243066061522945,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004793695633286684,
|
|
"loss": 5.5877,
|
|
"mean_token_accuracy": 0.14439464658498763,
|
|
"num_tokens": 4924647.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 5.796383953094482,
|
|
"epoch": 2.2481089258698943,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00047920332075141825,
|
|
"loss": 5.4375,
|
|
"mean_token_accuracy": 0.15170739442110062,
|
|
"num_tokens": 4934425.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 5.744509029388428,
|
|
"epoch": 2.253151790216843,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004790364435055367,
|
|
"loss": 5.4521,
|
|
"mean_token_accuracy": 0.1572572872042656,
|
|
"num_tokens": 4945045.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 5.751601648330689,
|
|
"epoch": 2.258194654563792,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000478868932109682,
|
|
"loss": 5.4538,
|
|
"mean_token_accuracy": 0.1518895298242569,
|
|
"num_tokens": 4956225.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 5.847568607330322,
|
|
"epoch": 2.263237518910741,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004787007870844835,
|
|
"loss": 5.4331,
|
|
"mean_token_accuracy": 0.15858909338712693,
|
|
"num_tokens": 4968591.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 5.770700931549072,
|
|
"epoch": 2.2682803832576903,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00047853200895253976,
|
|
"loss": 5.5117,
|
|
"mean_token_accuracy": 0.15076763331890106,
|
|
"num_tokens": 4978612.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 5.728929138183593,
|
|
"epoch": 2.2733232476046394,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00047836259823841716,
|
|
"loss": 5.3967,
|
|
"mean_token_accuracy": 0.15569760352373124,
|
|
"num_tokens": 4990039.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 5.8394993305206295,
|
|
"epoch": 2.2783661119515886,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00047819255546864797,
|
|
"loss": 5.5334,
|
|
"mean_token_accuracy": 0.15142289623618127,
|
|
"num_tokens": 5001978.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 5.773989820480347,
|
|
"epoch": 2.2834089762985377,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004780218811717293,
|
|
"loss": 5.4387,
|
|
"mean_token_accuracy": 0.1542734533548355,
|
|
"num_tokens": 5012559.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 5.786451053619385,
|
|
"epoch": 2.288451840645487,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00047785057587812057,
|
|
"loss": 5.4895,
|
|
"mean_token_accuracy": 0.15763220936059952,
|
|
"num_tokens": 5022944.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 5.834816360473633,
|
|
"epoch": 2.293494704992436,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00047767864012024266,
|
|
"loss": 5.4906,
|
|
"mean_token_accuracy": 0.15608912259340285,
|
|
"num_tokens": 5033771.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 5.748194980621338,
|
|
"epoch": 2.2985375693393846,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00047750607443247585,
|
|
"loss": 5.4629,
|
|
"mean_token_accuracy": 0.15616362243890763,
|
|
"num_tokens": 5045631.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 5.842692708969116,
|
|
"epoch": 2.3035804336863337,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00047733287935115837,
|
|
"loss": 5.4386,
|
|
"mean_token_accuracy": 0.1483730487525463,
|
|
"num_tokens": 5056392.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 5.736159992218018,
|
|
"epoch": 2.308623298033283,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004771590554145844,
|
|
"loss": 5.5127,
|
|
"mean_token_accuracy": 0.15482610762119292,
|
|
"num_tokens": 5066420.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 5.771495628356933,
|
|
"epoch": 2.313666162380232,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00047698460316300285,
|
|
"loss": 5.4948,
|
|
"mean_token_accuracy": 0.1513727217912674,
|
|
"num_tokens": 5075608.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 5.858992195129394,
|
|
"epoch": 2.318709026727181,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004768095231386153,
|
|
"loss": 5.4808,
|
|
"mean_token_accuracy": 0.15711421966552735,
|
|
"num_tokens": 5086304.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 5.822066783905029,
|
|
"epoch": 2.32375189107413,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004766338158855744,
|
|
"loss": 5.5389,
|
|
"mean_token_accuracy": 0.14951140880584718,
|
|
"num_tokens": 5096811.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 5.781665992736817,
|
|
"epoch": 2.3287947554210793,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004764574819499826,
|
|
"loss": 5.5377,
|
|
"mean_token_accuracy": 0.14901492297649382,
|
|
"num_tokens": 5107118.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 5.845264434814453,
|
|
"epoch": 2.3338376197680284,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00047628052187988955,
|
|
"loss": 5.5146,
|
|
"mean_token_accuracy": 0.1571188375353813,
|
|
"num_tokens": 5118362.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 5.801714611053467,
|
|
"epoch": 2.338880484114977,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004761029362252914,
|
|
"loss": 5.4965,
|
|
"mean_token_accuracy": 0.15613144487142563,
|
|
"num_tokens": 5128705.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 5.731414842605591,
|
|
"epoch": 2.343923348461926,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00047592472553812856,
|
|
"loss": 5.4049,
|
|
"mean_token_accuracy": 0.15618726015090942,
|
|
"num_tokens": 5139952.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 5.828210353851318,
|
|
"epoch": 2.3489662128088753,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004757458903722837,
|
|
"loss": 5.5705,
|
|
"mean_token_accuracy": 0.14632575511932372,
|
|
"num_tokens": 5149453.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 5.846777057647705,
|
|
"epoch": 2.3540090771558244,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00047556643128358075,
|
|
"loss": 5.5739,
|
|
"mean_token_accuracy": 0.151758673787117,
|
|
"num_tokens": 5160951.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 5.809584617614746,
|
|
"epoch": 2.3590519415027735,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00047538634882978277,
|
|
"loss": 5.5184,
|
|
"mean_token_accuracy": 0.14951531440019608,
|
|
"num_tokens": 5171172.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 5.904595851898193,
|
|
"epoch": 2.3640948058497226,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00047520564357059013,
|
|
"loss": 5.5472,
|
|
"mean_token_accuracy": 0.14950637370347977,
|
|
"num_tokens": 5183178.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 5.726088285446167,
|
|
"epoch": 2.3691376701966718,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004750243160676389,
|
|
"loss": 5.4694,
|
|
"mean_token_accuracy": 0.15826211422681807,
|
|
"num_tokens": 5193719.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 5.761527395248413,
|
|
"epoch": 2.374180534543621,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004748423668844992,
|
|
"loss": 5.4126,
|
|
"mean_token_accuracy": 0.16288568675518036,
|
|
"num_tokens": 5203873.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 5.756472539901734,
|
|
"epoch": 2.37922339889057,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004746597965866733,
|
|
"loss": 5.4822,
|
|
"mean_token_accuracy": 0.1508549377322197,
|
|
"num_tokens": 5213682.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 5.725824213027954,
|
|
"epoch": 2.384266263237519,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004744766057415938,
|
|
"loss": 5.4072,
|
|
"mean_token_accuracy": 0.15313855856657027,
|
|
"num_tokens": 5224850.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 5.775016498565674,
|
|
"epoch": 2.389309127584468,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004742927949186222,
|
|
"loss": 5.4885,
|
|
"mean_token_accuracy": 0.15430940687656403,
|
|
"num_tokens": 5237256.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 5.739787673950195,
|
|
"epoch": 2.394351991931417,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00047410836468904664,
|
|
"loss": 5.5326,
|
|
"mean_token_accuracy": 0.14594767913222312,
|
|
"num_tokens": 5247638.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 5.7555560111999515,
|
|
"epoch": 2.399394856278366,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00047392331562608073,
|
|
"loss": 5.4552,
|
|
"mean_token_accuracy": 0.15363992750644684,
|
|
"num_tokens": 5259438.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 5.838210725784302,
|
|
"epoch": 2.404437720625315,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004737376483048609,
|
|
"loss": 5.4469,
|
|
"mean_token_accuracy": 0.15905401185154916,
|
|
"num_tokens": 5269809.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 5.714035558700561,
|
|
"epoch": 2.4094805849722642,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004735513633024458,
|
|
"loss": 5.4964,
|
|
"mean_token_accuracy": 0.15178426951169968,
|
|
"num_tokens": 5280710.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 5.826812648773194,
|
|
"epoch": 2.4145234493192134,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.0004733644611978134,
|
|
"loss": 5.4946,
|
|
"mean_token_accuracy": 0.15057420432567598,
|
|
"num_tokens": 5293162.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 5.794790458679199,
|
|
"epoch": 2.4195663136661625,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004731769425718597,
|
|
"loss": 5.5328,
|
|
"mean_token_accuracy": 0.15011205822229384,
|
|
"num_tokens": 5304510.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 5.676998710632324,
|
|
"epoch": 2.4246091780131116,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000472988808007397,
|
|
"loss": 5.3985,
|
|
"mean_token_accuracy": 0.15580725222826003,
|
|
"num_tokens": 5314962.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 5.756295156478882,
|
|
"epoch": 2.4296520423600603,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00047280005808915176,
|
|
"loss": 5.4808,
|
|
"mean_token_accuracy": 0.14887662678956987,
|
|
"num_tokens": 5325314.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 5.852035331726074,
|
|
"epoch": 2.4346949067070094,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004726106934037631,
|
|
"loss": 5.5371,
|
|
"mean_token_accuracy": 0.15002442747354508,
|
|
"num_tokens": 5336408.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 5.760966157913208,
|
|
"epoch": 2.4397377710539585,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004724207145397809,
|
|
"loss": 5.4233,
|
|
"mean_token_accuracy": 0.15475991666316985,
|
|
"num_tokens": 5348014.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 5.703101778030396,
|
|
"epoch": 2.4447806354009076,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00047223012208766384,
|
|
"loss": 5.4472,
|
|
"mean_token_accuracy": 0.16211252957582473,
|
|
"num_tokens": 5357275.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 5.834553337097168,
|
|
"epoch": 2.4498234997478567,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004720389166397777,
|
|
"loss": 5.4293,
|
|
"mean_token_accuracy": 0.15148970633745193,
|
|
"num_tokens": 5367822.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 5.734790372848511,
|
|
"epoch": 2.454866364094806,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00047184709879039325,
|
|
"loss": 5.4705,
|
|
"mean_token_accuracy": 0.15349729210138321,
|
|
"num_tokens": 5379193.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 5.71196722984314,
|
|
"epoch": 2.459909228441755,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00047165466913568497,
|
|
"loss": 5.4403,
|
|
"mean_token_accuracy": 0.15672384798526764,
|
|
"num_tokens": 5389825.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 5.632027435302734,
|
|
"epoch": 2.464952092788704,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004714616282737285,
|
|
"loss": 5.3503,
|
|
"mean_token_accuracy": 0.1631494030356407,
|
|
"num_tokens": 5400318.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 5.746574640274048,
|
|
"epoch": 2.469994957135653,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004712679768044995,
|
|
"loss": 5.5016,
|
|
"mean_token_accuracy": 0.15050007700920104,
|
|
"num_tokens": 5411432.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 5.728002405166626,
|
|
"epoch": 2.4750378214826023,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.000471073715329871,
|
|
"loss": 5.5349,
|
|
"mean_token_accuracy": 0.151277057826519,
|
|
"num_tokens": 5422662.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 5.7511303424835205,
|
|
"epoch": 2.4800806858295514,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004708788444536123,
|
|
"loss": 5.5604,
|
|
"mean_token_accuracy": 0.15142618864774704,
|
|
"num_tokens": 5434976.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 5.813687133789062,
|
|
"epoch": 2.4851235501765,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004706833647813865,
|
|
"loss": 5.4447,
|
|
"mean_token_accuracy": 0.14822918325662612,
|
|
"num_tokens": 5445607.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 5.868877506256103,
|
|
"epoch": 2.4901664145234492,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00047048727692074893,
|
|
"loss": 5.5974,
|
|
"mean_token_accuracy": 0.14369846731424332,
|
|
"num_tokens": 5455793.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 5.750569772720337,
|
|
"epoch": 2.4952092788703983,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004702905814811452,
|
|
"loss": 5.5456,
|
|
"mean_token_accuracy": 0.14835046082735062,
|
|
"num_tokens": 5466800.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 5.769549560546875,
|
|
"epoch": 2.5002521432173475,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00047009327907390926,
|
|
"loss": 5.4905,
|
|
"mean_token_accuracy": 0.15407976657152175,
|
|
"num_tokens": 5477875.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 5.7594445705413815,
|
|
"epoch": 2.5052950075642966,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004698953703122616,
|
|
"loss": 5.5202,
|
|
"mean_token_accuracy": 0.15061493963003159,
|
|
"num_tokens": 5488193.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 5.930981731414795,
|
|
"epoch": 2.5103378719112457,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00046969685581130715,
|
|
"loss": 5.6699,
|
|
"mean_token_accuracy": 0.14534270614385605,
|
|
"num_tokens": 5500179.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 5.801249027252197,
|
|
"epoch": 2.515380736258195,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00046949773618803353,
|
|
"loss": 5.5361,
|
|
"mean_token_accuracy": 0.14681192636489868,
|
|
"num_tokens": 5513413.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 5.830780172348023,
|
|
"epoch": 2.5204236006051435,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00046929801206130925,
|
|
"loss": 5.5458,
|
|
"mean_token_accuracy": 0.1534034162759781,
|
|
"num_tokens": 5524697.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"epoch": 2.5204236006051435,
|
|
"eval_entropy": 5.592608311097434,
|
|
"eval_loss": 5.839509010314941,
|
|
"eval_mean_token_accuracy": 0.14263392531765628,
|
|
"eval_num_tokens": 5524697.0,
|
|
"eval_runtime": 2.0064,
|
|
"eval_samples_per_second": 1563.957,
|
|
"eval_steps_per_second": 195.868,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 5.831315755844116,
|
|
"epoch": 2.5254664649520926,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00046909768405188135,
|
|
"loss": 5.5966,
|
|
"mean_token_accuracy": 0.152495813369751,
|
|
"num_tokens": 5536816.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 5.793387174606323,
|
|
"epoch": 2.5305093292990417,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00046889675278237404,
|
|
"loss": 5.563,
|
|
"mean_token_accuracy": 0.15201305076479912,
|
|
"num_tokens": 5549329.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 5.723847055435181,
|
|
"epoch": 2.535552193645991,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004686952188772863,
|
|
"loss": 5.4199,
|
|
"mean_token_accuracy": 0.15851414501667022,
|
|
"num_tokens": 5560015.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 5.772934675216675,
|
|
"epoch": 2.54059505799294,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00046849308296299005,
|
|
"loss": 5.5512,
|
|
"mean_token_accuracy": 0.1466063901782036,
|
|
"num_tokens": 5571031.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 5.8602790355682375,
|
|
"epoch": 2.545637922339889,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004682903456677285,
|
|
"loss": 5.5197,
|
|
"mean_token_accuracy": 0.1541677251458168,
|
|
"num_tokens": 5582283.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 5.7739396572113035,
|
|
"epoch": 2.550680786686838,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00046808700762161385,
|
|
"loss": 5.5485,
|
|
"mean_token_accuracy": 0.15232178419828415,
|
|
"num_tokens": 5593185.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 5.727773284912109,
|
|
"epoch": 2.5557236510337873,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004678830694566253,
|
|
"loss": 5.443,
|
|
"mean_token_accuracy": 0.15851565152406694,
|
|
"num_tokens": 5604147.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 5.687780475616455,
|
|
"epoch": 2.5607665153807364,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004676785318066075,
|
|
"loss": 5.4492,
|
|
"mean_token_accuracy": 0.15316415429115296,
|
|
"num_tokens": 5615074.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 5.706979608535766,
|
|
"epoch": 2.5658093797276855,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004674733953072682,
|
|
"loss": 5.4952,
|
|
"mean_token_accuracy": 0.15433439165353774,
|
|
"num_tokens": 5626599.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 5.71798996925354,
|
|
"epoch": 2.5708522440746346,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004672676605961763,
|
|
"loss": 5.4109,
|
|
"mean_token_accuracy": 0.1548145070672035,
|
|
"num_tokens": 5637378.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 5.7578370571136475,
|
|
"epoch": 2.5758951084215833,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00046706132831276007,
|
|
"loss": 5.4573,
|
|
"mean_token_accuracy": 0.15992680490016936,
|
|
"num_tokens": 5649152.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 5.781821250915527,
|
|
"epoch": 2.5809379727685324,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00046685439909830486,
|
|
"loss": 5.5374,
|
|
"mean_token_accuracy": 0.14957017451524734,
|
|
"num_tokens": 5660476.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 5.686552333831787,
|
|
"epoch": 2.5859808371154815,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004666468735959516,
|
|
"loss": 5.3859,
|
|
"mean_token_accuracy": 0.16021039932966233,
|
|
"num_tokens": 5672178.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 5.751861190795898,
|
|
"epoch": 2.5910237014624307,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004664387524506943,
|
|
"loss": 5.4905,
|
|
"mean_token_accuracy": 0.1506672605872154,
|
|
"num_tokens": 5683988.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 5.813842916488648,
|
|
"epoch": 2.59606656580938,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004662300363093784,
|
|
"loss": 5.5207,
|
|
"mean_token_accuracy": 0.15429271310567855,
|
|
"num_tokens": 5695009.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 5.76313214302063,
|
|
"epoch": 2.601109430156329,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004660207258206983,
|
|
"loss": 5.5124,
|
|
"mean_token_accuracy": 0.15211643725633622,
|
|
"num_tokens": 5706838.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 5.7128678321838375,
|
|
"epoch": 2.606152294503278,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00046581082163519585,
|
|
"loss": 5.4025,
|
|
"mean_token_accuracy": 0.1623998686671257,
|
|
"num_tokens": 5717846.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 5.674966955184937,
|
|
"epoch": 2.6111951588502267,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004656003244052582,
|
|
"loss": 5.4086,
|
|
"mean_token_accuracy": 0.15809899121522902,
|
|
"num_tokens": 5727519.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 5.711035633087159,
|
|
"epoch": 2.616238023197176,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004653892347851155,
|
|
"loss": 5.4894,
|
|
"mean_token_accuracy": 0.1479623481631279,
|
|
"num_tokens": 5738717.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 5.829298400878907,
|
|
"epoch": 2.621280887544125,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004651775534308392,
|
|
"loss": 5.5748,
|
|
"mean_token_accuracy": 0.1517823338508606,
|
|
"num_tokens": 5750095.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 5.73341646194458,
|
|
"epoch": 2.626323751891074,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004649652810003398,
|
|
"loss": 5.4066,
|
|
"mean_token_accuracy": 0.1623570814728737,
|
|
"num_tokens": 5760282.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 5.715233659744262,
|
|
"epoch": 2.631366616238023,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00046475241815336504,
|
|
"loss": 5.4593,
|
|
"mean_token_accuracy": 0.15093406066298484,
|
|
"num_tokens": 5770547.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 5.733910846710205,
|
|
"epoch": 2.6364094805849723,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004645389655514974,
|
|
"loss": 5.4577,
|
|
"mean_token_accuracy": 0.1595013678073883,
|
|
"num_tokens": 5780801.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 5.796187686920166,
|
|
"epoch": 2.6414523449319214,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004643249238581525,
|
|
"loss": 5.4926,
|
|
"mean_token_accuracy": 0.14881011396646499,
|
|
"num_tokens": 5791851.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 5.74654221534729,
|
|
"epoch": 2.6464952092788705,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004641102937385769,
|
|
"loss": 5.5607,
|
|
"mean_token_accuracy": 0.1491996631026268,
|
|
"num_tokens": 5802551.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 5.71617021560669,
|
|
"epoch": 2.6515380736258196,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004638950758598461,
|
|
"loss": 5.4266,
|
|
"mean_token_accuracy": 0.15527755171060562,
|
|
"num_tokens": 5813376.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 5.730456876754761,
|
|
"epoch": 2.6565809379727687,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00046367927089086203,
|
|
"loss": 5.4781,
|
|
"mean_token_accuracy": 0.15348075479269027,
|
|
"num_tokens": 5824634.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 5.731200265884399,
|
|
"epoch": 2.661623802319718,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.00046346287950235163,
|
|
"loss": 5.3953,
|
|
"mean_token_accuracy": 0.1604924902319908,
|
|
"num_tokens": 5836771.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 5.7149101257324215,
|
|
"epoch": 2.6666666666666665,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004632459023668642,
|
|
"loss": 5.4142,
|
|
"mean_token_accuracy": 0.15535163134336472,
|
|
"num_tokens": 5847695.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 5.694322967529297,
|
|
"epoch": 2.6717095310136156,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004630283401587699,
|
|
"loss": 5.4941,
|
|
"mean_token_accuracy": 0.15507960841059684,
|
|
"num_tokens": 5859843.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 5.773857355117798,
|
|
"epoch": 2.6767523953605648,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004628101935542568,
|
|
"loss": 5.4785,
|
|
"mean_token_accuracy": 0.15669242292642593,
|
|
"num_tokens": 5871093.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 5.6854588985443115,
|
|
"epoch": 2.681795259707514,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00046259146323132986,
|
|
"loss": 5.4443,
|
|
"mean_token_accuracy": 0.16291586458683013,
|
|
"num_tokens": 5882235.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 5.627847146987915,
|
|
"epoch": 2.686838124054463,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00046237214986980776,
|
|
"loss": 5.3706,
|
|
"mean_token_accuracy": 0.1583113193511963,
|
|
"num_tokens": 5892567.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 5.8078423023223875,
|
|
"epoch": 2.691880988401412,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004621522541513216,
|
|
"loss": 5.534,
|
|
"mean_token_accuracy": 0.15646833181381226,
|
|
"num_tokens": 5904849.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 5.759836292266845,
|
|
"epoch": 2.696923852748361,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004619317767593123,
|
|
"loss": 5.4585,
|
|
"mean_token_accuracy": 0.15698864459991455,
|
|
"num_tokens": 5914873.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 5.778701591491699,
|
|
"epoch": 2.70196671709531,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00046171071837902873,
|
|
"loss": 5.4944,
|
|
"mean_token_accuracy": 0.15027303397655487,
|
|
"num_tokens": 5926312.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 5.70425443649292,
|
|
"epoch": 2.707009581442259,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004614890796975254,
|
|
"loss": 5.5446,
|
|
"mean_token_accuracy": 0.15716080814599992,
|
|
"num_tokens": 5939665.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 5.73945951461792,
|
|
"epoch": 2.712052445789208,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004612668614036605,
|
|
"loss": 5.4582,
|
|
"mean_token_accuracy": 0.1625843971967697,
|
|
"num_tokens": 5949999.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 5.6814555644989015,
|
|
"epoch": 2.7170953101361572,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004610440641880934,
|
|
"loss": 5.43,
|
|
"mean_token_accuracy": 0.16091229021549225,
|
|
"num_tokens": 5960026.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 5.6776024341583256,
|
|
"epoch": 2.7221381744831064,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00046082068874328324,
|
|
"loss": 5.4229,
|
|
"mean_token_accuracy": 0.15172194540500641,
|
|
"num_tokens": 5971505.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 5.7885068416595455,
|
|
"epoch": 2.7271810388300555,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00046059673576348596,
|
|
"loss": 5.4206,
|
|
"mean_token_accuracy": 0.15291666239500046,
|
|
"num_tokens": 5982919.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 5.690246200561523,
|
|
"epoch": 2.7322239031770046,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00046037220594475264,
|
|
"loss": 5.4181,
|
|
"mean_token_accuracy": 0.1541130945086479,
|
|
"num_tokens": 5993146.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 5.677676868438721,
|
|
"epoch": 2.7372667675239537,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000460147099984927,
|
|
"loss": 5.4411,
|
|
"mean_token_accuracy": 0.15763067305088044,
|
|
"num_tokens": 6004864.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 5.704281234741211,
|
|
"epoch": 2.742309631870903,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004599214185836438,
|
|
"loss": 5.4446,
|
|
"mean_token_accuracy": 0.1548577442765236,
|
|
"num_tokens": 6016440.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 5.667774200439453,
|
|
"epoch": 2.747352496217852,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00045969516244232574,
|
|
"loss": 5.3853,
|
|
"mean_token_accuracy": 0.16592003554105758,
|
|
"num_tokens": 6026227.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 5.672974157333374,
|
|
"epoch": 2.752395360564801,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.00045946833226418235,
|
|
"loss": 5.4411,
|
|
"mean_token_accuracy": 0.14951107129454613,
|
|
"num_tokens": 6038290.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 5.7321044921875,
|
|
"epoch": 2.7574382249117497,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004592409287542069,
|
|
"loss": 5.4785,
|
|
"mean_token_accuracy": 0.15505492836236953,
|
|
"num_tokens": 6049611.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 5.75576114654541,
|
|
"epoch": 2.762481089258699,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004590129526191748,
|
|
"loss": 5.4744,
|
|
"mean_token_accuracy": 0.15591098368167877,
|
|
"num_tokens": 6060270.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 5.712100553512573,
|
|
"epoch": 2.767523953605648,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004587844045676413,
|
|
"loss": 5.4833,
|
|
"mean_token_accuracy": 0.15551854074001312,
|
|
"num_tokens": 6071380.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 5.756773853302002,
|
|
"epoch": 2.772566817952597,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00045855528530993874,
|
|
"loss": 5.5192,
|
|
"mean_token_accuracy": 0.14727859497070311,
|
|
"num_tokens": 6082655.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 5.67363543510437,
|
|
"epoch": 2.777609682299546,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00045832559555817525,
|
|
"loss": 5.3987,
|
|
"mean_token_accuracy": 0.15972242504358292,
|
|
"num_tokens": 6094446.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 5.814352178573609,
|
|
"epoch": 2.7826525466464953,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00045809533602623185,
|
|
"loss": 5.5011,
|
|
"mean_token_accuracy": 0.15297414362430573,
|
|
"num_tokens": 6105606.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 5.644446802139282,
|
|
"epoch": 2.7876954109934444,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004578645074297603,
|
|
"loss": 5.4594,
|
|
"mean_token_accuracy": 0.15394866168498994,
|
|
"num_tokens": 6117743.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 5.717522144317627,
|
|
"epoch": 2.792738275340393,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00045763311048618144,
|
|
"loss": 5.5133,
|
|
"mean_token_accuracy": 0.15421534180641175,
|
|
"num_tokens": 6128324.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 5.728625106811523,
|
|
"epoch": 2.797781139687342,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004574011459146822,
|
|
"loss": 5.4733,
|
|
"mean_token_accuracy": 0.15346226394176482,
|
|
"num_tokens": 6138742.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 5.755233669281006,
|
|
"epoch": 2.8028240040342913,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00045716861443621386,
|
|
"loss": 5.4704,
|
|
"mean_token_accuracy": 0.1576635494828224,
|
|
"num_tokens": 6149990.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 5.668575716018677,
|
|
"epoch": 2.8078668683812404,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00045693551677348967,
|
|
"loss": 5.4862,
|
|
"mean_token_accuracy": 0.15542775243520737,
|
|
"num_tokens": 6160281.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 5.717032241821289,
|
|
"epoch": 2.8129097327281896,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004567018536509823,
|
|
"loss": 5.5101,
|
|
"mean_token_accuracy": 0.15712994039058686,
|
|
"num_tokens": 6171501.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 5.83033299446106,
|
|
"epoch": 2.8179525970751387,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00045646762579492245,
|
|
"loss": 5.5961,
|
|
"mean_token_accuracy": 0.14847124814987184,
|
|
"num_tokens": 6183401.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 5.6827051639556885,
|
|
"epoch": 2.822995461422088,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00045623283393329546,
|
|
"loss": 5.4071,
|
|
"mean_token_accuracy": 0.15656230449676514,
|
|
"num_tokens": 6194451.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 5.656178283691406,
|
|
"epoch": 2.828038325769037,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00045599747879584,
|
|
"loss": 5.372,
|
|
"mean_token_accuracy": 0.16056768000125884,
|
|
"num_tokens": 6204410.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 5.7147438526153564,
|
|
"epoch": 2.833081190115986,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004557615611140451,
|
|
"loss": 5.4197,
|
|
"mean_token_accuracy": 0.15314737409353257,
|
|
"num_tokens": 6215091.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 5.732995271682739,
|
|
"epoch": 2.838124054462935,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004555250816211485,
|
|
"loss": 5.4206,
|
|
"mean_token_accuracy": 0.15699324756860733,
|
|
"num_tokens": 6227201.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 5.727254390716553,
|
|
"epoch": 2.8431669188098843,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00045528804105213395,
|
|
"loss": 5.4894,
|
|
"mean_token_accuracy": 0.15545756816864015,
|
|
"num_tokens": 6237067.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 5.696281814575196,
|
|
"epoch": 2.848209783156833,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004550504401437289,
|
|
"loss": 5.4557,
|
|
"mean_token_accuracy": 0.16083767712116243,
|
|
"num_tokens": 6247269.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 5.741560411453247,
|
|
"epoch": 2.853252647503782,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00045481227963440235,
|
|
"loss": 5.5007,
|
|
"mean_token_accuracy": 0.15136996209621428,
|
|
"num_tokens": 6259094.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 5.688735246658325,
|
|
"epoch": 2.858295511850731,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00045457356026436295,
|
|
"loss": 5.4331,
|
|
"mean_token_accuracy": 0.15484428852796556,
|
|
"num_tokens": 6270012.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 5.675854349136353,
|
|
"epoch": 2.8633383761976803,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00045433428277555577,
|
|
"loss": 5.4437,
|
|
"mean_token_accuracy": 0.15472084283828735,
|
|
"num_tokens": 6280936.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 5.678705930709839,
|
|
"epoch": 2.8683812405446294,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004540944479116609,
|
|
"loss": 5.4328,
|
|
"mean_token_accuracy": 0.1551497146487236,
|
|
"num_tokens": 6291471.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 5.703444719314575,
|
|
"epoch": 2.8734241048915785,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.0004538540564180906,
|
|
"loss": 5.4985,
|
|
"mean_token_accuracy": 0.1513482004404068,
|
|
"num_tokens": 6303717.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 5.807976579666137,
|
|
"epoch": 2.8784669692385276,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004536131090419871,
|
|
"loss": 5.4788,
|
|
"mean_token_accuracy": 0.1565277710556984,
|
|
"num_tokens": 6312960.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 5.785744571685791,
|
|
"epoch": 2.8835098335854763,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004533716065322205,
|
|
"loss": 5.478,
|
|
"mean_token_accuracy": 0.15648596733808517,
|
|
"num_tokens": 6322558.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 5.783140516281128,
|
|
"epoch": 2.8885526979324254,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004531295496393861,
|
|
"loss": 5.5491,
|
|
"mean_token_accuracy": 0.15415573865175247,
|
|
"num_tokens": 6334763.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 5.720203304290772,
|
|
"epoch": 2.8935955622793745,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004528869391158023,
|
|
"loss": 5.4865,
|
|
"mean_token_accuracy": 0.15195681005716324,
|
|
"num_tokens": 6346678.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 5.7926136493682865,
|
|
"epoch": 2.8986384266263236,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00045264377571550834,
|
|
"loss": 5.4638,
|
|
"mean_token_accuracy": 0.15796593576669693,
|
|
"num_tokens": 6357495.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 5.6748387813568115,
|
|
"epoch": 2.9036812909732728,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004524000601942615,
|
|
"loss": 5.404,
|
|
"mean_token_accuracy": 0.15420129969716073,
|
|
"num_tokens": 6368557.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 5.678926944732666,
|
|
"epoch": 2.908724155320222,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004521557933095354,
|
|
"loss": 5.419,
|
|
"mean_token_accuracy": 0.1565443456172943,
|
|
"num_tokens": 6378471.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 5.691143369674682,
|
|
"epoch": 2.913767019667171,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000451910975820517,
|
|
"loss": 5.5043,
|
|
"mean_token_accuracy": 0.15166358798742294,
|
|
"num_tokens": 6390826.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 5.718291711807251,
|
|
"epoch": 2.91880988401412,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00045166560848810476,
|
|
"loss": 5.3554,
|
|
"mean_token_accuracy": 0.15250223875045776,
|
|
"num_tokens": 6400438.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 5.62859582901001,
|
|
"epoch": 2.9238527483610692,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00045141969207490606,
|
|
"loss": 5.4654,
|
|
"mean_token_accuracy": 0.15300080478191375,
|
|
"num_tokens": 6411533.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 5.635191631317139,
|
|
"epoch": 2.9288956127080183,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004511732273452347,
|
|
"loss": 5.4016,
|
|
"mean_token_accuracy": 0.16076576709747314,
|
|
"num_tokens": 6421409.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 5.789599132537842,
|
|
"epoch": 2.9339384770549675,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004509262150651089,
|
|
"loss": 5.528,
|
|
"mean_token_accuracy": 0.15272977948188782,
|
|
"num_tokens": 6432738.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 5.708388328552246,
|
|
"epoch": 2.938981341401916,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00045067865600224833,
|
|
"loss": 5.4285,
|
|
"mean_token_accuracy": 0.15610190331935883,
|
|
"num_tokens": 6443665.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 5.6582094669342045,
|
|
"epoch": 2.9440242057488653,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004504305509260724,
|
|
"loss": 5.3923,
|
|
"mean_token_accuracy": 0.1644108533859253,
|
|
"num_tokens": 6454849.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 5.660827159881592,
|
|
"epoch": 2.9490670700958144,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004501819006076974,
|
|
"loss": 5.4615,
|
|
"mean_token_accuracy": 0.1585303395986557,
|
|
"num_tokens": 6465791.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 5.6736462116241455,
|
|
"epoch": 2.9541099344427635,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00044993270581993413,
|
|
"loss": 5.3866,
|
|
"mean_token_accuracy": 0.1626739352941513,
|
|
"num_tokens": 6475740.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 5.662177610397339,
|
|
"epoch": 2.9591527987897126,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00044968296733728587,
|
|
"loss": 5.4365,
|
|
"mean_token_accuracy": 0.15786589086055755,
|
|
"num_tokens": 6487594.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 5.635574197769165,
|
|
"epoch": 2.9641956631366617,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004494326859359456,
|
|
"loss": 5.3978,
|
|
"mean_token_accuracy": 0.15217166393995285,
|
|
"num_tokens": 6498683.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 5.655746841430664,
|
|
"epoch": 2.969238527483611,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00044918186239379363,
|
|
"loss": 5.3981,
|
|
"mean_token_accuracy": 0.15589748024940492,
|
|
"num_tokens": 6509069.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 5.726499319076538,
|
|
"epoch": 2.9742813918305595,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004489304974903953,
|
|
"loss": 5.4381,
|
|
"mean_token_accuracy": 0.1575563818216324,
|
|
"num_tokens": 6519480.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 5.683734273910522,
|
|
"epoch": 2.9793242561775086,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004486785920069985,
|
|
"loss": 5.4503,
|
|
"mean_token_accuracy": 0.16029519587755203,
|
|
"num_tokens": 6531010.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 5.70301661491394,
|
|
"epoch": 2.9843671205244577,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004484261467265313,
|
|
"loss": 5.4097,
|
|
"mean_token_accuracy": 0.1510826602578163,
|
|
"num_tokens": 6540292.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 5.763415431976318,
|
|
"epoch": 2.989409984871407,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00044817316243359956,
|
|
"loss": 5.5666,
|
|
"mean_token_accuracy": 0.14824073538184165,
|
|
"num_tokens": 6553199.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 5.655461549758911,
|
|
"epoch": 2.994452849218356,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00044791963991448427,
|
|
"loss": 5.4304,
|
|
"mean_token_accuracy": 0.15923937857151033,
|
|
"num_tokens": 6564142.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 5.684079694747925,
|
|
"epoch": 2.999495713565305,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00044766557995713915,
|
|
"loss": 5.4241,
|
|
"mean_token_accuracy": 0.1565199539065361,
|
|
"num_tokens": 6574519.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 5.680332766638862,
|
|
"epoch": 3.0040342914775593,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004474109833511885,
|
|
"loss": 5.2251,
|
|
"mean_token_accuracy": 0.18075660036669838,
|
|
"num_tokens": 6583348.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 5.6684104919433596,
|
|
"epoch": 3.0090771558245084,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004471558508879244,
|
|
"loss": 5.2168,
|
|
"mean_token_accuracy": 0.17218788713216782,
|
|
"num_tokens": 6595704.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 5.640681171417237,
|
|
"epoch": 3.0141200201714575,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004469001833603045,
|
|
"loss": 5.1253,
|
|
"mean_token_accuracy": 0.17000921666622162,
|
|
"num_tokens": 6607177.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 5.66819200515747,
|
|
"epoch": 3.0191628845184066,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004466439815629493,
|
|
"loss": 5.162,
|
|
"mean_token_accuracy": 0.17252345532178878,
|
|
"num_tokens": 6620798.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 5.659558343887329,
|
|
"epoch": 3.0242057488653553,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00044638724629214,
|
|
"loss": 5.2435,
|
|
"mean_token_accuracy": 0.17568532526493072,
|
|
"num_tokens": 6632972.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 3.0242057488653553,
|
|
"eval_entropy": 5.5543622497383875,
|
|
"eval_loss": 5.758821964263916,
|
|
"eval_mean_token_accuracy": 0.14801666542470607,
|
|
"eval_num_tokens": 6632972.0,
|
|
"eval_runtime": 2.0078,
|
|
"eval_samples_per_second": 1562.938,
|
|
"eval_steps_per_second": 195.741,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"entropy": 5.6344273567199705,
|
|
"epoch": 3.0292486132123044,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00044612997834581557,
|
|
"loss": 5.0465,
|
|
"mean_token_accuracy": 0.1772995799779892,
|
|
"num_tokens": 6643259.0,
|
|
"step": 3005
|
|
},
|
|
{
|
|
"entropy": 5.544968223571777,
|
|
"epoch": 3.0342914775592535,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00044587217852357086,
|
|
"loss": 5.2312,
|
|
"mean_token_accuracy": 0.17549270987510682,
|
|
"num_tokens": 6655875.0,
|
|
"step": 3010
|
|
},
|
|
{
|
|
"entropy": 5.663616895675659,
|
|
"epoch": 3.0393343419062027,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00044561384762665363,
|
|
"loss": 5.1627,
|
|
"mean_token_accuracy": 0.17505645006895065,
|
|
"num_tokens": 6667377.0,
|
|
"step": 3015
|
|
},
|
|
{
|
|
"entropy": 5.604240417480469,
|
|
"epoch": 3.044377206253152,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004453549864579623,
|
|
"loss": 5.1276,
|
|
"mean_token_accuracy": 0.1773838996887207,
|
|
"num_tokens": 6678293.0,
|
|
"step": 3020
|
|
},
|
|
{
|
|
"entropy": 5.6146769523620605,
|
|
"epoch": 3.049420070600101,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004450955958220433,
|
|
"loss": 5.2004,
|
|
"mean_token_accuracy": 0.17052808105945588,
|
|
"num_tokens": 6690664.0,
|
|
"step": 3025
|
|
},
|
|
{
|
|
"entropy": 5.783908128738403,
|
|
"epoch": 3.05446293494705,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00044483567652508864,
|
|
"loss": 5.1861,
|
|
"mean_token_accuracy": 0.1718222200870514,
|
|
"num_tokens": 6701465.0,
|
|
"step": 3030
|
|
},
|
|
{
|
|
"entropy": 5.579598522186279,
|
|
"epoch": 3.059505799293999,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004445752293749337,
|
|
"loss": 5.1872,
|
|
"mean_token_accuracy": 0.1748570665717125,
|
|
"num_tokens": 6713290.0,
|
|
"step": 3035
|
|
},
|
|
{
|
|
"entropy": 5.6107306480407715,
|
|
"epoch": 3.0645486636409482,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004443142551810541,
|
|
"loss": 5.2065,
|
|
"mean_token_accuracy": 0.1741941511631012,
|
|
"num_tokens": 6723496.0,
|
|
"step": 3040
|
|
},
|
|
{
|
|
"entropy": 5.717791318893433,
|
|
"epoch": 3.069591527987897,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00044405275475456365,
|
|
"loss": 5.1889,
|
|
"mean_token_accuracy": 0.1693324014544487,
|
|
"num_tokens": 6734030.0,
|
|
"step": 3045
|
|
},
|
|
{
|
|
"entropy": 5.6288004398345945,
|
|
"epoch": 3.074634392334846,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004437907289082116,
|
|
"loss": 5.2149,
|
|
"mean_token_accuracy": 0.1699022576212883,
|
|
"num_tokens": 6745966.0,
|
|
"step": 3050
|
|
},
|
|
{
|
|
"entropy": 5.667066621780395,
|
|
"epoch": 3.079677256681795,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00044352817845638053,
|
|
"loss": 5.1171,
|
|
"mean_token_accuracy": 0.17061944007873536,
|
|
"num_tokens": 6756982.0,
|
|
"step": 3055
|
|
},
|
|
{
|
|
"entropy": 5.598727416992188,
|
|
"epoch": 3.0847201210287443,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004432651042150833,
|
|
"loss": 5.1428,
|
|
"mean_token_accuracy": 0.16678404957056045,
|
|
"num_tokens": 6768950.0,
|
|
"step": 3060
|
|
},
|
|
{
|
|
"entropy": 5.517779684066772,
|
|
"epoch": 3.0897629853756934,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00044300150700196054,
|
|
"loss": 5.1185,
|
|
"mean_token_accuracy": 0.16840976029634475,
|
|
"num_tokens": 6779748.0,
|
|
"step": 3065
|
|
},
|
|
{
|
|
"entropy": 5.645850419998169,
|
|
"epoch": 3.0948058497226425,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004427373876362786,
|
|
"loss": 5.2444,
|
|
"mean_token_accuracy": 0.16981570422649384,
|
|
"num_tokens": 6791456.0,
|
|
"step": 3070
|
|
},
|
|
{
|
|
"entropy": 5.568591070175171,
|
|
"epoch": 3.0998487140695916,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004424727469389265,
|
|
"loss": 5.1751,
|
|
"mean_token_accuracy": 0.17235083431005477,
|
|
"num_tokens": 6802813.0,
|
|
"step": 3075
|
|
},
|
|
{
|
|
"entropy": 5.676083755493164,
|
|
"epoch": 3.1048915784165407,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004422075857324137,
|
|
"loss": 5.226,
|
|
"mean_token_accuracy": 0.1703367382287979,
|
|
"num_tokens": 6813310.0,
|
|
"step": 3080
|
|
},
|
|
{
|
|
"entropy": 5.6751659393310545,
|
|
"epoch": 3.10993444276349,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004419419048408673,
|
|
"loss": 5.2554,
|
|
"mean_token_accuracy": 0.16963330656290054,
|
|
"num_tokens": 6825593.0,
|
|
"step": 3085
|
|
},
|
|
{
|
|
"entropy": 5.584595108032227,
|
|
"epoch": 3.1149773071104385,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004416757050900297,
|
|
"loss": 5.1262,
|
|
"mean_token_accuracy": 0.17636028975248336,
|
|
"num_tokens": 6836279.0,
|
|
"step": 3090
|
|
},
|
|
{
|
|
"entropy": 5.578235721588134,
|
|
"epoch": 3.1200201714573876,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004414089873072557,
|
|
"loss": 5.1211,
|
|
"mean_token_accuracy": 0.17816683799028396,
|
|
"num_tokens": 6846087.0,
|
|
"step": 3095
|
|
},
|
|
{
|
|
"entropy": 5.595646095275879,
|
|
"epoch": 3.1250630358043368,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004411417523215105,
|
|
"loss": 5.0995,
|
|
"mean_token_accuracy": 0.17462697625160217,
|
|
"num_tokens": 6857290.0,
|
|
"step": 3100
|
|
},
|
|
{
|
|
"entropy": 5.5459352970123295,
|
|
"epoch": 3.130105900151286,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004408740009633665,
|
|
"loss": 5.201,
|
|
"mean_token_accuracy": 0.17095047533512114,
|
|
"num_tokens": 6870060.0,
|
|
"step": 3105
|
|
},
|
|
{
|
|
"entropy": 5.644553804397583,
|
|
"epoch": 3.135148764498235,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00044060573406500117,
|
|
"loss": 5.122,
|
|
"mean_token_accuracy": 0.16597261726856233,
|
|
"num_tokens": 6880813.0,
|
|
"step": 3110
|
|
},
|
|
{
|
|
"entropy": 5.590497159957886,
|
|
"epoch": 3.140191628845184,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004403369524601941,
|
|
"loss": 5.111,
|
|
"mean_token_accuracy": 0.1813727870583534,
|
|
"num_tokens": 6891955.0,
|
|
"step": 3115
|
|
},
|
|
{
|
|
"entropy": 5.541271352767945,
|
|
"epoch": 3.145234493192133,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004400676569843249,
|
|
"loss": 5.1085,
|
|
"mean_token_accuracy": 0.17570345997810363,
|
|
"num_tokens": 6902873.0,
|
|
"step": 3120
|
|
},
|
|
{
|
|
"entropy": 5.626648044586181,
|
|
"epoch": 3.1502773575390823,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00043979784847436994,
|
|
"loss": 5.136,
|
|
"mean_token_accuracy": 0.1727659896016121,
|
|
"num_tokens": 6916104.0,
|
|
"step": 3125
|
|
},
|
|
{
|
|
"entropy": 5.626005983352661,
|
|
"epoch": 3.1553202218860315,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004395275277689004,
|
|
"loss": 5.1295,
|
|
"mean_token_accuracy": 0.1806250259280205,
|
|
"num_tokens": 6926117.0,
|
|
"step": 3130
|
|
},
|
|
{
|
|
"entropy": 5.573675155639648,
|
|
"epoch": 3.16036308623298,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004392566957080793,
|
|
"loss": 5.1353,
|
|
"mean_token_accuracy": 0.17382414340972902,
|
|
"num_tokens": 6936257.0,
|
|
"step": 3135
|
|
},
|
|
{
|
|
"entropy": 5.620882558822632,
|
|
"epoch": 3.1654059505799292,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00043898535313365884,
|
|
"loss": 5.2756,
|
|
"mean_token_accuracy": 0.17034963816404342,
|
|
"num_tokens": 6948161.0,
|
|
"step": 3140
|
|
},
|
|
{
|
|
"entropy": 5.593581104278565,
|
|
"epoch": 3.1704488149268784,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000438713500888978,
|
|
"loss": 5.151,
|
|
"mean_token_accuracy": 0.17283370345830917,
|
|
"num_tokens": 6959014.0,
|
|
"step": 3145
|
|
},
|
|
{
|
|
"entropy": 5.55266637802124,
|
|
"epoch": 3.1754916792738275,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004384411398189599,
|
|
"loss": 5.1184,
|
|
"mean_token_accuracy": 0.17197933048009872,
|
|
"num_tokens": 6969180.0,
|
|
"step": 3150
|
|
},
|
|
{
|
|
"entropy": 5.609263277053833,
|
|
"epoch": 3.1805345436207766,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000438168270770109,
|
|
"loss": 5.2663,
|
|
"mean_token_accuracy": 0.16960919350385667,
|
|
"num_tokens": 6980702.0,
|
|
"step": 3155
|
|
},
|
|
{
|
|
"entropy": 5.674694633483886,
|
|
"epoch": 3.1855774079677257,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004378948945905085,
|
|
"loss": 5.2344,
|
|
"mean_token_accuracy": 0.1671406552195549,
|
|
"num_tokens": 6992904.0,
|
|
"step": 3160
|
|
},
|
|
{
|
|
"entropy": 5.574981307983398,
|
|
"epoch": 3.190620272314675,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00043762101212981805,
|
|
"loss": 5.1502,
|
|
"mean_token_accuracy": 0.1726602241396904,
|
|
"num_tokens": 7004355.0,
|
|
"step": 3165
|
|
},
|
|
{
|
|
"entropy": 5.613319635391235,
|
|
"epoch": 3.195663136661624,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00043734662423927054,
|
|
"loss": 5.1484,
|
|
"mean_token_accuracy": 0.1709890753030777,
|
|
"num_tokens": 7014941.0,
|
|
"step": 3170
|
|
},
|
|
{
|
|
"entropy": 5.566009616851806,
|
|
"epoch": 3.200706001008573,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004370717317716699,
|
|
"loss": 5.1889,
|
|
"mean_token_accuracy": 0.16855269372463227,
|
|
"num_tokens": 7026208.0,
|
|
"step": 3175
|
|
},
|
|
{
|
|
"entropy": 5.562632369995117,
|
|
"epoch": 3.205748865355522,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004367963355813882,
|
|
"loss": 5.2607,
|
|
"mean_token_accuracy": 0.16157007962465286,
|
|
"num_tokens": 7038065.0,
|
|
"step": 3180
|
|
},
|
|
{
|
|
"entropy": 5.713967657089233,
|
|
"epoch": 3.210791729702471,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00043652043652436323,
|
|
"loss": 5.2221,
|
|
"mean_token_accuracy": 0.17069490253925323,
|
|
"num_tokens": 7049887.0,
|
|
"step": 3185
|
|
},
|
|
{
|
|
"entropy": 5.575469160079956,
|
|
"epoch": 3.21583459404942,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004362440354580956,
|
|
"loss": 5.1246,
|
|
"mean_token_accuracy": 0.17573699802160264,
|
|
"num_tokens": 7060056.0,
|
|
"step": 3190
|
|
},
|
|
{
|
|
"entropy": 5.538661527633667,
|
|
"epoch": 3.220877458396369,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004359671332416462,
|
|
"loss": 5.1584,
|
|
"mean_token_accuracy": 0.16958663016557693,
|
|
"num_tokens": 7071979.0,
|
|
"step": 3195
|
|
},
|
|
{
|
|
"entropy": 5.592955732345581,
|
|
"epoch": 3.225920322743318,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004356897307356335,
|
|
"loss": 5.1985,
|
|
"mean_token_accuracy": 0.17620886415243148,
|
|
"num_tokens": 7082534.0,
|
|
"step": 3200
|
|
},
|
|
{
|
|
"entropy": 5.550133085250854,
|
|
"epoch": 3.2309631870902673,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004354118288022309,
|
|
"loss": 5.2062,
|
|
"mean_token_accuracy": 0.1737912192940712,
|
|
"num_tokens": 7094737.0,
|
|
"step": 3205
|
|
},
|
|
{
|
|
"entropy": 5.597499799728394,
|
|
"epoch": 3.2360060514372164,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00043513342830516405,
|
|
"loss": 5.2059,
|
|
"mean_token_accuracy": 0.17379997968673705,
|
|
"num_tokens": 7105483.0,
|
|
"step": 3210
|
|
},
|
|
{
|
|
"entropy": 5.579964685440063,
|
|
"epoch": 3.2410489157841655,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000434854530109708,
|
|
"loss": 5.1136,
|
|
"mean_token_accuracy": 0.1714743360877037,
|
|
"num_tokens": 7115110.0,
|
|
"step": 3215
|
|
},
|
|
{
|
|
"entropy": 5.605909776687622,
|
|
"epoch": 3.2460917801311147,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00043457513508268487,
|
|
"loss": 5.1738,
|
|
"mean_token_accuracy": 0.17052750736474992,
|
|
"num_tokens": 7125216.0,
|
|
"step": 3220
|
|
},
|
|
{
|
|
"entropy": 5.617115163803101,
|
|
"epoch": 3.2511346444780633,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00043429524409246083,
|
|
"loss": 5.216,
|
|
"mean_token_accuracy": 0.1663753166794777,
|
|
"num_tokens": 7136457.0,
|
|
"step": 3225
|
|
},
|
|
{
|
|
"entropy": 5.546289443969727,
|
|
"epoch": 3.2561775088250124,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004340148580089436,
|
|
"loss": 5.1968,
|
|
"mean_token_accuracy": 0.17131324708461762,
|
|
"num_tokens": 7147315.0,
|
|
"step": 3230
|
|
},
|
|
{
|
|
"entropy": 5.66264066696167,
|
|
"epoch": 3.2612203731719616,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004337339777035794,
|
|
"loss": 5.1877,
|
|
"mean_token_accuracy": 0.16779913306236266,
|
|
"num_tokens": 7158436.0,
|
|
"step": 3235
|
|
},
|
|
{
|
|
"entropy": 5.637083053588867,
|
|
"epoch": 3.2662632375189107,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000433452604049351,
|
|
"loss": 5.2232,
|
|
"mean_token_accuracy": 0.16788240671157836,
|
|
"num_tokens": 7168705.0,
|
|
"step": 3240
|
|
},
|
|
{
|
|
"entropy": 5.558506965637207,
|
|
"epoch": 3.27130610186586,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00043317073792077394,
|
|
"loss": 5.1416,
|
|
"mean_token_accuracy": 0.17377690076828003,
|
|
"num_tokens": 7179200.0,
|
|
"step": 3245
|
|
},
|
|
{
|
|
"entropy": 5.623791217803955,
|
|
"epoch": 3.276348966212809,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004328883801938947,
|
|
"loss": 5.2167,
|
|
"mean_token_accuracy": 0.16812532246112824,
|
|
"num_tokens": 7189378.0,
|
|
"step": 3250
|
|
},
|
|
{
|
|
"entropy": 5.640508699417114,
|
|
"epoch": 3.281391830559758,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00043260553174628776,
|
|
"loss": 5.1124,
|
|
"mean_token_accuracy": 0.1789386674761772,
|
|
"num_tokens": 7198726.0,
|
|
"step": 3255
|
|
},
|
|
{
|
|
"entropy": 5.575011396408081,
|
|
"epoch": 3.286434694906707,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004323221934570524,
|
|
"loss": 5.1873,
|
|
"mean_token_accuracy": 0.16848568618297577,
|
|
"num_tokens": 7210547.0,
|
|
"step": 3260
|
|
},
|
|
{
|
|
"entropy": 5.504167604446411,
|
|
"epoch": 3.2914775592536563,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00043203836620681074,
|
|
"loss": 5.1018,
|
|
"mean_token_accuracy": 0.17515749335289002,
|
|
"num_tokens": 7221871.0,
|
|
"step": 3265
|
|
},
|
|
{
|
|
"entropy": 5.568135976791382,
|
|
"epoch": 3.2965204236006054,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00043175405087770435,
|
|
"loss": 5.1926,
|
|
"mean_token_accuracy": 0.17292053997516632,
|
|
"num_tokens": 7233928.0,
|
|
"step": 3270
|
|
},
|
|
{
|
|
"entropy": 5.581064987182617,
|
|
"epoch": 3.301563287947554,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00043146924835339186,
|
|
"loss": 5.2003,
|
|
"mean_token_accuracy": 0.16545283794403076,
|
|
"num_tokens": 7245263.0,
|
|
"step": 3275
|
|
},
|
|
{
|
|
"entropy": 5.704693555831909,
|
|
"epoch": 3.306606152294503,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00043118395951904605,
|
|
"loss": 5.2719,
|
|
"mean_token_accuracy": 0.1727494478225708,
|
|
"num_tokens": 7257941.0,
|
|
"step": 3280
|
|
},
|
|
{
|
|
"entropy": 5.6160472393035885,
|
|
"epoch": 3.3116490166414523,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004308981852613512,
|
|
"loss": 5.1661,
|
|
"mean_token_accuracy": 0.16646221727132798,
|
|
"num_tokens": 7267854.0,
|
|
"step": 3285
|
|
},
|
|
{
|
|
"entropy": 5.593245792388916,
|
|
"epoch": 3.3166918809884014,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004306119264685003,
|
|
"loss": 5.2396,
|
|
"mean_token_accuracy": 0.17379083186388017,
|
|
"num_tokens": 7279898.0,
|
|
"step": 3290
|
|
},
|
|
{
|
|
"entropy": 5.6669676303863525,
|
|
"epoch": 3.3217347453353505,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004303251840301923,
|
|
"loss": 5.231,
|
|
"mean_token_accuracy": 0.17253653556108475,
|
|
"num_tokens": 7291459.0,
|
|
"step": 3295
|
|
},
|
|
{
|
|
"entropy": 5.629107093811035,
|
|
"epoch": 3.3267776096822996,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00043003795883762933,
|
|
"loss": 5.2623,
|
|
"mean_token_accuracy": 0.17112715691328048,
|
|
"num_tokens": 7302709.0,
|
|
"step": 3300
|
|
},
|
|
{
|
|
"entropy": 5.535666561126709,
|
|
"epoch": 3.3318204740292487,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00042975025178351386,
|
|
"loss": 5.1606,
|
|
"mean_token_accuracy": 0.17082529515028,
|
|
"num_tokens": 7312541.0,
|
|
"step": 3305
|
|
},
|
|
{
|
|
"entropy": 5.560184144973755,
|
|
"epoch": 3.336863338376198,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000429462063762046,
|
|
"loss": 5.1982,
|
|
"mean_token_accuracy": 0.171163409948349,
|
|
"num_tokens": 7323444.0,
|
|
"step": 3310
|
|
},
|
|
{
|
|
"entropy": 5.597321176528931,
|
|
"epoch": 3.3419062027231465,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00042917339566892095,
|
|
"loss": 5.1318,
|
|
"mean_token_accuracy": 0.1789054661989212,
|
|
"num_tokens": 7334751.0,
|
|
"step": 3315
|
|
},
|
|
{
|
|
"entropy": 5.572882175445557,
|
|
"epoch": 3.3469490670700957,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004288842484013257,
|
|
"loss": 5.2281,
|
|
"mean_token_accuracy": 0.16511097252368928,
|
|
"num_tokens": 7347434.0,
|
|
"step": 3320
|
|
},
|
|
{
|
|
"entropy": 5.583644056320191,
|
|
"epoch": 3.3519919314170448,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004285946228579365,
|
|
"loss": 5.223,
|
|
"mean_token_accuracy": 0.17328801304101943,
|
|
"num_tokens": 7358240.0,
|
|
"step": 3325
|
|
},
|
|
{
|
|
"entropy": 5.58079833984375,
|
|
"epoch": 3.357034795763994,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004283045199389164,
|
|
"loss": 5.1985,
|
|
"mean_token_accuracy": 0.1712915286421776,
|
|
"num_tokens": 7369542.0,
|
|
"step": 3330
|
|
},
|
|
{
|
|
"entropy": 5.577418613433838,
|
|
"epoch": 3.362077660110943,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00042801394054591197,
|
|
"loss": 5.2423,
|
|
"mean_token_accuracy": 0.15943718701601028,
|
|
"num_tokens": 7381885.0,
|
|
"step": 3335
|
|
},
|
|
{
|
|
"entropy": 5.585267782211304,
|
|
"epoch": 3.367120524457892,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00042772288558205057,
|
|
"loss": 5.1436,
|
|
"mean_token_accuracy": 0.171712164580822,
|
|
"num_tokens": 7392042.0,
|
|
"step": 3340
|
|
},
|
|
{
|
|
"entropy": 5.626815605163574,
|
|
"epoch": 3.3721633888048412,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004274313559519377,
|
|
"loss": 5.2231,
|
|
"mean_token_accuracy": 0.1704651415348053,
|
|
"num_tokens": 7403388.0,
|
|
"step": 3345
|
|
},
|
|
{
|
|
"entropy": 5.563056755065918,
|
|
"epoch": 3.3772062531517903,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004271393525616544,
|
|
"loss": 5.1338,
|
|
"mean_token_accuracy": 0.17188099771738052,
|
|
"num_tokens": 7413501.0,
|
|
"step": 3350
|
|
},
|
|
{
|
|
"entropy": 5.508030366897583,
|
|
"epoch": 3.3822491174987395,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00042684687631875377,
|
|
"loss": 5.0388,
|
|
"mean_token_accuracy": 0.17996248453855515,
|
|
"num_tokens": 7424529.0,
|
|
"step": 3355
|
|
},
|
|
{
|
|
"entropy": 5.469664812088013,
|
|
"epoch": 3.3872919818456886,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00042655392813225864,
|
|
"loss": 5.2192,
|
|
"mean_token_accuracy": 0.17520563304424286,
|
|
"num_tokens": 7434551.0,
|
|
"step": 3360
|
|
},
|
|
{
|
|
"entropy": 5.644444131851197,
|
|
"epoch": 3.3923348461926373,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00042626050891265884,
|
|
"loss": 5.2033,
|
|
"mean_token_accuracy": 0.16889225244522094,
|
|
"num_tokens": 7443898.0,
|
|
"step": 3365
|
|
},
|
|
{
|
|
"entropy": 5.594678688049316,
|
|
"epoch": 3.3973777105395864,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004259666195719081,
|
|
"loss": 5.1951,
|
|
"mean_token_accuracy": 0.17015911489725113,
|
|
"num_tokens": 7454399.0,
|
|
"step": 3370
|
|
},
|
|
{
|
|
"entropy": 5.632461023330689,
|
|
"epoch": 3.4024205748865355,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00042567226102342134,
|
|
"loss": 5.2232,
|
|
"mean_token_accuracy": 0.17474526762962342,
|
|
"num_tokens": 7465564.0,
|
|
"step": 3375
|
|
},
|
|
{
|
|
"entropy": 5.511708402633667,
|
|
"epoch": 3.4074634392334846,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004253774341820715,
|
|
"loss": 5.168,
|
|
"mean_token_accuracy": 0.17088545560836793,
|
|
"num_tokens": 7476619.0,
|
|
"step": 3380
|
|
},
|
|
{
|
|
"entropy": 5.570131063461304,
|
|
"epoch": 3.4125063035804337,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004250821399641874,
|
|
"loss": 5.1567,
|
|
"mean_token_accuracy": 0.17556075602769852,
|
|
"num_tokens": 7487175.0,
|
|
"step": 3385
|
|
},
|
|
{
|
|
"entropy": 5.55647988319397,
|
|
"epoch": 3.417549167927383,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004247863792875502,
|
|
"loss": 5.1332,
|
|
"mean_token_accuracy": 0.17436463683843612,
|
|
"num_tokens": 7498152.0,
|
|
"step": 3390
|
|
},
|
|
{
|
|
"entropy": 5.6045814037323,
|
|
"epoch": 3.422592032274332,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00042449015307139076,
|
|
"loss": 5.1514,
|
|
"mean_token_accuracy": 0.17974659651517869,
|
|
"num_tokens": 7507914.0,
|
|
"step": 3395
|
|
},
|
|
{
|
|
"entropy": 5.553989219665527,
|
|
"epoch": 3.427634896621281,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00042419346223638725,
|
|
"loss": 5.2262,
|
|
"mean_token_accuracy": 0.16634046733379365,
|
|
"num_tokens": 7518941.0,
|
|
"step": 3400
|
|
},
|
|
{
|
|
"entropy": 5.527310371398926,
|
|
"epoch": 3.4326777609682297,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004238963077046616,
|
|
"loss": 5.0733,
|
|
"mean_token_accuracy": 0.1796322852373123,
|
|
"num_tokens": 7529315.0,
|
|
"step": 3405
|
|
},
|
|
{
|
|
"entropy": 5.5679820537567135,
|
|
"epoch": 3.437720625315179,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004235986903997767,
|
|
"loss": 5.1828,
|
|
"mean_token_accuracy": 0.17507729828357696,
|
|
"num_tokens": 7539888.0,
|
|
"step": 3410
|
|
},
|
|
{
|
|
"entropy": 5.595615339279175,
|
|
"epoch": 3.442763489662128,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00042330061124673437,
|
|
"loss": 5.1558,
|
|
"mean_token_accuracy": 0.1686934530735016,
|
|
"num_tokens": 7550429.0,
|
|
"step": 3415
|
|
},
|
|
{
|
|
"entropy": 5.540881061553955,
|
|
"epoch": 3.447806354009077,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00042300207117197116,
|
|
"loss": 5.1417,
|
|
"mean_token_accuracy": 0.1704073205590248,
|
|
"num_tokens": 7561051.0,
|
|
"step": 3420
|
|
},
|
|
{
|
|
"entropy": 5.5717075824737545,
|
|
"epoch": 3.452849218356026,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00042270307110335657,
|
|
"loss": 5.1663,
|
|
"mean_token_accuracy": 0.17200603187084199,
|
|
"num_tokens": 7570847.0,
|
|
"step": 3425
|
|
},
|
|
{
|
|
"entropy": 5.488401317596436,
|
|
"epoch": 3.4578920827029753,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004224036119701899,
|
|
"loss": 5.1541,
|
|
"mean_token_accuracy": 0.1766899436712265,
|
|
"num_tokens": 7581515.0,
|
|
"step": 3430
|
|
},
|
|
{
|
|
"entropy": 5.5144696712493895,
|
|
"epoch": 3.4629349470499244,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00042210369470319694,
|
|
"loss": 5.1638,
|
|
"mean_token_accuracy": 0.1747896119952202,
|
|
"num_tokens": 7592306.0,
|
|
"step": 3435
|
|
},
|
|
{
|
|
"entropy": 5.616534566879272,
|
|
"epoch": 3.4679778113968736,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004218033202345274,
|
|
"loss": 5.2739,
|
|
"mean_token_accuracy": 0.16605759114027024,
|
|
"num_tokens": 7603264.0,
|
|
"step": 3440
|
|
},
|
|
{
|
|
"entropy": 5.64601697921753,
|
|
"epoch": 3.4730206757438227,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004215024894977522,
|
|
"loss": 5.2276,
|
|
"mean_token_accuracy": 0.16955839842557907,
|
|
"num_tokens": 7615603.0,
|
|
"step": 3445
|
|
},
|
|
{
|
|
"entropy": 5.5407247066497805,
|
|
"epoch": 3.478063540090772,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004212012034278601,
|
|
"loss": 5.1311,
|
|
"mean_token_accuracy": 0.1764802634716034,
|
|
"num_tokens": 7626072.0,
|
|
"step": 3450
|
|
},
|
|
{
|
|
"entropy": 5.562434482574463,
|
|
"epoch": 3.4831064044377205,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00042089946296125513,
|
|
"loss": 5.1832,
|
|
"mean_token_accuracy": 0.16629791110754014,
|
|
"num_tokens": 7636321.0,
|
|
"step": 3455
|
|
},
|
|
{
|
|
"entropy": 5.605208253860473,
|
|
"epoch": 3.4881492687846696,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00042059726903575373,
|
|
"loss": 5.1707,
|
|
"mean_token_accuracy": 0.173575958609581,
|
|
"num_tokens": 7647709.0,
|
|
"step": 3460
|
|
},
|
|
{
|
|
"entropy": 5.559977340698242,
|
|
"epoch": 3.4931921331316187,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004202946225905815,
|
|
"loss": 5.1678,
|
|
"mean_token_accuracy": 0.17033482789993287,
|
|
"num_tokens": 7660195.0,
|
|
"step": 3465
|
|
},
|
|
{
|
|
"entropy": 5.549334526062012,
|
|
"epoch": 3.498234997478568,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004199915245663706,
|
|
"loss": 5.1158,
|
|
"mean_token_accuracy": 0.18043556958436965,
|
|
"num_tokens": 7670274.0,
|
|
"step": 3470
|
|
},
|
|
{
|
|
"entropy": 5.60921688079834,
|
|
"epoch": 3.503277861825517,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00041968797590515673,
|
|
"loss": 5.1614,
|
|
"mean_token_accuracy": 0.17460013180971146,
|
|
"num_tokens": 7680327.0,
|
|
"step": 3475
|
|
},
|
|
{
|
|
"entropy": 5.5384783267974855,
|
|
"epoch": 3.508320726172466,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004193839775503761,
|
|
"loss": 5.1986,
|
|
"mean_token_accuracy": 0.17059999704360962,
|
|
"num_tokens": 7691250.0,
|
|
"step": 3480
|
|
},
|
|
{
|
|
"entropy": 5.547350883483887,
|
|
"epoch": 3.513363590519415,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00041907953044686253,
|
|
"loss": 5.2145,
|
|
"mean_token_accuracy": 0.1730611115694046,
|
|
"num_tokens": 7703726.0,
|
|
"step": 3485
|
|
},
|
|
{
|
|
"entropy": 5.55848913192749,
|
|
"epoch": 3.518406454866364,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004187746355408448,
|
|
"loss": 5.2567,
|
|
"mean_token_accuracy": 0.16977688521146775,
|
|
"num_tokens": 7715179.0,
|
|
"step": 3490
|
|
},
|
|
{
|
|
"entropy": 5.6596705436706545,
|
|
"epoch": 3.523449319213313,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004184692937799431,
|
|
"loss": 5.2258,
|
|
"mean_token_accuracy": 0.17172631323337556,
|
|
"num_tokens": 7726586.0,
|
|
"step": 3495
|
|
},
|
|
{
|
|
"entropy": 5.560780143737793,
|
|
"epoch": 3.528492183560262,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00041816350611316693,
|
|
"loss": 5.1879,
|
|
"mean_token_accuracy": 0.17001775354146959,
|
|
"num_tokens": 7737350.0,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"epoch": 3.528492183560262,
|
|
"eval_entropy": 5.369396981392198,
|
|
"eval_loss": 5.73445987701416,
|
|
"eval_mean_token_accuracy": 0.14839879131423306,
|
|
"eval_num_tokens": 7737350.0,
|
|
"eval_runtime": 2.0018,
|
|
"eval_samples_per_second": 1567.606,
|
|
"eval_steps_per_second": 196.325,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"entropy": 5.549285745620727,
|
|
"epoch": 3.533535047907211,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00041785727349091117,
|
|
"loss": 5.189,
|
|
"mean_token_accuracy": 0.1749983012676239,
|
|
"num_tokens": 7749586.0,
|
|
"step": 3505
|
|
},
|
|
{
|
|
"entropy": 5.5008780002594,
|
|
"epoch": 3.5385779122541603,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000417550596864954,
|
|
"loss": 5.1844,
|
|
"mean_token_accuracy": 0.17288953214883804,
|
|
"num_tokens": 7761889.0,
|
|
"step": 3510
|
|
},
|
|
{
|
|
"entropy": 5.598424100875855,
|
|
"epoch": 3.5436207766011094,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00041724347718845344,
|
|
"loss": 5.1862,
|
|
"mean_token_accuracy": 0.1710773915052414,
|
|
"num_tokens": 7772088.0,
|
|
"step": 3515
|
|
},
|
|
{
|
|
"entropy": 5.575075244903564,
|
|
"epoch": 3.5486636409480585,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00041693591541594455,
|
|
"loss": 5.1688,
|
|
"mean_token_accuracy": 0.17212855219841003,
|
|
"num_tokens": 7783087.0,
|
|
"step": 3520
|
|
},
|
|
{
|
|
"entropy": 5.484617471694946,
|
|
"epoch": 3.5537065052950076,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004166279125033363,
|
|
"loss": 5.1371,
|
|
"mean_token_accuracy": 0.16855607330799102,
|
|
"num_tokens": 7792919.0,
|
|
"step": 3525
|
|
},
|
|
{
|
|
"entropy": 5.529237985610962,
|
|
"epoch": 3.5587493696419568,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00041631946940790893,
|
|
"loss": 5.1642,
|
|
"mean_token_accuracy": 0.17052251547574998,
|
|
"num_tokens": 7803905.0,
|
|
"step": 3530
|
|
},
|
|
{
|
|
"entropy": 5.549853134155273,
|
|
"epoch": 3.563792233988906,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00041601058708831077,
|
|
"loss": 5.1935,
|
|
"mean_token_accuracy": 0.1691523537039757,
|
|
"num_tokens": 7815651.0,
|
|
"step": 3535
|
|
},
|
|
{
|
|
"entropy": 5.556256866455078,
|
|
"epoch": 3.568835098335855,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000415701266504555,
|
|
"loss": 5.1763,
|
|
"mean_token_accuracy": 0.16975914239883422,
|
|
"num_tokens": 7827247.0,
|
|
"step": 3540
|
|
},
|
|
{
|
|
"entropy": 5.598951768875122,
|
|
"epoch": 3.5738779626828037,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004153915086180173,
|
|
"loss": 5.1876,
|
|
"mean_token_accuracy": 0.16911599040031433,
|
|
"num_tokens": 7838461.0,
|
|
"step": 3545
|
|
},
|
|
{
|
|
"entropy": 5.491204690933228,
|
|
"epoch": 3.578920827029753,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004150813143914323,
|
|
"loss": 5.1247,
|
|
"mean_token_accuracy": 0.17307860255241395,
|
|
"num_tokens": 7849824.0,
|
|
"step": 3550
|
|
},
|
|
{
|
|
"entropy": 5.503362417221069,
|
|
"epoch": 3.583963691376702,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004147706847888908,
|
|
"loss": 5.2137,
|
|
"mean_token_accuracy": 0.17098104059696198,
|
|
"num_tokens": 7862167.0,
|
|
"step": 3555
|
|
},
|
|
{
|
|
"entropy": 5.551947784423828,
|
|
"epoch": 3.589006555723651,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004144596207758367,
|
|
"loss": 5.1897,
|
|
"mean_token_accuracy": 0.1705412745475769,
|
|
"num_tokens": 7872320.0,
|
|
"step": 3560
|
|
},
|
|
{
|
|
"entropy": 5.577542543411255,
|
|
"epoch": 3.5940494200706,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004141481233190642,
|
|
"loss": 5.1474,
|
|
"mean_token_accuracy": 0.18331588208675384,
|
|
"num_tokens": 7883150.0,
|
|
"step": 3565
|
|
},
|
|
{
|
|
"entropy": 5.499377870559693,
|
|
"epoch": 3.5990922844175492,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004138361933867145,
|
|
"loss": 5.224,
|
|
"mean_token_accuracy": 0.16496764421463012,
|
|
"num_tokens": 7895371.0,
|
|
"step": 3570
|
|
},
|
|
{
|
|
"entropy": 5.4942014694213865,
|
|
"epoch": 3.6041351487644984,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004135238319482731,
|
|
"loss": 5.0919,
|
|
"mean_token_accuracy": 0.18165515959262848,
|
|
"num_tokens": 7905669.0,
|
|
"step": 3575
|
|
},
|
|
{
|
|
"entropy": 5.536538124084473,
|
|
"epoch": 3.609178013111447,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004132110399745666,
|
|
"loss": 5.115,
|
|
"mean_token_accuracy": 0.17324032187461852,
|
|
"num_tokens": 7915714.0,
|
|
"step": 3580
|
|
},
|
|
{
|
|
"entropy": 5.515967845916748,
|
|
"epoch": 3.614220877458396,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004128978184377596,
|
|
"loss": 5.1844,
|
|
"mean_token_accuracy": 0.16992210149765014,
|
|
"num_tokens": 7927110.0,
|
|
"step": 3585
|
|
},
|
|
{
|
|
"entropy": 5.577891540527344,
|
|
"epoch": 3.6192637418053453,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004125841683113518,
|
|
"loss": 5.2271,
|
|
"mean_token_accuracy": 0.16974231004714965,
|
|
"num_tokens": 7939170.0,
|
|
"step": 3590
|
|
},
|
|
{
|
|
"entropy": 5.535889959335327,
|
|
"epoch": 3.6243066061522944,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.00041227009057017523,
|
|
"loss": 5.2515,
|
|
"mean_token_accuracy": 0.16570890843868255,
|
|
"num_tokens": 7952615.0,
|
|
"step": 3595
|
|
},
|
|
{
|
|
"entropy": 5.586013078689575,
|
|
"epoch": 3.6293494704992435,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004119555861903906,
|
|
"loss": 5.206,
|
|
"mean_token_accuracy": 0.1657826915383339,
|
|
"num_tokens": 7963946.0,
|
|
"step": 3600
|
|
},
|
|
{
|
|
"entropy": 5.53160924911499,
|
|
"epoch": 3.6343923348461926,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004116406561494847,
|
|
"loss": 5.1753,
|
|
"mean_token_accuracy": 0.1703537657856941,
|
|
"num_tokens": 7975270.0,
|
|
"step": 3605
|
|
},
|
|
{
|
|
"entropy": 5.450020408630371,
|
|
"epoch": 3.6394351991931417,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00041132530142626756,
|
|
"loss": 5.1179,
|
|
"mean_token_accuracy": 0.1767469197511673,
|
|
"num_tokens": 7986008.0,
|
|
"step": 3610
|
|
},
|
|
{
|
|
"entropy": 5.5850231647491455,
|
|
"epoch": 3.644478063540091,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00041100952300086884,
|
|
"loss": 5.2954,
|
|
"mean_token_accuracy": 0.16496750712394714,
|
|
"num_tokens": 7997416.0,
|
|
"step": 3615
|
|
},
|
|
{
|
|
"entropy": 5.524045991897583,
|
|
"epoch": 3.64952092788704,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00041069332185473514,
|
|
"loss": 5.0927,
|
|
"mean_token_accuracy": 0.18004596680402757,
|
|
"num_tokens": 8009120.0,
|
|
"step": 3620
|
|
},
|
|
{
|
|
"entropy": 5.517886161804199,
|
|
"epoch": 3.654563792233989,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000410376698970627,
|
|
"loss": 5.2037,
|
|
"mean_token_accuracy": 0.16733744442462922,
|
|
"num_tokens": 8020325.0,
|
|
"step": 3625
|
|
},
|
|
{
|
|
"entropy": 5.51771764755249,
|
|
"epoch": 3.659606656580938,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00041005965533261554,
|
|
"loss": 5.1455,
|
|
"mean_token_accuracy": 0.17291317731142045,
|
|
"num_tokens": 8031456.0,
|
|
"step": 3630
|
|
},
|
|
{
|
|
"entropy": 5.566720819473266,
|
|
"epoch": 3.6646495209278873,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00040974219192607973,
|
|
"loss": 5.1863,
|
|
"mean_token_accuracy": 0.17562130242586135,
|
|
"num_tokens": 8042823.0,
|
|
"step": 3635
|
|
},
|
|
{
|
|
"entropy": 5.576432228088379,
|
|
"epoch": 3.669692385274836,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00040942430973770323,
|
|
"loss": 5.2634,
|
|
"mean_token_accuracy": 0.1694312125444412,
|
|
"num_tokens": 8054516.0,
|
|
"step": 3640
|
|
},
|
|
{
|
|
"entropy": 5.546356248855591,
|
|
"epoch": 3.674735249621785,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004091060097554712,
|
|
"loss": 5.1489,
|
|
"mean_token_accuracy": 0.17199642658233644,
|
|
"num_tokens": 8064783.0,
|
|
"step": 3645
|
|
},
|
|
{
|
|
"entropy": 5.504854154586792,
|
|
"epoch": 3.679778113968734,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004087872929686673,
|
|
"loss": 5.1279,
|
|
"mean_token_accuracy": 0.17425395399332047,
|
|
"num_tokens": 8076085.0,
|
|
"step": 3650
|
|
},
|
|
{
|
|
"entropy": 5.540618562698365,
|
|
"epoch": 3.6848209783156833,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00040846816036787056,
|
|
"loss": 5.2578,
|
|
"mean_token_accuracy": 0.16293734610080718,
|
|
"num_tokens": 8088218.0,
|
|
"step": 3655
|
|
},
|
|
{
|
|
"entropy": 5.491701459884643,
|
|
"epoch": 3.6898638426626325,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004081486129449527,
|
|
"loss": 5.1834,
|
|
"mean_token_accuracy": 0.16654185503721236,
|
|
"num_tokens": 8099768.0,
|
|
"step": 3660
|
|
},
|
|
{
|
|
"entropy": 5.525410223007202,
|
|
"epoch": 3.6949067070095816,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00040782865169307423,
|
|
"loss": 5.1673,
|
|
"mean_token_accuracy": 0.17311840653419494,
|
|
"num_tokens": 8109858.0,
|
|
"step": 3665
|
|
},
|
|
{
|
|
"entropy": 5.635105133056641,
|
|
"epoch": 3.6999495713565302,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004075082776066822,
|
|
"loss": 5.2699,
|
|
"mean_token_accuracy": 0.17256860733032225,
|
|
"num_tokens": 8120262.0,
|
|
"step": 3670
|
|
},
|
|
{
|
|
"entropy": 5.633985710144043,
|
|
"epoch": 3.7049924357034794,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004071874916815068,
|
|
"loss": 5.2252,
|
|
"mean_token_accuracy": 0.1700965315103531,
|
|
"num_tokens": 8131483.0,
|
|
"step": 3675
|
|
},
|
|
{
|
|
"entropy": 5.565285682678223,
|
|
"epoch": 3.7100353000504285,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00040686629491455787,
|
|
"loss": 5.2592,
|
|
"mean_token_accuracy": 0.16653827726840972,
|
|
"num_tokens": 8143431.0,
|
|
"step": 3680
|
|
},
|
|
{
|
|
"entropy": 5.641773891448975,
|
|
"epoch": 3.7150781643973776,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00040654468830412247,
|
|
"loss": 5.2758,
|
|
"mean_token_accuracy": 0.16420862525701524,
|
|
"num_tokens": 8154596.0,
|
|
"step": 3685
|
|
},
|
|
{
|
|
"entropy": 5.557222461700439,
|
|
"epoch": 3.7201210287443267,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00040622267284976134,
|
|
"loss": 5.2412,
|
|
"mean_token_accuracy": 0.16766396015882493,
|
|
"num_tokens": 8165509.0,
|
|
"step": 3690
|
|
},
|
|
{
|
|
"entropy": 5.5014101505279545,
|
|
"epoch": 3.725163893091276,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00040590024955230596,
|
|
"loss": 5.1346,
|
|
"mean_token_accuracy": 0.1745723605155945,
|
|
"num_tokens": 8176275.0,
|
|
"step": 3695
|
|
},
|
|
{
|
|
"entropy": 5.546860265731811,
|
|
"epoch": 3.730206757438225,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004055774194138554,
|
|
"loss": 5.2458,
|
|
"mean_token_accuracy": 0.16376573592424393,
|
|
"num_tokens": 8187820.0,
|
|
"step": 3700
|
|
},
|
|
{
|
|
"entropy": 5.487030458450318,
|
|
"epoch": 3.735249621785174,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004052541834377731,
|
|
"loss": 5.1298,
|
|
"mean_token_accuracy": 0.17279367297887802,
|
|
"num_tokens": 8197750.0,
|
|
"step": 3705
|
|
},
|
|
{
|
|
"entropy": 5.52157416343689,
|
|
"epoch": 3.740292486132123,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00040493054262868397,
|
|
"loss": 5.2294,
|
|
"mean_token_accuracy": 0.16765167117118834,
|
|
"num_tokens": 8209275.0,
|
|
"step": 3710
|
|
},
|
|
{
|
|
"entropy": 5.623435354232788,
|
|
"epoch": 3.7453353504790723,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00040460649799247116,
|
|
"loss": 5.274,
|
|
"mean_token_accuracy": 0.16619176417589188,
|
|
"num_tokens": 8221280.0,
|
|
"step": 3715
|
|
},
|
|
{
|
|
"entropy": 5.5495246887207035,
|
|
"epoch": 3.7503782148260214,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004042820505362728,
|
|
"loss": 5.1579,
|
|
"mean_token_accuracy": 0.17330808341503143,
|
|
"num_tokens": 8231679.0,
|
|
"step": 3720
|
|
},
|
|
{
|
|
"entropy": 5.518253946304322,
|
|
"epoch": 3.7554210791729705,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004039572012684792,
|
|
"loss": 5.1609,
|
|
"mean_token_accuracy": 0.1804614931344986,
|
|
"num_tokens": 8242014.0,
|
|
"step": 3725
|
|
},
|
|
{
|
|
"entropy": 5.509087562561035,
|
|
"epoch": 3.760463943519919,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00040363195119872916,
|
|
"loss": 5.1316,
|
|
"mean_token_accuracy": 0.17738175839185716,
|
|
"num_tokens": 8253128.0,
|
|
"step": 3730
|
|
},
|
|
{
|
|
"entropy": 5.481418514251709,
|
|
"epoch": 3.7655068078668683,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00040330630133790757,
|
|
"loss": 5.0826,
|
|
"mean_token_accuracy": 0.1816011145710945,
|
|
"num_tokens": 8264643.0,
|
|
"step": 3735
|
|
},
|
|
{
|
|
"entropy": 5.557744121551513,
|
|
"epoch": 3.7705496722138174,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00040298025269814165,
|
|
"loss": 5.2476,
|
|
"mean_token_accuracy": 0.1693367749452591,
|
|
"num_tokens": 8275009.0,
|
|
"step": 3740
|
|
},
|
|
{
|
|
"entropy": 5.485255146026612,
|
|
"epoch": 3.7755925365607665,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004026538062927981,
|
|
"loss": 5.1841,
|
|
"mean_token_accuracy": 0.17268425226211548,
|
|
"num_tokens": 8286321.0,
|
|
"step": 3745
|
|
},
|
|
{
|
|
"entropy": 5.526268100738525,
|
|
"epoch": 3.7806354009077157,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00040232696313647985,
|
|
"loss": 5.1882,
|
|
"mean_token_accuracy": 0.16982749551534654,
|
|
"num_tokens": 8297216.0,
|
|
"step": 3750
|
|
},
|
|
{
|
|
"entropy": 5.589543390274048,
|
|
"epoch": 3.7856782652546648,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004019997242450231,
|
|
"loss": 5.1807,
|
|
"mean_token_accuracy": 0.1681605100631714,
|
|
"num_tokens": 8309053.0,
|
|
"step": 3755
|
|
},
|
|
{
|
|
"entropy": 5.585790061950684,
|
|
"epoch": 3.7907211296016134,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004016720906354937,
|
|
"loss": 5.167,
|
|
"mean_token_accuracy": 0.1676460698246956,
|
|
"num_tokens": 8318840.0,
|
|
"step": 3760
|
|
},
|
|
{
|
|
"entropy": 5.488352680206299,
|
|
"epoch": 3.7957639939485626,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004013440633261845,
|
|
"loss": 5.1861,
|
|
"mean_token_accuracy": 0.17753762751817703,
|
|
"num_tokens": 8329127.0,
|
|
"step": 3765
|
|
},
|
|
{
|
|
"entropy": 5.540678644180298,
|
|
"epoch": 3.8008068582955117,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00040101564333661213,
|
|
"loss": 5.1762,
|
|
"mean_token_accuracy": 0.16890554130077362,
|
|
"num_tokens": 8340651.0,
|
|
"step": 3770
|
|
},
|
|
{
|
|
"entropy": 5.52897891998291,
|
|
"epoch": 3.805849722642461,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00040068683168751316,
|
|
"loss": 5.1679,
|
|
"mean_token_accuracy": 0.1740403175354004,
|
|
"num_tokens": 8351399.0,
|
|
"step": 3775
|
|
},
|
|
{
|
|
"entropy": 5.552410459518432,
|
|
"epoch": 3.81089258698941,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00040035762940084205,
|
|
"loss": 5.1664,
|
|
"mean_token_accuracy": 0.1713647425174713,
|
|
"num_tokens": 8361756.0,
|
|
"step": 3780
|
|
},
|
|
{
|
|
"entropy": 5.575224256515503,
|
|
"epoch": 3.815935451336359,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000400028037499767,
|
|
"loss": 5.1949,
|
|
"mean_token_accuracy": 0.16735757291316986,
|
|
"num_tokens": 8372408.0,
|
|
"step": 3785
|
|
},
|
|
{
|
|
"entropy": 5.491192865371704,
|
|
"epoch": 3.820978315683308,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003996980570086673,
|
|
"loss": 5.1288,
|
|
"mean_token_accuracy": 0.17830123901367187,
|
|
"num_tokens": 8382788.0,
|
|
"step": 3790
|
|
},
|
|
{
|
|
"entropy": 5.456447315216065,
|
|
"epoch": 3.8260211800302573,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00039936768895312987,
|
|
"loss": 5.2046,
|
|
"mean_token_accuracy": 0.17088985294103623,
|
|
"num_tokens": 8393693.0,
|
|
"step": 3795
|
|
},
|
|
{
|
|
"entropy": 5.543635463714599,
|
|
"epoch": 3.8310640443772064,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00039903693435994626,
|
|
"loss": 5.1799,
|
|
"mean_token_accuracy": 0.1809312805533409,
|
|
"num_tokens": 8405163.0,
|
|
"step": 3800
|
|
},
|
|
{
|
|
"entropy": 5.5416360855102536,
|
|
"epoch": 3.8361069087241555,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00039870579425710937,
|
|
"loss": 5.1815,
|
|
"mean_token_accuracy": 0.16874978542327881,
|
|
"num_tokens": 8416987.0,
|
|
"step": 3805
|
|
},
|
|
{
|
|
"entropy": 5.543004035949707,
|
|
"epoch": 3.8411497730711046,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003983742696738103,
|
|
"loss": 5.2159,
|
|
"mean_token_accuracy": 0.16193017065525056,
|
|
"num_tokens": 8428864.0,
|
|
"step": 3810
|
|
},
|
|
{
|
|
"entropy": 5.472354793548584,
|
|
"epoch": 3.8461926374180537,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003980423616404351,
|
|
"loss": 5.1145,
|
|
"mean_token_accuracy": 0.1718842253088951,
|
|
"num_tokens": 8438970.0,
|
|
"step": 3815
|
|
},
|
|
{
|
|
"entropy": 5.6912829875946045,
|
|
"epoch": 3.8512355017650024,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003977100711885616,
|
|
"loss": 5.27,
|
|
"mean_token_accuracy": 0.1604915589094162,
|
|
"num_tokens": 8448935.0,
|
|
"step": 3820
|
|
},
|
|
{
|
|
"entropy": 5.571341705322266,
|
|
"epoch": 3.8562783661119515,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00039737739935095614,
|
|
"loss": 5.2292,
|
|
"mean_token_accuracy": 0.17025845050811766,
|
|
"num_tokens": 8458908.0,
|
|
"step": 3825
|
|
},
|
|
{
|
|
"entropy": 5.46253547668457,
|
|
"epoch": 3.8613212304589006,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003970443471615705,
|
|
"loss": 5.1226,
|
|
"mean_token_accuracy": 0.1824743166565895,
|
|
"num_tokens": 8469161.0,
|
|
"step": 3830
|
|
},
|
|
{
|
|
"entropy": 5.526420211791992,
|
|
"epoch": 3.8663640948058497,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003967109156555386,
|
|
"loss": 5.1266,
|
|
"mean_token_accuracy": 0.17558418214321136,
|
|
"num_tokens": 8479821.0,
|
|
"step": 3835
|
|
},
|
|
{
|
|
"entropy": 5.533747005462646,
|
|
"epoch": 3.871406959152799,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00039637710586917324,
|
|
"loss": 5.2273,
|
|
"mean_token_accuracy": 0.16653116643428803,
|
|
"num_tokens": 8489837.0,
|
|
"step": 3840
|
|
},
|
|
{
|
|
"entropy": 5.5383467197418215,
|
|
"epoch": 3.876449823499748,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00039604291883996287,
|
|
"loss": 5.2157,
|
|
"mean_token_accuracy": 0.17093627005815507,
|
|
"num_tokens": 8501241.0,
|
|
"step": 3845
|
|
},
|
|
{
|
|
"entropy": 5.579620122909546,
|
|
"epoch": 3.8814926878466967,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00039570835560656854,
|
|
"loss": 5.2384,
|
|
"mean_token_accuracy": 0.17254853248596191,
|
|
"num_tokens": 8512558.0,
|
|
"step": 3850
|
|
},
|
|
{
|
|
"entropy": 5.501360273361206,
|
|
"epoch": 3.8865355521936458,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003953734172088205,
|
|
"loss": 5.1082,
|
|
"mean_token_accuracy": 0.17222679555416107,
|
|
"num_tokens": 8522870.0,
|
|
"step": 3855
|
|
},
|
|
{
|
|
"entropy": 5.489282083511353,
|
|
"epoch": 3.891578416540595,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00039503810468771494,
|
|
"loss": 5.1833,
|
|
"mean_token_accuracy": 0.16986328959465027,
|
|
"num_tokens": 8533749.0,
|
|
"step": 3860
|
|
},
|
|
{
|
|
"entropy": 5.596812009811401,
|
|
"epoch": 3.896621280887544,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003947024190854111,
|
|
"loss": 5.2081,
|
|
"mean_token_accuracy": 0.17518100887537003,
|
|
"num_tokens": 8544700.0,
|
|
"step": 3865
|
|
},
|
|
{
|
|
"entropy": 5.515973043441773,
|
|
"epoch": 3.901664145234493,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00039436636144522737,
|
|
"loss": 5.2052,
|
|
"mean_token_accuracy": 0.16411028504371644,
|
|
"num_tokens": 8555236.0,
|
|
"step": 3870
|
|
},
|
|
{
|
|
"entropy": 5.50197205543518,
|
|
"epoch": 3.9067070095814422,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00039402993281163875,
|
|
"loss": 5.2238,
|
|
"mean_token_accuracy": 0.1677716538310051,
|
|
"num_tokens": 8566597.0,
|
|
"step": 3875
|
|
},
|
|
{
|
|
"entropy": 5.565766859054565,
|
|
"epoch": 3.9117498739283914,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003936931342302732,
|
|
"loss": 5.2006,
|
|
"mean_token_accuracy": 0.17089226096868515,
|
|
"num_tokens": 8577297.0,
|
|
"step": 3880
|
|
},
|
|
{
|
|
"entropy": 5.615969038009643,
|
|
"epoch": 3.9167927382753405,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003933559667479084,
|
|
"loss": 5.2114,
|
|
"mean_token_accuracy": 0.17481039613485336,
|
|
"num_tokens": 8587145.0,
|
|
"step": 3885
|
|
},
|
|
{
|
|
"entropy": 5.523828649520874,
|
|
"epoch": 3.9218356026222896,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00039301843141246876,
|
|
"loss": 5.0917,
|
|
"mean_token_accuracy": 0.1800573080778122,
|
|
"num_tokens": 8597410.0,
|
|
"step": 3890
|
|
},
|
|
{
|
|
"entropy": 5.486855506896973,
|
|
"epoch": 3.9268784669692387,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00039268052927302185,
|
|
"loss": 5.152,
|
|
"mean_token_accuracy": 0.17523978948593139,
|
|
"num_tokens": 8607839.0,
|
|
"step": 3895
|
|
},
|
|
{
|
|
"entropy": 5.507344198226929,
|
|
"epoch": 3.931921331316188,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003923422613797753,
|
|
"loss": 5.145,
|
|
"mean_token_accuracy": 0.17821708023548127,
|
|
"num_tokens": 8618517.0,
|
|
"step": 3900
|
|
},
|
|
{
|
|
"entropy": 5.443444013595581,
|
|
"epoch": 3.936964195663137,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003920036287840734,
|
|
"loss": 5.1196,
|
|
"mean_token_accuracy": 0.17776491343975068,
|
|
"num_tokens": 8629120.0,
|
|
"step": 3905
|
|
},
|
|
{
|
|
"entropy": 5.507380771636963,
|
|
"epoch": 3.9420070600100856,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00039166463253839434,
|
|
"loss": 5.1807,
|
|
"mean_token_accuracy": 0.17569509148597717,
|
|
"num_tokens": 8640503.0,
|
|
"step": 3910
|
|
},
|
|
{
|
|
"entropy": 5.548609113693237,
|
|
"epoch": 3.9470499243570347,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003913252736963462,
|
|
"loss": 5.1845,
|
|
"mean_token_accuracy": 0.18213451653718948,
|
|
"num_tokens": 8652575.0,
|
|
"step": 3915
|
|
},
|
|
{
|
|
"entropy": 5.547990179061889,
|
|
"epoch": 3.952092788703984,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000390985553312664,
|
|
"loss": 5.103,
|
|
"mean_token_accuracy": 0.17633057981729508,
|
|
"num_tokens": 8663646.0,
|
|
"step": 3920
|
|
},
|
|
{
|
|
"entropy": 5.468158292770386,
|
|
"epoch": 3.957135653050933,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003906454724432065,
|
|
"loss": 5.1768,
|
|
"mean_token_accuracy": 0.1767694115638733,
|
|
"num_tokens": 8673953.0,
|
|
"step": 3925
|
|
},
|
|
{
|
|
"entropy": 5.503249216079712,
|
|
"epoch": 3.962178517397882,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00039030503214495296,
|
|
"loss": 5.1754,
|
|
"mean_token_accuracy": 0.17349360436201094,
|
|
"num_tokens": 8685388.0,
|
|
"step": 3930
|
|
},
|
|
{
|
|
"entropy": 5.510561561584472,
|
|
"epoch": 3.967221381744831,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003899642334759996,
|
|
"loss": 5.1526,
|
|
"mean_token_accuracy": 0.16797081828117372,
|
|
"num_tokens": 8695205.0,
|
|
"step": 3935
|
|
},
|
|
{
|
|
"entropy": 5.522814512252808,
|
|
"epoch": 3.97226424609178,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003896230774955567,
|
|
"loss": 5.1497,
|
|
"mean_token_accuracy": 0.17399344295263292,
|
|
"num_tokens": 8706118.0,
|
|
"step": 3940
|
|
},
|
|
{
|
|
"entropy": 5.531307697296143,
|
|
"epoch": 3.977307110438729,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00038928156526394476,
|
|
"loss": 5.1456,
|
|
"mean_token_accuracy": 0.17701290547847748,
|
|
"num_tokens": 8716926.0,
|
|
"step": 3945
|
|
},
|
|
{
|
|
"entropy": 5.55611481666565,
|
|
"epoch": 3.982349974785678,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0003889396978425918,
|
|
"loss": 5.2677,
|
|
"mean_token_accuracy": 0.17022654563188552,
|
|
"num_tokens": 8729409.0,
|
|
"step": 3950
|
|
},
|
|
{
|
|
"entropy": 5.540706205368042,
|
|
"epoch": 3.987392839132627,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003885974762940295,
|
|
"loss": 5.1903,
|
|
"mean_token_accuracy": 0.17912053614854812,
|
|
"num_tokens": 8740140.0,
|
|
"step": 3955
|
|
},
|
|
{
|
|
"entropy": 5.536376667022705,
|
|
"epoch": 3.9924357034795763,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003882549016818905,
|
|
"loss": 5.0777,
|
|
"mean_token_accuracy": 0.1807361587882042,
|
|
"num_tokens": 8749176.0,
|
|
"step": 3960
|
|
},
|
|
{
|
|
"entropy": 5.494466400146484,
|
|
"epoch": 3.9974785678265254,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00038791197507090456,
|
|
"loss": 5.2207,
|
|
"mean_token_accuracy": 0.1752302274107933,
|
|
"num_tokens": 8761575.0,
|
|
"step": 3965
|
|
},
|
|
{
|
|
"entropy": 5.447997570037842,
|
|
"epoch": 4.00201714573878,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00038756869752689527,
|
|
"loss": 5.0317,
|
|
"mean_token_accuracy": 0.1796759201420678,
|
|
"num_tokens": 8771147.0,
|
|
"step": 3970
|
|
},
|
|
{
|
|
"entropy": 5.479295158386231,
|
|
"epoch": 4.007060010085729,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003872250701167775,
|
|
"loss": 4.8945,
|
|
"mean_token_accuracy": 0.19214107245206832,
|
|
"num_tokens": 8781936.0,
|
|
"step": 3975
|
|
},
|
|
{
|
|
"entropy": 5.523778438568115,
|
|
"epoch": 4.012102874432678,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00038688109390855297,
|
|
"loss": 4.9124,
|
|
"mean_token_accuracy": 0.18693623542785645,
|
|
"num_tokens": 8791492.0,
|
|
"step": 3980
|
|
},
|
|
{
|
|
"entropy": 5.490765380859375,
|
|
"epoch": 4.017145738779627,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003865367699713077,
|
|
"loss": 4.915,
|
|
"mean_token_accuracy": 0.19355802237987518,
|
|
"num_tokens": 8802552.0,
|
|
"step": 3985
|
|
},
|
|
{
|
|
"entropy": 5.573561906814575,
|
|
"epoch": 4.022188603126576,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00038619209937520836,
|
|
"loss": 4.9609,
|
|
"mean_token_accuracy": 0.19315825253725052,
|
|
"num_tokens": 8813885.0,
|
|
"step": 3990
|
|
},
|
|
{
|
|
"entropy": 5.545063018798828,
|
|
"epoch": 4.027231467473525,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00038584708319149924,
|
|
"loss": 4.924,
|
|
"mean_token_accuracy": 0.1915197879076004,
|
|
"num_tokens": 8825503.0,
|
|
"step": 3995
|
|
},
|
|
{
|
|
"entropy": 5.532052516937256,
|
|
"epoch": 4.032274331820474,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00038550172249249856,
|
|
"loss": 4.9654,
|
|
"mean_token_accuracy": 0.18269565850496292,
|
|
"num_tokens": 8835581.0,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"epoch": 4.032274331820474,
|
|
"eval_entropy": 5.2951136175306095,
|
|
"eval_loss": 5.6981201171875,
|
|
"eval_mean_token_accuracy": 0.1518737894243563,
|
|
"eval_num_tokens": 8835581.0,
|
|
"eval_runtime": 2.0087,
|
|
"eval_samples_per_second": 1562.169,
|
|
"eval_steps_per_second": 195.644,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"entropy": 5.441691541671753,
|
|
"epoch": 4.0373171961674235,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003851560183515953,
|
|
"loss": 4.8364,
|
|
"mean_token_accuracy": 0.1995864525437355,
|
|
"num_tokens": 8847453.0,
|
|
"step": 4005
|
|
},
|
|
{
|
|
"entropy": 5.478153371810913,
|
|
"epoch": 4.042360060514373,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000384809971843246,
|
|
"loss": 4.943,
|
|
"mean_token_accuracy": 0.19371214061975478,
|
|
"num_tokens": 8858098.0,
|
|
"step": 4010
|
|
},
|
|
{
|
|
"entropy": 5.4701759815216064,
|
|
"epoch": 4.047402924861321,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000384463584042971,
|
|
"loss": 4.8983,
|
|
"mean_token_accuracy": 0.19070684611797334,
|
|
"num_tokens": 8869601.0,
|
|
"step": 4015
|
|
},
|
|
{
|
|
"entropy": 5.404227542877197,
|
|
"epoch": 4.05244578920827,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00038411685602735175,
|
|
"loss": 4.8598,
|
|
"mean_token_accuracy": 0.1950615182518959,
|
|
"num_tokens": 8881926.0,
|
|
"step": 4020
|
|
},
|
|
{
|
|
"entropy": 5.497881841659546,
|
|
"epoch": 4.057488653555219,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00038376978887402687,
|
|
"loss": 4.9857,
|
|
"mean_token_accuracy": 0.1911414608359337,
|
|
"num_tokens": 8892714.0,
|
|
"step": 4025
|
|
},
|
|
{
|
|
"entropy": 5.548092031478882,
|
|
"epoch": 4.062531517902168,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003834223836616891,
|
|
"loss": 4.9201,
|
|
"mean_token_accuracy": 0.19523752927780152,
|
|
"num_tokens": 8905407.0,
|
|
"step": 4030
|
|
},
|
|
{
|
|
"entropy": 5.482694101333618,
|
|
"epoch": 4.067574382249117,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00038307464147008176,
|
|
"loss": 4.9763,
|
|
"mean_token_accuracy": 0.18707808405160903,
|
|
"num_tokens": 8915474.0,
|
|
"step": 4035
|
|
},
|
|
{
|
|
"entropy": 5.473349285125733,
|
|
"epoch": 4.072617246596066,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003827265633799957,
|
|
"loss": 4.9164,
|
|
"mean_token_accuracy": 0.18727621883153917,
|
|
"num_tokens": 8926168.0,
|
|
"step": 4040
|
|
},
|
|
{
|
|
"entropy": 5.5510602474212645,
|
|
"epoch": 4.0776601109430155,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003823781504732657,
|
|
"loss": 4.9872,
|
|
"mean_token_accuracy": 0.18850124329328538,
|
|
"num_tokens": 8936995.0,
|
|
"step": 4045
|
|
},
|
|
{
|
|
"entropy": 5.502942848205566,
|
|
"epoch": 4.082702975289965,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003820294038327669,
|
|
"loss": 5.0291,
|
|
"mean_token_accuracy": 0.17579314410686492,
|
|
"num_tokens": 8947675.0,
|
|
"step": 4050
|
|
},
|
|
{
|
|
"entropy": 5.43959641456604,
|
|
"epoch": 4.087745839636914,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003816803245424121,
|
|
"loss": 4.9583,
|
|
"mean_token_accuracy": 0.18968660086393357,
|
|
"num_tokens": 8958486.0,
|
|
"step": 4055
|
|
},
|
|
{
|
|
"entropy": 5.6069999694824215,
|
|
"epoch": 4.092788703983863,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003813309136871478,
|
|
"loss": 4.9437,
|
|
"mean_token_accuracy": 0.18789149820804596,
|
|
"num_tokens": 8968208.0,
|
|
"step": 4060
|
|
},
|
|
{
|
|
"entropy": 5.42473406791687,
|
|
"epoch": 4.097831568330812,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00038098117235295083,
|
|
"loss": 4.8809,
|
|
"mean_token_accuracy": 0.19628652930259705,
|
|
"num_tokens": 8978327.0,
|
|
"step": 4065
|
|
},
|
|
{
|
|
"entropy": 5.450071191787719,
|
|
"epoch": 4.102874432677761,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003806311016268254,
|
|
"loss": 5.0098,
|
|
"mean_token_accuracy": 0.18902334868907927,
|
|
"num_tokens": 8990944.0,
|
|
"step": 4070
|
|
},
|
|
{
|
|
"entropy": 5.515642833709717,
|
|
"epoch": 4.10791729702471,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003802807025967995,
|
|
"loss": 4.9525,
|
|
"mean_token_accuracy": 0.18901822715997696,
|
|
"num_tokens": 9002505.0,
|
|
"step": 4075
|
|
},
|
|
{
|
|
"entropy": 5.492382621765136,
|
|
"epoch": 4.112960161371659,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003799299763519213,
|
|
"loss": 4.9218,
|
|
"mean_token_accuracy": 0.18582296818494798,
|
|
"num_tokens": 9013453.0,
|
|
"step": 4080
|
|
},
|
|
{
|
|
"entropy": 5.464056396484375,
|
|
"epoch": 4.118003025718608,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003795789239822561,
|
|
"loss": 4.9046,
|
|
"mean_token_accuracy": 0.1943138062953949,
|
|
"num_tokens": 9025318.0,
|
|
"step": 4085
|
|
},
|
|
{
|
|
"entropy": 5.504607772827148,
|
|
"epoch": 4.1230458900655576,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003792275465788826,
|
|
"loss": 4.8736,
|
|
"mean_token_accuracy": 0.19997980892658235,
|
|
"num_tokens": 9034334.0,
|
|
"step": 4090
|
|
},
|
|
{
|
|
"entropy": 5.426046752929688,
|
|
"epoch": 4.128088754412507,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00037887584523389013,
|
|
"loss": 4.9473,
|
|
"mean_token_accuracy": 0.1910230666399002,
|
|
"num_tokens": 9046096.0,
|
|
"step": 4095
|
|
},
|
|
{
|
|
"entropy": 5.422527837753296,
|
|
"epoch": 4.133131618759456,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00037852382104037454,
|
|
"loss": 4.9216,
|
|
"mean_token_accuracy": 0.1813015878200531,
|
|
"num_tokens": 9057540.0,
|
|
"step": 4100
|
|
},
|
|
{
|
|
"entropy": 5.500780868530273,
|
|
"epoch": 4.138174483106404,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00037817147509243497,
|
|
"loss": 4.9006,
|
|
"mean_token_accuracy": 0.18728822469711304,
|
|
"num_tokens": 9068770.0,
|
|
"step": 4105
|
|
},
|
|
{
|
|
"entropy": 5.52485876083374,
|
|
"epoch": 4.143217347453353,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003778188084851709,
|
|
"loss": 5.0102,
|
|
"mean_token_accuracy": 0.17882611155509948,
|
|
"num_tokens": 9080026.0,
|
|
"step": 4110
|
|
},
|
|
{
|
|
"entropy": 5.423612928390503,
|
|
"epoch": 4.148260211800302,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00037746582231467816,
|
|
"loss": 4.9117,
|
|
"mean_token_accuracy": 0.19607280492782592,
|
|
"num_tokens": 9092013.0,
|
|
"step": 4115
|
|
},
|
|
{
|
|
"entropy": 5.480008506774903,
|
|
"epoch": 4.153303076147251,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003771125176780461,
|
|
"loss": 4.9532,
|
|
"mean_token_accuracy": 0.18666686713695527,
|
|
"num_tokens": 9104739.0,
|
|
"step": 4120
|
|
},
|
|
{
|
|
"entropy": 5.438631057739258,
|
|
"epoch": 4.1583459404942005,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00037675889567335345,
|
|
"loss": 4.9032,
|
|
"mean_token_accuracy": 0.1882699802517891,
|
|
"num_tokens": 9116321.0,
|
|
"step": 4125
|
|
},
|
|
{
|
|
"entropy": 5.413467502593994,
|
|
"epoch": 4.16338880484115,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00037640495739966573,
|
|
"loss": 4.8726,
|
|
"mean_token_accuracy": 0.19468048214912415,
|
|
"num_tokens": 9126844.0,
|
|
"step": 4130
|
|
},
|
|
{
|
|
"entropy": 5.429811573028564,
|
|
"epoch": 4.168431669188099,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00037605070395703124,
|
|
"loss": 4.9145,
|
|
"mean_token_accuracy": 0.19117997139692305,
|
|
"num_tokens": 9137887.0,
|
|
"step": 4135
|
|
},
|
|
{
|
|
"entropy": 5.440280389785767,
|
|
"epoch": 4.173474533535048,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00037569613644647783,
|
|
"loss": 4.8592,
|
|
"mean_token_accuracy": 0.19679671227931977,
|
|
"num_tokens": 9148087.0,
|
|
"step": 4140
|
|
},
|
|
{
|
|
"entropy": 5.348052024841309,
|
|
"epoch": 4.178517397881997,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003753412559700094,
|
|
"loss": 4.8474,
|
|
"mean_token_accuracy": 0.19631566852331161,
|
|
"num_tokens": 9159628.0,
|
|
"step": 4145
|
|
},
|
|
{
|
|
"entropy": 5.40684404373169,
|
|
"epoch": 4.183560262228946,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00037498606363060285,
|
|
"loss": 4.9295,
|
|
"mean_token_accuracy": 0.18776677399873734,
|
|
"num_tokens": 9171295.0,
|
|
"step": 4150
|
|
},
|
|
{
|
|
"entropy": 5.421151542663575,
|
|
"epoch": 4.188603126575895,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003746305605322041,
|
|
"loss": 4.8839,
|
|
"mean_token_accuracy": 0.189324226975441,
|
|
"num_tokens": 9181619.0,
|
|
"step": 4155
|
|
},
|
|
{
|
|
"entropy": 5.4041276454925535,
|
|
"epoch": 4.193645990922844,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00037427474777972477,
|
|
"loss": 4.8893,
|
|
"mean_token_accuracy": 0.19542089104652405,
|
|
"num_tokens": 9192184.0,
|
|
"step": 4160
|
|
},
|
|
{
|
|
"entropy": 5.407623958587647,
|
|
"epoch": 4.198688855269793,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00037391862647903943,
|
|
"loss": 4.9579,
|
|
"mean_token_accuracy": 0.18405826836824418,
|
|
"num_tokens": 9201452.0,
|
|
"step": 4165
|
|
},
|
|
{
|
|
"entropy": 5.436958503723145,
|
|
"epoch": 4.2037317196167425,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00037356219773698106,
|
|
"loss": 4.931,
|
|
"mean_token_accuracy": 0.18935900330543518,
|
|
"num_tokens": 9210945.0,
|
|
"step": 4170
|
|
},
|
|
{
|
|
"entropy": 5.422917175292969,
|
|
"epoch": 4.208774583963692,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00037320546266133854,
|
|
"loss": 4.8019,
|
|
"mean_token_accuracy": 0.1974889099597931,
|
|
"num_tokens": 9221577.0,
|
|
"step": 4175
|
|
},
|
|
{
|
|
"entropy": 5.359337902069091,
|
|
"epoch": 4.213817448310641,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003728484223608526,
|
|
"loss": 4.9397,
|
|
"mean_token_accuracy": 0.18506143540143966,
|
|
"num_tokens": 9232353.0,
|
|
"step": 4180
|
|
},
|
|
{
|
|
"entropy": 5.474852514266968,
|
|
"epoch": 4.21886031265759,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00037249107794521286,
|
|
"loss": 4.999,
|
|
"mean_token_accuracy": 0.18769071996212006,
|
|
"num_tokens": 9243157.0,
|
|
"step": 4185
|
|
},
|
|
{
|
|
"entropy": 5.494953155517578,
|
|
"epoch": 4.223903177004539,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00037213343052505394,
|
|
"loss": 5.0012,
|
|
"mean_token_accuracy": 0.18003073781728746,
|
|
"num_tokens": 9256750.0,
|
|
"step": 4190
|
|
},
|
|
{
|
|
"entropy": 5.472083234786988,
|
|
"epoch": 4.228946041351487,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003717754812119523,
|
|
"loss": 4.956,
|
|
"mean_token_accuracy": 0.1967128187417984,
|
|
"num_tokens": 9268436.0,
|
|
"step": 4195
|
|
},
|
|
{
|
|
"entropy": 5.4630804538726805,
|
|
"epoch": 4.233988905698436,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003714172311184226,
|
|
"loss": 4.9528,
|
|
"mean_token_accuracy": 0.18292782902717591,
|
|
"num_tokens": 9278326.0,
|
|
"step": 4200
|
|
},
|
|
{
|
|
"entropy": 5.484615135192871,
|
|
"epoch": 4.2390317700453854,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003710586813579144,
|
|
"loss": 4.9238,
|
|
"mean_token_accuracy": 0.19408095479011536,
|
|
"num_tokens": 9289788.0,
|
|
"step": 4205
|
|
},
|
|
{
|
|
"entropy": 5.494198989868164,
|
|
"epoch": 4.244074634392335,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00037069983304480875,
|
|
"loss": 4.9014,
|
|
"mean_token_accuracy": 0.18106771111488343,
|
|
"num_tokens": 9300932.0,
|
|
"step": 4210
|
|
},
|
|
{
|
|
"entropy": 5.444451522827149,
|
|
"epoch": 4.249117498739284,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003703406872944144,
|
|
"loss": 4.9871,
|
|
"mean_token_accuracy": 0.1897885501384735,
|
|
"num_tokens": 9313043.0,
|
|
"step": 4215
|
|
},
|
|
{
|
|
"entropy": 5.472175931930542,
|
|
"epoch": 4.254160363086233,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003699812452229645,
|
|
"loss": 5.0034,
|
|
"mean_token_accuracy": 0.18668656200170516,
|
|
"num_tokens": 9324122.0,
|
|
"step": 4220
|
|
},
|
|
{
|
|
"entropy": 5.4078141212463375,
|
|
"epoch": 4.259203227433182,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00036962150794761354,
|
|
"loss": 4.8625,
|
|
"mean_token_accuracy": 0.19438074082136153,
|
|
"num_tokens": 9335745.0,
|
|
"step": 4225
|
|
},
|
|
{
|
|
"entropy": 5.443893241882324,
|
|
"epoch": 4.264246091780131,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003692614765864331,
|
|
"loss": 4.9353,
|
|
"mean_token_accuracy": 0.19140463024377824,
|
|
"num_tokens": 9346843.0,
|
|
"step": 4230
|
|
},
|
|
{
|
|
"entropy": 5.42264347076416,
|
|
"epoch": 4.26928895612708,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00036890115225840904,
|
|
"loss": 4.9312,
|
|
"mean_token_accuracy": 0.1884590283036232,
|
|
"num_tokens": 9357782.0,
|
|
"step": 4235
|
|
},
|
|
{
|
|
"entropy": 5.500774240493774,
|
|
"epoch": 4.274331820474029,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003685405360834376,
|
|
"loss": 5.066,
|
|
"mean_token_accuracy": 0.18616201281547545,
|
|
"num_tokens": 9370704.0,
|
|
"step": 4240
|
|
},
|
|
{
|
|
"entropy": 5.496207666397095,
|
|
"epoch": 4.279374684820978,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00036817962918232234,
|
|
"loss": 4.9956,
|
|
"mean_token_accuracy": 0.18022577613592147,
|
|
"num_tokens": 9382174.0,
|
|
"step": 4245
|
|
},
|
|
{
|
|
"entropy": 5.461671495437622,
|
|
"epoch": 4.2844175491679275,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00036781843267677,
|
|
"loss": 4.891,
|
|
"mean_token_accuracy": 0.19812634885311126,
|
|
"num_tokens": 9393380.0,
|
|
"step": 4250
|
|
},
|
|
{
|
|
"entropy": 5.511389446258545,
|
|
"epoch": 4.289460413514877,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003674569476893878,
|
|
"loss": 4.9947,
|
|
"mean_token_accuracy": 0.1872989922761917,
|
|
"num_tokens": 9404194.0,
|
|
"step": 4255
|
|
},
|
|
{
|
|
"entropy": 5.418403482437133,
|
|
"epoch": 4.294503277861826,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003670951753436794,
|
|
"loss": 4.8961,
|
|
"mean_token_accuracy": 0.19839798361063005,
|
|
"num_tokens": 9415488.0,
|
|
"step": 4260
|
|
},
|
|
{
|
|
"entropy": 5.469805717468262,
|
|
"epoch": 4.299546142208775,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00036673311676404163,
|
|
"loss": 4.9808,
|
|
"mean_token_accuracy": 0.18969418555498124,
|
|
"num_tokens": 9426351.0,
|
|
"step": 4265
|
|
},
|
|
{
|
|
"entropy": 5.486623811721802,
|
|
"epoch": 4.304589006555724,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003663707730757608,
|
|
"loss": 4.9876,
|
|
"mean_token_accuracy": 0.18251553624868394,
|
|
"num_tokens": 9438302.0,
|
|
"step": 4270
|
|
},
|
|
{
|
|
"entropy": 5.405372142791748,
|
|
"epoch": 4.309631870902673,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00036600814540500956,
|
|
"loss": 4.8864,
|
|
"mean_token_accuracy": 0.1842934623360634,
|
|
"num_tokens": 9449004.0,
|
|
"step": 4275
|
|
},
|
|
{
|
|
"entropy": 5.371332120895386,
|
|
"epoch": 4.314674735249622,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000365645234878843,
|
|
"loss": 4.8976,
|
|
"mean_token_accuracy": 0.1928465932607651,
|
|
"num_tokens": 9459624.0,
|
|
"step": 4280
|
|
},
|
|
{
|
|
"entropy": 5.409686946868897,
|
|
"epoch": 4.31971759959657,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00036528204262519545,
|
|
"loss": 4.9067,
|
|
"mean_token_accuracy": 0.1930206909775734,
|
|
"num_tokens": 9470380.0,
|
|
"step": 4285
|
|
},
|
|
{
|
|
"entropy": 5.430053758621216,
|
|
"epoch": 4.3247604639435195,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003649185697728768,
|
|
"loss": 4.899,
|
|
"mean_token_accuracy": 0.18901572674512862,
|
|
"num_tokens": 9480802.0,
|
|
"step": 4290
|
|
},
|
|
{
|
|
"entropy": 5.416997241973877,
|
|
"epoch": 4.329803328290469,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00036455481745156906,
|
|
"loss": 4.9041,
|
|
"mean_token_accuracy": 0.18799829632043838,
|
|
"num_tokens": 9491842.0,
|
|
"step": 4295
|
|
},
|
|
{
|
|
"entropy": 5.3991118431091305,
|
|
"epoch": 4.334846192637418,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003641907867918228,
|
|
"loss": 4.9211,
|
|
"mean_token_accuracy": 0.19682487845420837,
|
|
"num_tokens": 9503398.0,
|
|
"step": 4300
|
|
},
|
|
{
|
|
"entropy": 5.5496655941009525,
|
|
"epoch": 4.339889056984367,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003638264789250538,
|
|
"loss": 5.0603,
|
|
"mean_token_accuracy": 0.18734301924705504,
|
|
"num_tokens": 9513915.0,
|
|
"step": 4305
|
|
},
|
|
{
|
|
"entropy": 5.445239114761352,
|
|
"epoch": 4.344931921331316,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00036346189498353914,
|
|
"loss": 4.92,
|
|
"mean_token_accuracy": 0.19160290211439132,
|
|
"num_tokens": 9524630.0,
|
|
"step": 4310
|
|
},
|
|
{
|
|
"entropy": 5.447795438766479,
|
|
"epoch": 4.349974785678265,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003630970361004143,
|
|
"loss": 4.918,
|
|
"mean_token_accuracy": 0.19554964900016786,
|
|
"num_tokens": 9535135.0,
|
|
"step": 4315
|
|
},
|
|
{
|
|
"entropy": 5.484377765655518,
|
|
"epoch": 4.355017650025214,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003627319034096691,
|
|
"loss": 4.9662,
|
|
"mean_token_accuracy": 0.1896781951189041,
|
|
"num_tokens": 9544845.0,
|
|
"step": 4320
|
|
},
|
|
{
|
|
"entropy": 5.486665153503418,
|
|
"epoch": 4.360060514372163,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003623664980461443,
|
|
"loss": 4.9834,
|
|
"mean_token_accuracy": 0.1834908902645111,
|
|
"num_tokens": 9555498.0,
|
|
"step": 4325
|
|
},
|
|
{
|
|
"entropy": 5.4672764301300045,
|
|
"epoch": 4.3651033787191125,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00036200082114552806,
|
|
"loss": 4.9537,
|
|
"mean_token_accuracy": 0.1898433819413185,
|
|
"num_tokens": 9567995.0,
|
|
"step": 4330
|
|
},
|
|
{
|
|
"entropy": 5.413251781463623,
|
|
"epoch": 4.370146243066062,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00036163487384435294,
|
|
"loss": 4.9757,
|
|
"mean_token_accuracy": 0.18344618380069733,
|
|
"num_tokens": 9578199.0,
|
|
"step": 4335
|
|
},
|
|
{
|
|
"entropy": 5.532832288742066,
|
|
"epoch": 4.375189107413011,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003612686572799914,
|
|
"loss": 5.0579,
|
|
"mean_token_accuracy": 0.18481167405843735,
|
|
"num_tokens": 9588633.0,
|
|
"step": 4340
|
|
},
|
|
{
|
|
"entropy": 5.512619161605835,
|
|
"epoch": 4.38023197175996,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003609021725906529,
|
|
"loss": 4.9773,
|
|
"mean_token_accuracy": 0.19319147616624832,
|
|
"num_tokens": 9600269.0,
|
|
"step": 4345
|
|
},
|
|
{
|
|
"entropy": 5.452920627593994,
|
|
"epoch": 4.385274836106909,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00036053542091538045,
|
|
"loss": 4.9411,
|
|
"mean_token_accuracy": 0.1868438258767128,
|
|
"num_tokens": 9610699.0,
|
|
"step": 4350
|
|
},
|
|
{
|
|
"entropy": 5.481621026992798,
|
|
"epoch": 4.390317700453858,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00036016840339404664,
|
|
"loss": 4.9502,
|
|
"mean_token_accuracy": 0.19051971286535263,
|
|
"num_tokens": 9620541.0,
|
|
"step": 4355
|
|
},
|
|
{
|
|
"entropy": 5.414801406860351,
|
|
"epoch": 4.395360564800807,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003598011211673505,
|
|
"loss": 4.8956,
|
|
"mean_token_accuracy": 0.19339224845170974,
|
|
"num_tokens": 9632048.0,
|
|
"step": 4360
|
|
},
|
|
{
|
|
"entropy": 5.43272294998169,
|
|
"epoch": 4.400403429147756,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0003594335753768135,
|
|
"loss": 4.9385,
|
|
"mean_token_accuracy": 0.18861111402511596,
|
|
"num_tokens": 9643187.0,
|
|
"step": 4365
|
|
},
|
|
{
|
|
"entropy": 5.466538143157959,
|
|
"epoch": 4.405446293494705,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003590657671647766,
|
|
"loss": 4.9841,
|
|
"mean_token_accuracy": 0.18604212254285812,
|
|
"num_tokens": 9654375.0,
|
|
"step": 4370
|
|
},
|
|
{
|
|
"entropy": 5.455693197250366,
|
|
"epoch": 4.4104891578416545,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00035869769767439606,
|
|
"loss": 5.0398,
|
|
"mean_token_accuracy": 0.17666561752557755,
|
|
"num_tokens": 9666702.0,
|
|
"step": 4375
|
|
},
|
|
{
|
|
"entropy": 5.581379508972168,
|
|
"epoch": 4.415532022188603,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003583293680496404,
|
|
"loss": 5.0989,
|
|
"mean_token_accuracy": 0.1818246841430664,
|
|
"num_tokens": 9677989.0,
|
|
"step": 4380
|
|
},
|
|
{
|
|
"entropy": 5.456424427032471,
|
|
"epoch": 4.420574886535552,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003579607794352867,
|
|
"loss": 4.9003,
|
|
"mean_token_accuracy": 0.194338421523571,
|
|
"num_tokens": 9688763.0,
|
|
"step": 4385
|
|
},
|
|
{
|
|
"entropy": 5.405390357971191,
|
|
"epoch": 4.425617750882501,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00035759193297691674,
|
|
"loss": 4.9336,
|
|
"mean_token_accuracy": 0.18550542891025543,
|
|
"num_tokens": 9698457.0,
|
|
"step": 4390
|
|
},
|
|
{
|
|
"entropy": 5.374500799179077,
|
|
"epoch": 4.43066061522945,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00035722282982091373,
|
|
"loss": 4.9292,
|
|
"mean_token_accuracy": 0.18963116258382798,
|
|
"num_tokens": 9709564.0,
|
|
"step": 4395
|
|
},
|
|
{
|
|
"entropy": 5.378555965423584,
|
|
"epoch": 4.435703479576399,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003568534711144591,
|
|
"loss": 4.9044,
|
|
"mean_token_accuracy": 0.19540732204914094,
|
|
"num_tokens": 9719231.0,
|
|
"step": 4400
|
|
},
|
|
{
|
|
"entropy": 5.466189575195313,
|
|
"epoch": 4.440746343923348,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000356483858005528,
|
|
"loss": 4.9401,
|
|
"mean_token_accuracy": 0.18661027848720552,
|
|
"num_tokens": 9729785.0,
|
|
"step": 4405
|
|
},
|
|
{
|
|
"entropy": 5.419613313674927,
|
|
"epoch": 4.445789208270297,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00035611399164288654,
|
|
"loss": 4.8999,
|
|
"mean_token_accuracy": 0.1899822771549225,
|
|
"num_tokens": 9740292.0,
|
|
"step": 4410
|
|
},
|
|
{
|
|
"entropy": 5.45462851524353,
|
|
"epoch": 4.450832072617247,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0003557438731760879,
|
|
"loss": 4.9517,
|
|
"mean_token_accuracy": 0.193515445291996,
|
|
"num_tokens": 9750062.0,
|
|
"step": 4415
|
|
},
|
|
{
|
|
"entropy": 5.407767200469971,
|
|
"epoch": 4.455874936964196,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003553735037554689,
|
|
"loss": 4.9616,
|
|
"mean_token_accuracy": 0.19140054583549498,
|
|
"num_tokens": 9760427.0,
|
|
"step": 4420
|
|
},
|
|
{
|
|
"entropy": 5.45370831489563,
|
|
"epoch": 4.460917801311145,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003550028845321461,
|
|
"loss": 4.9871,
|
|
"mean_token_accuracy": 0.1904370054602623,
|
|
"num_tokens": 9771842.0,
|
|
"step": 4425
|
|
},
|
|
{
|
|
"entropy": 5.474357604980469,
|
|
"epoch": 4.465960665658094,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003546320166580127,
|
|
"loss": 4.9407,
|
|
"mean_token_accuracy": 0.18953434228897095,
|
|
"num_tokens": 9782860.0,
|
|
"step": 4430
|
|
},
|
|
{
|
|
"entropy": 5.478485345840454,
|
|
"epoch": 4.471003530005043,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00035426090128573464,
|
|
"loss": 4.9559,
|
|
"mean_token_accuracy": 0.18774261474609374,
|
|
"num_tokens": 9793361.0,
|
|
"step": 4435
|
|
},
|
|
{
|
|
"entropy": 5.357384872436524,
|
|
"epoch": 4.476046394351992,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00035388953956874686,
|
|
"loss": 4.9149,
|
|
"mean_token_accuracy": 0.18590909540653228,
|
|
"num_tokens": 9805017.0,
|
|
"step": 4440
|
|
},
|
|
{
|
|
"entropy": 5.478418445587158,
|
|
"epoch": 4.481089258698941,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003535179326612504,
|
|
"loss": 5.028,
|
|
"mean_token_accuracy": 0.18700236678123475,
|
|
"num_tokens": 9816333.0,
|
|
"step": 4445
|
|
},
|
|
{
|
|
"entropy": 5.552411794662476,
|
|
"epoch": 4.48613212304589,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00035314608171820787,
|
|
"loss": 5.1035,
|
|
"mean_token_accuracy": 0.18107048124074937,
|
|
"num_tokens": 9828835.0,
|
|
"step": 4450
|
|
},
|
|
{
|
|
"entropy": 5.484770631790161,
|
|
"epoch": 4.4911749873928395,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00035277398789534075,
|
|
"loss": 5.0053,
|
|
"mean_token_accuracy": 0.19101633429527282,
|
|
"num_tokens": 9839321.0,
|
|
"step": 4455
|
|
},
|
|
{
|
|
"entropy": 5.481580114364624,
|
|
"epoch": 4.496217851739789,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003524016523491252,
|
|
"loss": 4.9446,
|
|
"mean_token_accuracy": 0.1967537522315979,
|
|
"num_tokens": 9849913.0,
|
|
"step": 4460
|
|
},
|
|
{
|
|
"entropy": 5.419422292709351,
|
|
"epoch": 4.501260716086737,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00035202907623678865,
|
|
"loss": 4.9584,
|
|
"mean_token_accuracy": 0.18906479179859162,
|
|
"num_tokens": 9861981.0,
|
|
"step": 4465
|
|
},
|
|
{
|
|
"entropy": 5.490814256668091,
|
|
"epoch": 4.506303580433686,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003516562607163062,
|
|
"loss": 5.0118,
|
|
"mean_token_accuracy": 0.1882714733481407,
|
|
"num_tokens": 9873189.0,
|
|
"step": 4470
|
|
},
|
|
{
|
|
"entropy": 5.501973104476929,
|
|
"epoch": 4.511346444780635,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003512832069463971,
|
|
"loss": 4.9954,
|
|
"mean_token_accuracy": 0.182276713848114,
|
|
"num_tokens": 9884263.0,
|
|
"step": 4475
|
|
},
|
|
{
|
|
"entropy": 5.490594863891602,
|
|
"epoch": 4.516389309127584,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00035090991608652125,
|
|
"loss": 5.0661,
|
|
"mean_token_accuracy": 0.18492345809936522,
|
|
"num_tokens": 9894958.0,
|
|
"step": 4480
|
|
},
|
|
{
|
|
"entropy": 5.430255556106568,
|
|
"epoch": 4.521432173474533,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00035053638929687504,
|
|
"loss": 4.9228,
|
|
"mean_token_accuracy": 0.1863362818956375,
|
|
"num_tokens": 9906846.0,
|
|
"step": 4485
|
|
},
|
|
{
|
|
"entropy": 5.469951725006103,
|
|
"epoch": 4.526475037821482,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003501626277383884,
|
|
"loss": 4.996,
|
|
"mean_token_accuracy": 0.18500665277242662,
|
|
"num_tokens": 9917696.0,
|
|
"step": 4490
|
|
},
|
|
{
|
|
"entropy": 5.479888916015625,
|
|
"epoch": 4.5315179021684315,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000349788632572721,
|
|
"loss": 5.0186,
|
|
"mean_token_accuracy": 0.18784863352775574,
|
|
"num_tokens": 9928907.0,
|
|
"step": 4495
|
|
},
|
|
{
|
|
"entropy": 5.398473262786865,
|
|
"epoch": 4.536560766515381,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003494144049622582,
|
|
"loss": 4.9328,
|
|
"mean_token_accuracy": 0.19004513770341874,
|
|
"num_tokens": 9939239.0,
|
|
"step": 4500
|
|
},
|
|
{
|
|
"epoch": 4.536560766515381,
|
|
"eval_entropy": 5.287213412859968,
|
|
"eval_loss": 5.680631160736084,
|
|
"eval_mean_token_accuracy": 0.15380664844701006,
|
|
"eval_num_tokens": 9939239.0,
|
|
"eval_runtime": 2.0095,
|
|
"eval_samples_per_second": 1561.611,
|
|
"eval_steps_per_second": 195.575,
|
|
"step": 4500
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 9910,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 10,
|
|
"save_steps": 500,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 2455610010071040.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|