024d74a8bd
Model: fpadovani/ita-latn-10mb-after-ppt-Dp-10mb-ckpt500_seed10 Source: Original Platform
3068 lines
82 KiB
JSON
3068 lines
82 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 1.5123550176500253,
|
|
"eval_steps": 500,
|
|
"global_step": 1500,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 7.2617237091064455,
|
|
"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
|
|
}
|
|
],
|
|
"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": 817034222684160.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|