024d74a8bd
Model: fpadovani/ita-latn-10mb-after-ppt-Dp-10mb-ckpt500_seed10 Source: Original Platform
1046 lines
28 KiB
JSON
1046 lines
28 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.5042864346949067,
|
|
"eval_steps": 500,
|
|
"global_step": 500,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 7.2617237091064455,
|
|
"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
|
|
}
|
|
],
|
|
"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": 267475425361920.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|