7ba63f0c50
Model: fpadovani/ita-latn-10mb-after-ppt-Dp-10mb-ckpt500_seed3407 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": 7.140286159515381,
|
|
"epoch": 0.005042864346949067,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 2e-06,
|
|
"loss": 6.8508,
|
|
"mean_token_accuracy": 0.08954758420586587,
|
|
"num_tokens": 11445.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 7.160407781600952,
|
|
"epoch": 0.010085728693898134,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 6.8381,
|
|
"mean_token_accuracy": 0.09056923538446426,
|
|
"num_tokens": 22095.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 7.163180255889893,
|
|
"epoch": 0.015128593040847202,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 7e-06,
|
|
"loss": 6.8499,
|
|
"mean_token_accuracy": 0.08891279846429825,
|
|
"num_tokens": 33847.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 7.216903209686279,
|
|
"epoch": 0.020171457387796268,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 6.8564,
|
|
"mean_token_accuracy": 0.09377839192748069,
|
|
"num_tokens": 44495.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 7.231993389129639,
|
|
"epoch": 0.025214321734745335,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 6.8481,
|
|
"mean_token_accuracy": 0.0926659069955349,
|
|
"num_tokens": 55298.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 7.239791536331177,
|
|
"epoch": 0.030257186081694403,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 6.8024,
|
|
"mean_token_accuracy": 0.09430743008852005,
|
|
"num_tokens": 66391.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 7.254619121551514,
|
|
"epoch": 0.03530005042864347,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 6.7592,
|
|
"mean_token_accuracy": 0.09942321330308915,
|
|
"num_tokens": 77624.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 7.30181918144226,
|
|
"epoch": 0.040342914775592535,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 6.8325,
|
|
"mean_token_accuracy": 0.09075544625520707,
|
|
"num_tokens": 89426.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 7.293350648880005,
|
|
"epoch": 0.0453857791225416,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 6.8901,
|
|
"mean_token_accuracy": 0.09258394688367844,
|
|
"num_tokens": 100252.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 7.353130722045899,
|
|
"epoch": 0.05042864346949067,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 6.7851,
|
|
"mean_token_accuracy": 0.09203029945492744,
|
|
"num_tokens": 110578.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 7.304513311386108,
|
|
"epoch": 0.05547150781643974,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 6.9065,
|
|
"mean_token_accuracy": 0.09156164526939392,
|
|
"num_tokens": 121631.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 7.341808700561524,
|
|
"epoch": 0.060514372163388806,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 6.9075,
|
|
"mean_token_accuracy": 0.09541143402457238,
|
|
"num_tokens": 134254.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 7.3094686508178714,
|
|
"epoch": 0.06555723651033787,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 6.8583,
|
|
"mean_token_accuracy": 0.08802076503634453,
|
|
"num_tokens": 144928.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 7.32190637588501,
|
|
"epoch": 0.07060010085728693,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 6.7789,
|
|
"mean_token_accuracy": 0.09283096343278885,
|
|
"num_tokens": 155518.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 7.300724887847901,
|
|
"epoch": 0.07564296520423601,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 6.819,
|
|
"mean_token_accuracy": 0.09886608645319939,
|
|
"num_tokens": 165850.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 7.25584135055542,
|
|
"epoch": 0.08068582955118507,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 6.8076,
|
|
"mean_token_accuracy": 0.09424731731414795,
|
|
"num_tokens": 176787.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 7.303829574584961,
|
|
"epoch": 0.08572869389813415,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 6.7915,
|
|
"mean_token_accuracy": 0.0980073258280754,
|
|
"num_tokens": 188345.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 7.327994871139526,
|
|
"epoch": 0.0907715582450832,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 6.8242,
|
|
"mean_token_accuracy": 0.0911541111767292,
|
|
"num_tokens": 199315.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 7.294112253189087,
|
|
"epoch": 0.09581442259203228,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 6.7153,
|
|
"mean_token_accuracy": 0.09620574191212654,
|
|
"num_tokens": 209627.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 7.2472625255584715,
|
|
"epoch": 0.10085728693898134,
|
|
"grad_norm": 0.8203125,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 6.7516,
|
|
"mean_token_accuracy": 0.09449133202433586,
|
|
"num_tokens": 220872.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 7.2352982521057125,
|
|
"epoch": 0.1059001512859304,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 6.7707,
|
|
"mean_token_accuracy": 0.10300672203302383,
|
|
"num_tokens": 233272.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 7.241355466842651,
|
|
"epoch": 0.11094301563287948,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 6.7697,
|
|
"mean_token_accuracy": 0.09401320964097977,
|
|
"num_tokens": 243529.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 7.283736085891723,
|
|
"epoch": 0.11598587997982854,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 6.7884,
|
|
"mean_token_accuracy": 0.09377909824252129,
|
|
"num_tokens": 254942.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 7.26002721786499,
|
|
"epoch": 0.12102874432677761,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 6.7897,
|
|
"mean_token_accuracy": 0.09622539505362511,
|
|
"num_tokens": 266552.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 7.222865438461303,
|
|
"epoch": 0.1260716086737267,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 6.748,
|
|
"mean_token_accuracy": 0.09950896799564361,
|
|
"num_tokens": 277906.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 7.261881589889526,
|
|
"epoch": 0.13111447302067575,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 6.7379,
|
|
"mean_token_accuracy": 0.0997304804623127,
|
|
"num_tokens": 289079.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 7.321910429000854,
|
|
"epoch": 0.1361573373676248,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 6.8065,
|
|
"mean_token_accuracy": 0.09663683921098709,
|
|
"num_tokens": 300107.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 7.179061460494995,
|
|
"epoch": 0.14120020171457387,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 6.7372,
|
|
"mean_token_accuracy": 0.10018184557557105,
|
|
"num_tokens": 310779.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 7.288008165359497,
|
|
"epoch": 0.14624306606152296,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 6.8281,
|
|
"mean_token_accuracy": 0.10092653110623359,
|
|
"num_tokens": 323478.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 7.338018560409546,
|
|
"epoch": 0.15128593040847202,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 6.7599,
|
|
"mean_token_accuracy": 0.1008948564529419,
|
|
"num_tokens": 335255.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 7.256061553955078,
|
|
"epoch": 0.15632879475542108,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 6.7086,
|
|
"mean_token_accuracy": 0.09725899025797843,
|
|
"num_tokens": 344824.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 7.258427906036377,
|
|
"epoch": 0.16137165910237014,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 6.7495,
|
|
"mean_token_accuracy": 0.09951818436384201,
|
|
"num_tokens": 356283.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 7.266161060333252,
|
|
"epoch": 0.1664145234493192,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 6.7592,
|
|
"mean_token_accuracy": 0.09592765346169471,
|
|
"num_tokens": 368098.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 7.238630914688111,
|
|
"epoch": 0.1714573877962683,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 6.7977,
|
|
"mean_token_accuracy": 0.09837751016020775,
|
|
"num_tokens": 380948.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 7.225790119171142,
|
|
"epoch": 0.17650025214321735,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 6.7359,
|
|
"mean_token_accuracy": 0.0987349271774292,
|
|
"num_tokens": 390392.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 7.300117444992066,
|
|
"epoch": 0.1815431164901664,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 6.8281,
|
|
"mean_token_accuracy": 0.10141217559576035,
|
|
"num_tokens": 400328.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 7.300943756103516,
|
|
"epoch": 0.18658598083711547,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 6.8062,
|
|
"mean_token_accuracy": 0.09373326599597931,
|
|
"num_tokens": 410909.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 7.274401330947876,
|
|
"epoch": 0.19162884518406456,
|
|
"grad_norm": 0.7890625,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 6.817,
|
|
"mean_token_accuracy": 0.09608622565865517,
|
|
"num_tokens": 422726.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 7.216585302352906,
|
|
"epoch": 0.19667170953101362,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 6.7439,
|
|
"mean_token_accuracy": 0.10276417583227157,
|
|
"num_tokens": 433376.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 7.360825824737549,
|
|
"epoch": 0.20171457387796268,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 6.7932,
|
|
"mean_token_accuracy": 0.09293813556432724,
|
|
"num_tokens": 443940.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 7.150915622711182,
|
|
"epoch": 0.20675743822491174,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.000102,
|
|
"loss": 6.686,
|
|
"mean_token_accuracy": 0.09526368230581284,
|
|
"num_tokens": 455016.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 7.291905736923217,
|
|
"epoch": 0.2118003025718608,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 6.7718,
|
|
"mean_token_accuracy": 0.10028775706887245,
|
|
"num_tokens": 466380.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 7.096459913253784,
|
|
"epoch": 0.2168431669188099,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000107,
|
|
"loss": 6.6202,
|
|
"mean_token_accuracy": 0.10174242556095123,
|
|
"num_tokens": 476915.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 7.275387096405029,
|
|
"epoch": 0.22188603126575895,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 6.7345,
|
|
"mean_token_accuracy": 0.1033494584262371,
|
|
"num_tokens": 488924.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 7.2009937286376955,
|
|
"epoch": 0.22692889561270801,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000112,
|
|
"loss": 6.6908,
|
|
"mean_token_accuracy": 0.10189548060297966,
|
|
"num_tokens": 498965.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 7.218687295913696,
|
|
"epoch": 0.23197175995965708,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 6.7348,
|
|
"mean_token_accuracy": 0.0991855151951313,
|
|
"num_tokens": 509755.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 7.226504278182984,
|
|
"epoch": 0.23701462430660616,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 6.6762,
|
|
"mean_token_accuracy": 0.0955890141427517,
|
|
"num_tokens": 521221.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 7.219500541687012,
|
|
"epoch": 0.24205748865355523,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 6.8369,
|
|
"mean_token_accuracy": 0.09283191710710526,
|
|
"num_tokens": 533595.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.376226377487183,
|
|
"epoch": 0.24710035300050429,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000122,
|
|
"loss": 6.7942,
|
|
"mean_token_accuracy": 0.09729663655161858,
|
|
"num_tokens": 543672.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.155847024917603,
|
|
"epoch": 0.2521432173474534,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 6.6626,
|
|
"mean_token_accuracy": 0.09860076606273652,
|
|
"num_tokens": 554906.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 7.288216876983642,
|
|
"epoch": 0.25718608169440244,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000127,
|
|
"loss": 6.7774,
|
|
"mean_token_accuracy": 0.09813973531126977,
|
|
"num_tokens": 566419.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.1866679191589355,
|
|
"epoch": 0.2622289460413515,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 6.7309,
|
|
"mean_token_accuracy": 0.09767505675554275,
|
|
"num_tokens": 578848.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.233185148239135,
|
|
"epoch": 0.26727181038830056,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.000132,
|
|
"loss": 6.7021,
|
|
"mean_token_accuracy": 0.10028551891446114,
|
|
"num_tokens": 590601.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.194158220291138,
|
|
"epoch": 0.2723146747352496,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 6.7672,
|
|
"mean_token_accuracy": 0.09758391752839088,
|
|
"num_tokens": 602021.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.340858078002929,
|
|
"epoch": 0.2773575390821987,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 6.8642,
|
|
"mean_token_accuracy": 0.09155096188187599,
|
|
"num_tokens": 613966.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.167615747451782,
|
|
"epoch": 0.28240040342914774,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 6.6849,
|
|
"mean_token_accuracy": 0.0972824864089489,
|
|
"num_tokens": 624851.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.212220478057861,
|
|
"epoch": 0.2874432677760968,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 6.8036,
|
|
"mean_token_accuracy": 0.09324645102024079,
|
|
"num_tokens": 635580.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.35611047744751,
|
|
"epoch": 0.2924861321230459,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 6.8484,
|
|
"mean_token_accuracy": 0.0951171413064003,
|
|
"num_tokens": 646927.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.219331979751587,
|
|
"epoch": 0.297528996469995,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000147,
|
|
"loss": 6.6477,
|
|
"mean_token_accuracy": 0.09922101944684983,
|
|
"num_tokens": 658116.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.274383974075318,
|
|
"epoch": 0.30257186081694404,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 6.6747,
|
|
"mean_token_accuracy": 0.09747742339968682,
|
|
"num_tokens": 669054.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.159107351303101,
|
|
"epoch": 0.3076147251638931,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000152,
|
|
"loss": 6.7491,
|
|
"mean_token_accuracy": 0.10271781459450721,
|
|
"num_tokens": 680439.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.223165512084961,
|
|
"epoch": 0.31265758951084216,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 6.7262,
|
|
"mean_token_accuracy": 0.10282548666000366,
|
|
"num_tokens": 690909.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.201652526855469,
|
|
"epoch": 0.3177004538577912,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000157,
|
|
"loss": 6.7207,
|
|
"mean_token_accuracy": 0.10077903121709823,
|
|
"num_tokens": 701744.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.211695003509521,
|
|
"epoch": 0.3227433182047403,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 6.7612,
|
|
"mean_token_accuracy": 0.09715296998620034,
|
|
"num_tokens": 712268.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.288143825531006,
|
|
"epoch": 0.32778618255168934,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000162,
|
|
"loss": 6.7867,
|
|
"mean_token_accuracy": 0.09628506749868393,
|
|
"num_tokens": 722978.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.225297594070435,
|
|
"epoch": 0.3328290468986384,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 6.7812,
|
|
"mean_token_accuracy": 0.0981549084186554,
|
|
"num_tokens": 734278.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.247027444839477,
|
|
"epoch": 0.3378719112455875,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 6.6968,
|
|
"mean_token_accuracy": 0.09765793681144715,
|
|
"num_tokens": 746109.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.143796396255493,
|
|
"epoch": 0.3429147755925366,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 6.63,
|
|
"mean_token_accuracy": 0.09958403259515762,
|
|
"num_tokens": 757519.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.164006567001342,
|
|
"epoch": 0.34795763993948564,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 6.6246,
|
|
"mean_token_accuracy": 0.10218961685895919,
|
|
"num_tokens": 767665.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.145311164855957,
|
|
"epoch": 0.3530005042864347,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 6.6681,
|
|
"mean_token_accuracy": 0.10835013091564179,
|
|
"num_tokens": 778880.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.140744829177857,
|
|
"epoch": 0.35804336863338376,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.000177,
|
|
"loss": 6.6781,
|
|
"mean_token_accuracy": 0.10356075391173362,
|
|
"num_tokens": 790360.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.131025409698486,
|
|
"epoch": 0.3630862329803328,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 6.6595,
|
|
"mean_token_accuracy": 0.10246514827013016,
|
|
"num_tokens": 800889.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.184374713897705,
|
|
"epoch": 0.3681290973272819,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000182,
|
|
"loss": 6.6523,
|
|
"mean_token_accuracy": 0.09842721745371819,
|
|
"num_tokens": 811414.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.105099868774414,
|
|
"epoch": 0.37317196167423095,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 6.6653,
|
|
"mean_token_accuracy": 0.09902059137821198,
|
|
"num_tokens": 822822.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.043937063217163,
|
|
"epoch": 0.37821482602118,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000187,
|
|
"loss": 6.5729,
|
|
"mean_token_accuracy": 0.10602593421936035,
|
|
"num_tokens": 833145.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.152944040298462,
|
|
"epoch": 0.3832576903681291,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 6.6975,
|
|
"mean_token_accuracy": 0.10423027276992798,
|
|
"num_tokens": 844015.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.114481735229492,
|
|
"epoch": 0.3883005547150782,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000192,
|
|
"loss": 6.6216,
|
|
"mean_token_accuracy": 0.10629610195755959,
|
|
"num_tokens": 855806.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.063960886001587,
|
|
"epoch": 0.39334341906202724,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 6.5439,
|
|
"mean_token_accuracy": 0.1060337856411934,
|
|
"num_tokens": 865750.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.154440927505493,
|
|
"epoch": 0.3983862834089763,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 6.6872,
|
|
"mean_token_accuracy": 0.10153400599956512,
|
|
"num_tokens": 877630.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.007752466201782,
|
|
"epoch": 0.40342914775592537,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 6.5811,
|
|
"mean_token_accuracy": 0.10793254896998405,
|
|
"num_tokens": 888667.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.195240163803101,
|
|
"epoch": 0.4084720121028744,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000202,
|
|
"loss": 6.6621,
|
|
"mean_token_accuracy": 0.09717055335640908,
|
|
"num_tokens": 900989.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.0393874168396,
|
|
"epoch": 0.4135148764498235,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 6.5594,
|
|
"mean_token_accuracy": 0.09989590123295784,
|
|
"num_tokens": 911479.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.06795744895935,
|
|
"epoch": 0.41855774079677255,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000207,
|
|
"loss": 6.5759,
|
|
"mean_token_accuracy": 0.10631833747029304,
|
|
"num_tokens": 922434.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.052112197875976,
|
|
"epoch": 0.4236006051437216,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 6.5854,
|
|
"mean_token_accuracy": 0.10236857533454895,
|
|
"num_tokens": 933405.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.114569473266601,
|
|
"epoch": 0.4286434694906707,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000212,
|
|
"loss": 6.5826,
|
|
"mean_token_accuracy": 0.10237419381737708,
|
|
"num_tokens": 943567.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 6.984195518493652,
|
|
"epoch": 0.4336863338376198,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 6.6261,
|
|
"mean_token_accuracy": 0.09788121804594993,
|
|
"num_tokens": 955123.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.152405214309693,
|
|
"epoch": 0.43872919818456885,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 6.6689,
|
|
"mean_token_accuracy": 0.09843425154685974,
|
|
"num_tokens": 965565.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.103341150283813,
|
|
"epoch": 0.4437720625315179,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 6.5858,
|
|
"mean_token_accuracy": 0.10635014995932579,
|
|
"num_tokens": 977101.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.027708673477173,
|
|
"epoch": 0.44881492687846697,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000222,
|
|
"loss": 6.6377,
|
|
"mean_token_accuracy": 0.10166379287838936,
|
|
"num_tokens": 987445.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.0129906177520756,
|
|
"epoch": 0.45385779122541603,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 6.5091,
|
|
"mean_token_accuracy": 0.10711757391691208,
|
|
"num_tokens": 999379.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 6.951913785934448,
|
|
"epoch": 0.4589006555723651,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 6.4679,
|
|
"mean_token_accuracy": 0.10711723566055298,
|
|
"num_tokens": 1010744.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.059710884094239,
|
|
"epoch": 0.46394351991931415,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 6.5676,
|
|
"mean_token_accuracy": 0.10420158058404923,
|
|
"num_tokens": 1021808.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.073292541503906,
|
|
"epoch": 0.4689863842662632,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 6.6029,
|
|
"mean_token_accuracy": 0.10069075301289558,
|
|
"num_tokens": 1033694.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 6.95826940536499,
|
|
"epoch": 0.47402924861321233,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.5898,
|
|
"mean_token_accuracy": 0.1029825672507286,
|
|
"num_tokens": 1044690.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.098178434371948,
|
|
"epoch": 0.4790721129601614,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.5965,
|
|
"mean_token_accuracy": 0.10506486594676971,
|
|
"num_tokens": 1055557.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 6.981123685836792,
|
|
"epoch": 0.48411497730711045,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.5391,
|
|
"mean_token_accuracy": 0.10544369742274284,
|
|
"num_tokens": 1067304.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.022891092300415,
|
|
"epoch": 0.4891578416540595,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.5635,
|
|
"mean_token_accuracy": 0.10295396745204925,
|
|
"num_tokens": 1078955.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 6.96609411239624,
|
|
"epoch": 0.49420070600100857,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.6139,
|
|
"mean_token_accuracy": 0.1050510697066784,
|
|
"num_tokens": 1091405.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.0190895080566404,
|
|
"epoch": 0.49924357034795763,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.5892,
|
|
"mean_token_accuracy": 0.10599951222538948,
|
|
"num_tokens": 1103227.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.028829288482666,
|
|
"epoch": 0.5042864346949067,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 6.5202,
|
|
"mean_token_accuracy": 0.10993084460496902,
|
|
"num_tokens": 1114485.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 0.5042864346949067,
|
|
"eval_entropy": 6.8225840927687,
|
|
"eval_loss": 6.599184989929199,
|
|
"eval_mean_token_accuracy": 0.10669155084260244,
|
|
"eval_num_tokens": 1114485.0,
|
|
"eval_runtime": 2.131,
|
|
"eval_samples_per_second": 1472.565,
|
|
"eval_steps_per_second": 184.423,
|
|
"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": 276978743377920.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|