7ba63f0c50
Model: fpadovani/ita-latn-10mb-after-ppt-Dp-10mb-ckpt500_seed3407 Source: Original Platform
4079 lines
110 KiB
JSON
4079 lines
110 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 2.016137165910237,
|
|
"eval_steps": 500,
|
|
"global_step": 2000,
|
|
"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
|
|
},
|
|
{
|
|
"entropy": 6.961726331710816,
|
|
"epoch": 0.5093292990418558,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.5673,
|
|
"mean_token_accuracy": 0.10170895382761955,
|
|
"num_tokens": 1126789.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 6.986708450317383,
|
|
"epoch": 0.5143721633888049,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.5409,
|
|
"mean_token_accuracy": 0.10625190362334251,
|
|
"num_tokens": 1138192.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 6.920113372802734,
|
|
"epoch": 0.5194150277357539,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.4838,
|
|
"mean_token_accuracy": 0.10268188640475273,
|
|
"num_tokens": 1148526.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.071941947937011,
|
|
"epoch": 0.524457892082703,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.5906,
|
|
"mean_token_accuracy": 0.09835154190659523,
|
|
"num_tokens": 1159194.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 6.885069465637207,
|
|
"epoch": 0.529500756429652,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.4986,
|
|
"mean_token_accuracy": 0.10742172673344612,
|
|
"num_tokens": 1170601.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 7.027553844451904,
|
|
"epoch": 0.5345436207766011,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.5016,
|
|
"mean_token_accuracy": 0.10665301531553269,
|
|
"num_tokens": 1182441.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 6.889227390289307,
|
|
"epoch": 0.5395864851235502,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.4023,
|
|
"mean_token_accuracy": 0.11396976485848427,
|
|
"num_tokens": 1192530.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 6.831969308853149,
|
|
"epoch": 0.5446293494704992,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.5142,
|
|
"mean_token_accuracy": 0.10788194462656975,
|
|
"num_tokens": 1203831.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 6.912649297714234,
|
|
"epoch": 0.5496722138174484,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.4504,
|
|
"mean_token_accuracy": 0.10705123022198677,
|
|
"num_tokens": 1214164.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 6.927637386322021,
|
|
"epoch": 0.5547150781643974,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.4916,
|
|
"mean_token_accuracy": 0.10291663855314255,
|
|
"num_tokens": 1226932.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 7.036397933959961,
|
|
"epoch": 0.5597579425113465,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.5515,
|
|
"mean_token_accuracy": 0.1012086533010006,
|
|
"num_tokens": 1237212.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 6.870266580581665,
|
|
"epoch": 0.5648008068582955,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.473,
|
|
"mean_token_accuracy": 0.10597705841064453,
|
|
"num_tokens": 1249340.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 6.893102836608887,
|
|
"epoch": 0.5698436712052446,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.4801,
|
|
"mean_token_accuracy": 0.10842250436544418,
|
|
"num_tokens": 1260778.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 6.866225051879883,
|
|
"epoch": 0.5748865355521936,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.4787,
|
|
"mean_token_accuracy": 0.1063090831041336,
|
|
"num_tokens": 1272216.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 6.98068904876709,
|
|
"epoch": 0.5799293998991427,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.5465,
|
|
"mean_token_accuracy": 0.1084217220544815,
|
|
"num_tokens": 1283558.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 6.992026376724243,
|
|
"epoch": 0.5849722642460918,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.6296,
|
|
"mean_token_accuracy": 0.09711046032607555,
|
|
"num_tokens": 1294638.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 7.050863647460938,
|
|
"epoch": 0.5900151285930408,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.526,
|
|
"mean_token_accuracy": 0.10724521428346634,
|
|
"num_tokens": 1304697.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 6.845832347869873,
|
|
"epoch": 0.59505799293999,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.4895,
|
|
"mean_token_accuracy": 0.10360805690288544,
|
|
"num_tokens": 1316618.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 6.966832494735717,
|
|
"epoch": 0.600100857286939,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.6021,
|
|
"mean_token_accuracy": 0.09591096639633179,
|
|
"num_tokens": 1328134.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 6.980519390106201,
|
|
"epoch": 0.6051437216338881,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.5161,
|
|
"mean_token_accuracy": 0.10199217274785041,
|
|
"num_tokens": 1339101.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 6.954841566085816,
|
|
"epoch": 0.6101865859808371,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.3913,
|
|
"mean_token_accuracy": 0.11074580848217011,
|
|
"num_tokens": 1349181.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 6.819632625579834,
|
|
"epoch": 0.6152294503277862,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.5304,
|
|
"mean_token_accuracy": 0.10506919622421265,
|
|
"num_tokens": 1361443.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 6.891585016250611,
|
|
"epoch": 0.6202723146747352,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.4873,
|
|
"mean_token_accuracy": 0.11039864644408226,
|
|
"num_tokens": 1372807.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 6.961631441116333,
|
|
"epoch": 0.6253151790216843,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.5536,
|
|
"mean_token_accuracy": 0.10718053579330444,
|
|
"num_tokens": 1385509.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 6.873125696182251,
|
|
"epoch": 0.6303580433686334,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.4554,
|
|
"mean_token_accuracy": 0.10879878252744675,
|
|
"num_tokens": 1396035.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 6.814496755599976,
|
|
"epoch": 0.6354009077155824,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.4004,
|
|
"mean_token_accuracy": 0.11274328380823136,
|
|
"num_tokens": 1406839.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 6.782590436935425,
|
|
"epoch": 0.6404437720625316,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.3721,
|
|
"mean_token_accuracy": 0.10830515772104263,
|
|
"num_tokens": 1417708.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 6.925254392623901,
|
|
"epoch": 0.6454866364094806,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.5092,
|
|
"mean_token_accuracy": 0.10417326539754868,
|
|
"num_tokens": 1428985.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 6.864048671722412,
|
|
"epoch": 0.6505295007564297,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.5278,
|
|
"mean_token_accuracy": 0.1116502083837986,
|
|
"num_tokens": 1440232.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 6.749512767791748,
|
|
"epoch": 0.6555723651033787,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.2722,
|
|
"mean_token_accuracy": 0.11440228670835495,
|
|
"num_tokens": 1451735.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 6.767019939422608,
|
|
"epoch": 0.6606152294503278,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.4566,
|
|
"mean_token_accuracy": 0.1066243976354599,
|
|
"num_tokens": 1463477.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 6.886978673934936,
|
|
"epoch": 0.6656580937972768,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.4628,
|
|
"mean_token_accuracy": 0.10322247892618179,
|
|
"num_tokens": 1473802.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 6.786734676361084,
|
|
"epoch": 0.6707009581442259,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.4699,
|
|
"mean_token_accuracy": 0.10582903772592545,
|
|
"num_tokens": 1485519.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 6.832977151870727,
|
|
"epoch": 0.675743822491175,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.4649,
|
|
"mean_token_accuracy": 0.1091465823352337,
|
|
"num_tokens": 1496957.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 6.882336664199829,
|
|
"epoch": 0.680786686838124,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.4699,
|
|
"mean_token_accuracy": 0.11220887899398804,
|
|
"num_tokens": 1508336.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 6.780294466018677,
|
|
"epoch": 0.6858295511850732,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.4558,
|
|
"mean_token_accuracy": 0.10774369463324547,
|
|
"num_tokens": 1520016.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 6.71516900062561,
|
|
"epoch": 0.6908724155320222,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.2974,
|
|
"mean_token_accuracy": 0.11265372037887574,
|
|
"num_tokens": 1531877.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 6.664650774002075,
|
|
"epoch": 0.6959152798789713,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.295,
|
|
"mean_token_accuracy": 0.11770892515778542,
|
|
"num_tokens": 1541553.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 6.810173177719117,
|
|
"epoch": 0.7009581442259203,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.4674,
|
|
"mean_token_accuracy": 0.11161921173334122,
|
|
"num_tokens": 1552491.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 6.704865646362305,
|
|
"epoch": 0.7060010085728694,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.368,
|
|
"mean_token_accuracy": 0.11017820984125137,
|
|
"num_tokens": 1563695.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 6.858102321624756,
|
|
"epoch": 0.7110438729198184,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.3911,
|
|
"mean_token_accuracy": 0.1099718302488327,
|
|
"num_tokens": 1573467.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 6.678512668609619,
|
|
"epoch": 0.7160867372667675,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.3853,
|
|
"mean_token_accuracy": 0.1186108984053135,
|
|
"num_tokens": 1584367.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 6.745659399032593,
|
|
"epoch": 0.7211296016137166,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.3118,
|
|
"mean_token_accuracy": 0.11698892638087273,
|
|
"num_tokens": 1595301.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 6.733723211288452,
|
|
"epoch": 0.7261724659606656,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.3898,
|
|
"mean_token_accuracy": 0.10888881608843803,
|
|
"num_tokens": 1605916.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 6.798846769332886,
|
|
"epoch": 0.7312153303076148,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.3867,
|
|
"mean_token_accuracy": 0.11005102321505547,
|
|
"num_tokens": 1616455.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 6.668491792678833,
|
|
"epoch": 0.7362581946545638,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.3685,
|
|
"mean_token_accuracy": 0.11659844517707825,
|
|
"num_tokens": 1627396.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 6.883508205413818,
|
|
"epoch": 0.7413010590015129,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.5013,
|
|
"mean_token_accuracy": 0.1054186426103115,
|
|
"num_tokens": 1639073.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.682006788253784,
|
|
"epoch": 0.7463439233484619,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.3675,
|
|
"mean_token_accuracy": 0.11053554564714432,
|
|
"num_tokens": 1649787.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 6.725868129730225,
|
|
"epoch": 0.751386787695411,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.3089,
|
|
"mean_token_accuracy": 0.10525414422154426,
|
|
"num_tokens": 1658883.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.616196966171264,
|
|
"epoch": 0.75642965204236,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.2851,
|
|
"mean_token_accuracy": 0.11788807809352875,
|
|
"num_tokens": 1669099.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 6.656661128997802,
|
|
"epoch": 0.7614725163893091,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.2785,
|
|
"mean_token_accuracy": 0.11349497437477112,
|
|
"num_tokens": 1680227.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 6.659192895889282,
|
|
"epoch": 0.7665153807362582,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.2911,
|
|
"mean_token_accuracy": 0.11143397390842438,
|
|
"num_tokens": 1689901.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.674802637100219,
|
|
"epoch": 0.7715582450832073,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.3671,
|
|
"mean_token_accuracy": 0.11457237899303437,
|
|
"num_tokens": 1700498.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.6554454326629635,
|
|
"epoch": 0.7766011094301564,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.2704,
|
|
"mean_token_accuracy": 0.11751708760857582,
|
|
"num_tokens": 1710715.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.724909543991089,
|
|
"epoch": 0.7816439737771054,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.3937,
|
|
"mean_token_accuracy": 0.11218245178461075,
|
|
"num_tokens": 1723858.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.652149772644043,
|
|
"epoch": 0.7866868381240545,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.3675,
|
|
"mean_token_accuracy": 0.10926554128527641,
|
|
"num_tokens": 1734678.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.643021297454834,
|
|
"epoch": 0.7917297024710035,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.2785,
|
|
"mean_token_accuracy": 0.11664367765188217,
|
|
"num_tokens": 1743869.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.746792840957641,
|
|
"epoch": 0.7967725668179526,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.3355,
|
|
"mean_token_accuracy": 0.11328683719038964,
|
|
"num_tokens": 1755266.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.558908605575562,
|
|
"epoch": 0.8018154311649016,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.3026,
|
|
"mean_token_accuracy": 0.11433365866541863,
|
|
"num_tokens": 1766089.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.660847854614258,
|
|
"epoch": 0.8068582955118507,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.3085,
|
|
"mean_token_accuracy": 0.11278970018029214,
|
|
"num_tokens": 1776057.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.678964233398437,
|
|
"epoch": 0.8119011598587998,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.3612,
|
|
"mean_token_accuracy": 0.10903131514787674,
|
|
"num_tokens": 1786729.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.682238483428955,
|
|
"epoch": 0.8169440242057489,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.2865,
|
|
"mean_token_accuracy": 0.11454379260540008,
|
|
"num_tokens": 1797708.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.576850557327271,
|
|
"epoch": 0.821986888552698,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.3184,
|
|
"mean_token_accuracy": 0.12089119926095009,
|
|
"num_tokens": 1807326.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.605403757095337,
|
|
"epoch": 0.827029752899647,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.2645,
|
|
"mean_token_accuracy": 0.11531168594956398,
|
|
"num_tokens": 1817266.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.650342416763306,
|
|
"epoch": 0.8320726172465961,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.283,
|
|
"mean_token_accuracy": 0.12210415899753571,
|
|
"num_tokens": 1827840.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.4851631164550785,
|
|
"epoch": 0.8371154815935451,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.2653,
|
|
"mean_token_accuracy": 0.11239074468612671,
|
|
"num_tokens": 1838158.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.684498739242554,
|
|
"epoch": 0.8421583459404942,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.3504,
|
|
"mean_token_accuracy": 0.1102371446788311,
|
|
"num_tokens": 1848317.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.569194793701172,
|
|
"epoch": 0.8472012102874432,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.277,
|
|
"mean_token_accuracy": 0.11234809830784798,
|
|
"num_tokens": 1858751.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.672533845901489,
|
|
"epoch": 0.8522440746343923,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.2743,
|
|
"mean_token_accuracy": 0.1156908892095089,
|
|
"num_tokens": 1870739.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.670135545730591,
|
|
"epoch": 0.8572869389813415,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.407,
|
|
"mean_token_accuracy": 0.1146222934126854,
|
|
"num_tokens": 1880995.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.589824962615967,
|
|
"epoch": 0.8623298033282905,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.353,
|
|
"mean_token_accuracy": 0.11284212246537209,
|
|
"num_tokens": 1892633.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.740231466293335,
|
|
"epoch": 0.8673726676752396,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.4102,
|
|
"mean_token_accuracy": 0.10714741796255112,
|
|
"num_tokens": 1904802.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.678654336929322,
|
|
"epoch": 0.8724155320221886,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.4171,
|
|
"mean_token_accuracy": 0.11096260994672776,
|
|
"num_tokens": 1915755.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.564265584945678,
|
|
"epoch": 0.8774583963691377,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.1608,
|
|
"mean_token_accuracy": 0.11736172437667847,
|
|
"num_tokens": 1926098.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.612742137908936,
|
|
"epoch": 0.8825012607160867,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.386,
|
|
"mean_token_accuracy": 0.11219031587243081,
|
|
"num_tokens": 1936607.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.597411489486694,
|
|
"epoch": 0.8875441250630358,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.2794,
|
|
"mean_token_accuracy": 0.1165696531534195,
|
|
"num_tokens": 1948596.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.572381401062012,
|
|
"epoch": 0.8925869894099848,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.346,
|
|
"mean_token_accuracy": 0.11880208030343056,
|
|
"num_tokens": 1960322.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.577315855026245,
|
|
"epoch": 0.8976298537569339,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.2788,
|
|
"mean_token_accuracy": 0.11647847816348075,
|
|
"num_tokens": 1971052.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.658908319473267,
|
|
"epoch": 0.902672718103883,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.289,
|
|
"mean_token_accuracy": 0.11131321266293526,
|
|
"num_tokens": 1980917.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.534201812744141,
|
|
"epoch": 0.9077155824508321,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.2499,
|
|
"mean_token_accuracy": 0.11574188694357872,
|
|
"num_tokens": 1991035.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.536944723129272,
|
|
"epoch": 0.9127584467977812,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.2626,
|
|
"mean_token_accuracy": 0.11412332057952881,
|
|
"num_tokens": 2000587.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.589452266693115,
|
|
"epoch": 0.9178013111447302,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.2204,
|
|
"mean_token_accuracy": 0.12138074561953545,
|
|
"num_tokens": 2011087.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.517997121810913,
|
|
"epoch": 0.9228441754916793,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.2726,
|
|
"mean_token_accuracy": 0.11250934302806855,
|
|
"num_tokens": 2023472.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.592492294311524,
|
|
"epoch": 0.9278870398386283,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.2621,
|
|
"mean_token_accuracy": 0.11691212207078934,
|
|
"num_tokens": 2033783.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.471665382385254,
|
|
"epoch": 0.9329299041855774,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.2791,
|
|
"mean_token_accuracy": 0.11752638220787048,
|
|
"num_tokens": 2045882.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.6013233184814455,
|
|
"epoch": 0.9379727685325264,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.2954,
|
|
"mean_token_accuracy": 0.11021574363112449,
|
|
"num_tokens": 2056549.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.6011061668396,
|
|
"epoch": 0.9430156328794755,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.3005,
|
|
"mean_token_accuracy": 0.1199602648615837,
|
|
"num_tokens": 2067540.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.461949396133423,
|
|
"epoch": 0.9480584972264247,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.1988,
|
|
"mean_token_accuracy": 0.11865604519844056,
|
|
"num_tokens": 2077386.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.524080276489258,
|
|
"epoch": 0.9531013615733737,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.1647,
|
|
"mean_token_accuracy": 0.12025642693042755,
|
|
"num_tokens": 2087777.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.380746269226075,
|
|
"epoch": 0.9581442259203228,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.1626,
|
|
"mean_token_accuracy": 0.11587570384144782,
|
|
"num_tokens": 2099071.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.422477722167969,
|
|
"epoch": 0.9631870902672718,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.2323,
|
|
"mean_token_accuracy": 0.11468488499522209,
|
|
"num_tokens": 2109168.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.764794063568115,
|
|
"epoch": 0.9682299546142209,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.333,
|
|
"mean_token_accuracy": 0.11822190582752228,
|
|
"num_tokens": 2120696.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.390703725814819,
|
|
"epoch": 0.9732728189611699,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.2922,
|
|
"mean_token_accuracy": 0.11532932966947555,
|
|
"num_tokens": 2132427.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.523982810974121,
|
|
"epoch": 0.978315683308119,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.2758,
|
|
"mean_token_accuracy": 0.11771768406033516,
|
|
"num_tokens": 2143120.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.630441951751709,
|
|
"epoch": 0.983358547655068,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.2674,
|
|
"mean_token_accuracy": 0.1124331571161747,
|
|
"num_tokens": 2154154.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.468868446350098,
|
|
"epoch": 0.9884014120020171,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.3614,
|
|
"mean_token_accuracy": 0.11102845743298531,
|
|
"num_tokens": 2167000.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.588601922988891,
|
|
"epoch": 0.9934442763489663,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.2969,
|
|
"mean_token_accuracy": 0.11907721534371377,
|
|
"num_tokens": 2178100.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.4566107273101805,
|
|
"epoch": 0.9984871406959153,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.2409,
|
|
"mean_token_accuracy": 0.11127112805843353,
|
|
"num_tokens": 2189063.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.666820261213514,
|
|
"epoch": 1.0030257186081695,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.1897,
|
|
"mean_token_accuracy": 0.11703457931677501,
|
|
"num_tokens": 2198302.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.405033683776855,
|
|
"epoch": 1.0080685829551186,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.1489,
|
|
"mean_token_accuracy": 0.12198151051998138,
|
|
"num_tokens": 2209648.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 1.0080685829551186,
|
|
"eval_entropy": 6.3889404944791135,
|
|
"eval_loss": 6.29405403137207,
|
|
"eval_mean_token_accuracy": 0.11920730208470924,
|
|
"eval_num_tokens": 2209648.0,
|
|
"eval_runtime": 2.0392,
|
|
"eval_samples_per_second": 1538.862,
|
|
"eval_steps_per_second": 192.726,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.4559520244598385,
|
|
"epoch": 1.0131114473020675,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999997762223182,
|
|
"loss": 6.0731,
|
|
"mean_token_accuracy": 0.12329766899347305,
|
|
"num_tokens": 2220806.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.391281938552856,
|
|
"epoch": 1.0181543116490166,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999988671262489,
|
|
"loss": 6.0518,
|
|
"mean_token_accuracy": 0.12401486709713935,
|
|
"num_tokens": 2231806.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.45139217376709,
|
|
"epoch": 1.0231971759959657,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999972587285105,
|
|
"loss": 6.1204,
|
|
"mean_token_accuracy": 0.13016737401485443,
|
|
"num_tokens": 2243835.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.346062660217285,
|
|
"epoch": 1.0282400403429148,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999949510341015,
|
|
"loss": 6.0243,
|
|
"mean_token_accuracy": 0.1229862093925476,
|
|
"num_tokens": 2254722.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.361986684799194,
|
|
"epoch": 1.033282904689864,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999919440501947,
|
|
"loss": 5.9939,
|
|
"mean_token_accuracy": 0.1316454030573368,
|
|
"num_tokens": 2264806.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.5396524429321286,
|
|
"epoch": 1.0383257690368128,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004999882377861355,
|
|
"loss": 6.1738,
|
|
"mean_token_accuracy": 0.12066846787929535,
|
|
"num_tokens": 2277037.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.373451137542725,
|
|
"epoch": 1.043368633383762,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004999838322534433,
|
|
"loss": 6.1222,
|
|
"mean_token_accuracy": 0.11222040578722954,
|
|
"num_tokens": 2289829.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.477676343917847,
|
|
"epoch": 1.048411497730711,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004999787274658107,
|
|
"loss": 6.0197,
|
|
"mean_token_accuracy": 0.12076854482293128,
|
|
"num_tokens": 2300340.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.352199077606201,
|
|
"epoch": 1.0534543620776602,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999729234391032,
|
|
"loss": 6.0411,
|
|
"mean_token_accuracy": 0.128338123857975,
|
|
"num_tokens": 2310009.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.3622044086456295,
|
|
"epoch": 1.058497226424609,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00049996642019136,
|
|
"loss": 6.0789,
|
|
"mean_token_accuracy": 0.12038798034191131,
|
|
"num_tokens": 2319290.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.475778961181641,
|
|
"epoch": 1.0635400907715582,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004999592177427933,
|
|
"loss": 6.0314,
|
|
"mean_token_accuracy": 0.12897789776325225,
|
|
"num_tokens": 2329170.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.307097339630127,
|
|
"epoch": 1.0685829551185073,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004999513161157885,
|
|
"loss": 6.0295,
|
|
"mean_token_accuracy": 0.12456534877419471,
|
|
"num_tokens": 2339617.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.410826921463013,
|
|
"epoch": 1.0736258194654564,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999427153349041,
|
|
"loss": 6.0221,
|
|
"mean_token_accuracy": 0.131358290463686,
|
|
"num_tokens": 2349548.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.284106302261352,
|
|
"epoch": 1.0786686838124055,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999334154268714,
|
|
"loss": 6.0214,
|
|
"mean_token_accuracy": 0.1201727457344532,
|
|
"num_tokens": 2359116.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.455397367477417,
|
|
"epoch": 1.0837115481593544,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004999234164205949,
|
|
"loss": 6.077,
|
|
"mean_token_accuracy": 0.128146930038929,
|
|
"num_tokens": 2369412.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.31950888633728,
|
|
"epoch": 1.0887544125063036,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004999127183471517,
|
|
"loss": 6.1372,
|
|
"mean_token_accuracy": 0.12115996405482292,
|
|
"num_tokens": 2381276.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.459614086151123,
|
|
"epoch": 1.0937972768532527,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004999013212397915,
|
|
"loss": 6.0449,
|
|
"mean_token_accuracy": 0.12767613232135772,
|
|
"num_tokens": 2392135.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.249569988250732,
|
|
"epoch": 1.0988401412002018,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004998892251339368,
|
|
"loss": 5.9614,
|
|
"mean_token_accuracy": 0.13299826681613922,
|
|
"num_tokens": 2402957.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.355391168594361,
|
|
"epoch": 1.1038830055471507,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004998764300671827,
|
|
"loss": 6.0912,
|
|
"mean_token_accuracy": 0.1270129896700382,
|
|
"num_tokens": 2413386.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.3538957118988035,
|
|
"epoch": 1.1089258698940998,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004998629360792965,
|
|
"loss": 6.0877,
|
|
"mean_token_accuracy": 0.12099160179495812,
|
|
"num_tokens": 2424441.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.348247289657593,
|
|
"epoch": 1.113968734241049,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004998487432122179,
|
|
"loss": 6.0927,
|
|
"mean_token_accuracy": 0.1191978357732296,
|
|
"num_tokens": 2436144.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.384111166000366,
|
|
"epoch": 1.119011598587998,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004998338515100585,
|
|
"loss": 6.018,
|
|
"mean_token_accuracy": 0.13143668174743653,
|
|
"num_tokens": 2447393.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.262911605834961,
|
|
"epoch": 1.1240544629349472,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004998182610191022,
|
|
"loss": 5.9724,
|
|
"mean_token_accuracy": 0.12895440310239792,
|
|
"num_tokens": 2458632.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.391802740097046,
|
|
"epoch": 1.129097327281896,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004998019717878047,
|
|
"loss": 6.0638,
|
|
"mean_token_accuracy": 0.12387155443429947,
|
|
"num_tokens": 2469331.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.402612590789795,
|
|
"epoch": 1.1341401916288452,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000499784983866793,
|
|
"loss": 6.104,
|
|
"mean_token_accuracy": 0.12408280819654464,
|
|
"num_tokens": 2479501.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.263928079605103,
|
|
"epoch": 1.1391830559757943,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004997672973088661,
|
|
"loss": 5.9657,
|
|
"mean_token_accuracy": 0.12689762637019159,
|
|
"num_tokens": 2490087.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.365678787231445,
|
|
"epoch": 1.1442259203227434,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004997489121689944,
|
|
"loss": 6.0444,
|
|
"mean_token_accuracy": 0.12256940007209778,
|
|
"num_tokens": 2501130.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.414411735534668,
|
|
"epoch": 1.1492687846696923,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004997298285043189,
|
|
"loss": 6.1479,
|
|
"mean_token_accuracy": 0.11699996069073677,
|
|
"num_tokens": 2513999.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.347586917877197,
|
|
"epoch": 1.1543116490166414,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004997100463741527,
|
|
"loss": 6.0692,
|
|
"mean_token_accuracy": 0.12473845109343529,
|
|
"num_tokens": 2525579.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.3394937992095945,
|
|
"epoch": 1.1593545133635905,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004996895658399785,
|
|
"loss": 6.0364,
|
|
"mean_token_accuracy": 0.12666600346565246,
|
|
"num_tokens": 2536091.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.336071920394898,
|
|
"epoch": 1.1643973777105396,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004996683869654507,
|
|
"loss": 6.0728,
|
|
"mean_token_accuracy": 0.12960560843348504,
|
|
"num_tokens": 2546073.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.374319171905517,
|
|
"epoch": 1.1694402420574885,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004996465098163934,
|
|
"loss": 6.0249,
|
|
"mean_token_accuracy": 0.12398967444896698,
|
|
"num_tokens": 2556626.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.286091995239258,
|
|
"epoch": 1.1744831064044376,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004996239344608015,
|
|
"loss": 6.0368,
|
|
"mean_token_accuracy": 0.130549568682909,
|
|
"num_tokens": 2567304.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.302768659591675,
|
|
"epoch": 1.1795259707513868,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004996006609688398,
|
|
"loss": 5.9231,
|
|
"mean_token_accuracy": 0.12457310259342194,
|
|
"num_tokens": 2577242.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.277439069747925,
|
|
"epoch": 1.1845688350983359,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004995766894128426,
|
|
"loss": 6.0239,
|
|
"mean_token_accuracy": 0.11836419180035591,
|
|
"num_tokens": 2588330.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.297782897949219,
|
|
"epoch": 1.189611699445285,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004995520198673143,
|
|
"loss": 6.0412,
|
|
"mean_token_accuracy": 0.12364854291081429,
|
|
"num_tokens": 2598471.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.36807279586792,
|
|
"epoch": 1.194654563792234,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004995266524089283,
|
|
"loss": 6.1099,
|
|
"mean_token_accuracy": 0.12131383642554283,
|
|
"num_tokens": 2610238.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.306796598434448,
|
|
"epoch": 1.199697428139183,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004995005871165271,
|
|
"loss": 5.9793,
|
|
"mean_token_accuracy": 0.12928211241960524,
|
|
"num_tokens": 2619870.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.336686420440674,
|
|
"epoch": 1.2047402924861321,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004994738240711225,
|
|
"loss": 6.1157,
|
|
"mean_token_accuracy": 0.12358368933200836,
|
|
"num_tokens": 2631541.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.3283501148223875,
|
|
"epoch": 1.2097831568330812,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004994463633558945,
|
|
"loss": 5.9661,
|
|
"mean_token_accuracy": 0.1275870643556118,
|
|
"num_tokens": 2641896.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.21545786857605,
|
|
"epoch": 1.2148260211800301,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004994182050561917,
|
|
"loss": 6.042,
|
|
"mean_token_accuracy": 0.12058580294251442,
|
|
"num_tokens": 2654208.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.33097243309021,
|
|
"epoch": 1.2198688855269793,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004993893492595306,
|
|
"loss": 6.0159,
|
|
"mean_token_accuracy": 0.1281864397227764,
|
|
"num_tokens": 2665115.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.338105392456055,
|
|
"epoch": 1.2249117498739284,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004993597960555959,
|
|
"loss": 6.1094,
|
|
"mean_token_accuracy": 0.12263126000761986,
|
|
"num_tokens": 2676093.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.3116168022155765,
|
|
"epoch": 1.2299546142208775,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004993295455362395,
|
|
"loss": 6.0505,
|
|
"mean_token_accuracy": 0.1321282334625721,
|
|
"num_tokens": 2687866.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.302728509902954,
|
|
"epoch": 1.2349974785678266,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004992985977954808,
|
|
"loss": 6.0691,
|
|
"mean_token_accuracy": 0.12222623154520988,
|
|
"num_tokens": 2699814.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.31167459487915,
|
|
"epoch": 1.2400403429147757,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000499266952929506,
|
|
"loss": 6.04,
|
|
"mean_token_accuracy": 0.13241140097379683,
|
|
"num_tokens": 2710384.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.292968320846557,
|
|
"epoch": 1.2450832072617246,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004992346110366682,
|
|
"loss": 6.005,
|
|
"mean_token_accuracy": 0.12606269121170044,
|
|
"num_tokens": 2721442.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.2806806564331055,
|
|
"epoch": 1.2501260716086737,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004992015722174865,
|
|
"loss": 6.0363,
|
|
"mean_token_accuracy": 0.12223082482814789,
|
|
"num_tokens": 2733605.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.310382080078125,
|
|
"epoch": 1.2551689359556228,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004991678365746466,
|
|
"loss": 6.0663,
|
|
"mean_token_accuracy": 0.1318654552102089,
|
|
"num_tokens": 2744853.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.200191354751587,
|
|
"epoch": 1.2602118003025717,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004991334042129996,
|
|
"loss": 5.9704,
|
|
"mean_token_accuracy": 0.13543844670057298,
|
|
"num_tokens": 2754477.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.297356128692627,
|
|
"epoch": 1.2652546646495209,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004990982752395618,
|
|
"loss": 5.9907,
|
|
"mean_token_accuracy": 0.12480146735906601,
|
|
"num_tokens": 2765895.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.247424459457397,
|
|
"epoch": 1.27029752899647,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004990624497635152,
|
|
"loss": 6.0287,
|
|
"mean_token_accuracy": 0.12800199538469315,
|
|
"num_tokens": 2776897.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.351614952087402,
|
|
"epoch": 1.275340393343419,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.000499025927896206,
|
|
"loss": 6.0447,
|
|
"mean_token_accuracy": 0.12583597749471664,
|
|
"num_tokens": 2788666.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.213450241088867,
|
|
"epoch": 1.2803832576903682,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004989887097511451,
|
|
"loss": 5.9402,
|
|
"mean_token_accuracy": 0.12620697543025017,
|
|
"num_tokens": 2799167.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.208583068847656,
|
|
"epoch": 1.2854261220373173,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004989507954440074,
|
|
"loss": 5.9497,
|
|
"mean_token_accuracy": 0.125148519128561,
|
|
"num_tokens": 2809774.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.248659753799439,
|
|
"epoch": 1.2904689863842662,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004989121850926313,
|
|
"loss": 6.0665,
|
|
"mean_token_accuracy": 0.12353873103857041,
|
|
"num_tokens": 2820515.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.255040407180786,
|
|
"epoch": 1.2955118507312153,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004988728788170188,
|
|
"loss": 6.0036,
|
|
"mean_token_accuracy": 0.12691570073366165,
|
|
"num_tokens": 2832163.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.244554233551026,
|
|
"epoch": 1.3005547150781644,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004988328767393346,
|
|
"loss": 5.9936,
|
|
"mean_token_accuracy": 0.12777815014123917,
|
|
"num_tokens": 2843717.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.182911777496338,
|
|
"epoch": 1.3055975794251133,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004987921789839061,
|
|
"loss": 5.9551,
|
|
"mean_token_accuracy": 0.1249240666627884,
|
|
"num_tokens": 2854489.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.2384546279907225,
|
|
"epoch": 1.3106404437720625,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004987507856772229,
|
|
"loss": 5.9983,
|
|
"mean_token_accuracy": 0.1261892832815647,
|
|
"num_tokens": 2865317.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.236528825759888,
|
|
"epoch": 1.3156833081190116,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004987086969479364,
|
|
"loss": 5.9279,
|
|
"mean_token_accuracy": 0.12764144465327262,
|
|
"num_tokens": 2877466.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.200816583633423,
|
|
"epoch": 1.3207261724659607,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004986659129268592,
|
|
"loss": 5.9469,
|
|
"mean_token_accuracy": 0.1273733749985695,
|
|
"num_tokens": 2887698.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.093001461029052,
|
|
"epoch": 1.3257690368129098,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004986224337469652,
|
|
"loss": 5.911,
|
|
"mean_token_accuracy": 0.13541385903954506,
|
|
"num_tokens": 2899339.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.184291648864746,
|
|
"epoch": 1.330811901159859,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004985782595433886,
|
|
"loss": 5.9541,
|
|
"mean_token_accuracy": 0.1313224196434021,
|
|
"num_tokens": 2908340.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.297305583953857,
|
|
"epoch": 1.3358547655068078,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004985333904534237,
|
|
"loss": 6.0437,
|
|
"mean_token_accuracy": 0.12720400467514992,
|
|
"num_tokens": 2920432.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.2412254333496096,
|
|
"epoch": 1.340897629853757,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000498487826616525,
|
|
"loss": 5.9973,
|
|
"mean_token_accuracy": 0.12265557274222375,
|
|
"num_tokens": 2930385.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.255727338790893,
|
|
"epoch": 1.345940494200706,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004984415681743056,
|
|
"loss": 5.9564,
|
|
"mean_token_accuracy": 0.12846655994653702,
|
|
"num_tokens": 2940760.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.201718139648437,
|
|
"epoch": 1.350983358547655,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004983946152705382,
|
|
"loss": 5.9618,
|
|
"mean_token_accuracy": 0.13474117815494538,
|
|
"num_tokens": 2952441.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.288907527923584,
|
|
"epoch": 1.356026222894604,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004983469680511531,
|
|
"loss": 5.9944,
|
|
"mean_token_accuracy": 0.12811118066310884,
|
|
"num_tokens": 2962766.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.08305811882019,
|
|
"epoch": 1.3610690872415532,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004982986266642393,
|
|
"loss": 5.8789,
|
|
"mean_token_accuracy": 0.13739963695406915,
|
|
"num_tokens": 2974239.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.2541156768798825,
|
|
"epoch": 1.3661119515885023,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004982495912600426,
|
|
"loss": 5.9234,
|
|
"mean_token_accuracy": 0.1332431972026825,
|
|
"num_tokens": 2984497.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.171270656585693,
|
|
"epoch": 1.3711548159354514,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004981998619909664,
|
|
"loss": 5.9094,
|
|
"mean_token_accuracy": 0.1351434640586376,
|
|
"num_tokens": 2995367.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.187997531890869,
|
|
"epoch": 1.3761976802824005,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004981494390115704,
|
|
"loss": 5.9337,
|
|
"mean_token_accuracy": 0.13198790550231934,
|
|
"num_tokens": 3007687.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.221469449996948,
|
|
"epoch": 1.3812405446293494,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004980983224785704,
|
|
"loss": 5.9315,
|
|
"mean_token_accuracy": 0.12946036010980605,
|
|
"num_tokens": 3017505.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.105452346801758,
|
|
"epoch": 1.3862834089762985,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004980465125508376,
|
|
"loss": 5.8702,
|
|
"mean_token_accuracy": 0.13039664775133133,
|
|
"num_tokens": 3028271.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.187194681167602,
|
|
"epoch": 1.3913262733232477,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004979940093893986,
|
|
"loss": 5.9277,
|
|
"mean_token_accuracy": 0.13038016557693483,
|
|
"num_tokens": 3040440.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.109730768203735,
|
|
"epoch": 1.3963691376701965,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004979408131574345,
|
|
"loss": 5.9059,
|
|
"mean_token_accuracy": 0.1315137878060341,
|
|
"num_tokens": 3050999.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.239810371398926,
|
|
"epoch": 1.4014120020171457,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004978869240202802,
|
|
"loss": 5.9602,
|
|
"mean_token_accuracy": 0.1289587862789631,
|
|
"num_tokens": 3063374.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.167391586303711,
|
|
"epoch": 1.4064548663640948,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004978323421454246,
|
|
"loss": 5.9927,
|
|
"mean_token_accuracy": 0.13119542375206947,
|
|
"num_tokens": 3075464.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.129855394363403,
|
|
"epoch": 1.411497730711044,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004977770677025092,
|
|
"loss": 5.8591,
|
|
"mean_token_accuracy": 0.13318418264389037,
|
|
"num_tokens": 3086398.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.136085891723633,
|
|
"epoch": 1.416540595057993,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004977211008633284,
|
|
"loss": 5.9911,
|
|
"mean_token_accuracy": 0.12132739946246147,
|
|
"num_tokens": 3097777.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.179863452911377,
|
|
"epoch": 1.4215834594049421,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004976644418018285,
|
|
"loss": 5.9536,
|
|
"mean_token_accuracy": 0.13351338282227515,
|
|
"num_tokens": 3108711.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.222500562667847,
|
|
"epoch": 1.426626323751891,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000497607090694107,
|
|
"loss": 5.9831,
|
|
"mean_token_accuracy": 0.1274914562702179,
|
|
"num_tokens": 3119589.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.082250499725342,
|
|
"epoch": 1.4316691880988401,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004975490477184126,
|
|
"loss": 5.9187,
|
|
"mean_token_accuracy": 0.1321844443678856,
|
|
"num_tokens": 3131744.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.268057537078858,
|
|
"epoch": 1.4367120524457893,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000497490313055144,
|
|
"loss": 5.955,
|
|
"mean_token_accuracy": 0.12634534537792205,
|
|
"num_tokens": 3143526.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.115379571914673,
|
|
"epoch": 1.4417549167927382,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004974308868868501,
|
|
"loss": 5.8917,
|
|
"mean_token_accuracy": 0.12933965548872947,
|
|
"num_tokens": 3154561.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.187546491622925,
|
|
"epoch": 1.4467977811396873,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004973707693982287,
|
|
"loss": 6.0571,
|
|
"mean_token_accuracy": 0.12402074038982391,
|
|
"num_tokens": 3166052.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.280006074905396,
|
|
"epoch": 1.4518406454866364,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.0004973099607761264,
|
|
"loss": 6.0143,
|
|
"mean_token_accuracy": 0.12498047798871995,
|
|
"num_tokens": 3179335.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.124026155471801,
|
|
"epoch": 1.4568835098335855,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004972484612095378,
|
|
"loss": 5.8948,
|
|
"mean_token_accuracy": 0.1293070510029793,
|
|
"num_tokens": 3192129.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.247459745407104,
|
|
"epoch": 1.4619263741805346,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000497186270889605,
|
|
"loss": 6.0099,
|
|
"mean_token_accuracy": 0.12838999554514885,
|
|
"num_tokens": 3202792.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.181798028945923,
|
|
"epoch": 1.4669692385274837,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004971233900096169,
|
|
"loss": 5.9532,
|
|
"mean_token_accuracy": 0.1265900619328022,
|
|
"num_tokens": 3213707.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.254209232330322,
|
|
"epoch": 1.4720121028744326,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004970598187650085,
|
|
"loss": 5.9438,
|
|
"mean_token_accuracy": 0.13058415800333023,
|
|
"num_tokens": 3224531.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.14162049293518,
|
|
"epoch": 1.4770549672213817,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000496995557353361,
|
|
"loss": 5.9229,
|
|
"mean_token_accuracy": 0.13472673147916794,
|
|
"num_tokens": 3234656.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.159527111053467,
|
|
"epoch": 1.4820978315683309,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004969306059744004,
|
|
"loss": 5.9529,
|
|
"mean_token_accuracy": 0.12983041852712632,
|
|
"num_tokens": 3246745.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.079238557815552,
|
|
"epoch": 1.4871406959152798,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004968649648299968,
|
|
"loss": 5.9325,
|
|
"mean_token_accuracy": 0.13039209172129632,
|
|
"num_tokens": 3257277.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.221942520141601,
|
|
"epoch": 1.4921835602622289,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004967986341241646,
|
|
"loss": 5.8996,
|
|
"mean_token_accuracy": 0.1321016862988472,
|
|
"num_tokens": 3267704.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.176310968399048,
|
|
"epoch": 1.497226424609178,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004967316140630611,
|
|
"loss": 5.9687,
|
|
"mean_token_accuracy": 0.13029878064990044,
|
|
"num_tokens": 3279772.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.158623075485229,
|
|
"epoch": 1.502269288956127,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004966639048549862,
|
|
"loss": 5.8634,
|
|
"mean_token_accuracy": 0.1383678749203682,
|
|
"num_tokens": 3290536.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.0815863609313965,
|
|
"epoch": 1.5073121533030762,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004965955067103816,
|
|
"loss": 5.9482,
|
|
"mean_token_accuracy": 0.136953729391098,
|
|
"num_tokens": 3302051.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.164208698272705,
|
|
"epoch": 1.5123550176500253,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004965264198418305,
|
|
"loss": 5.9239,
|
|
"mean_token_accuracy": 0.12362579479813576,
|
|
"num_tokens": 3314324.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"epoch": 1.5123550176500253,
|
|
"eval_entropy": 5.92913255072732,
|
|
"eval_loss": 6.065815448760986,
|
|
"eval_mean_token_accuracy": 0.13071360590727882,
|
|
"eval_num_tokens": 3314324.0,
|
|
"eval_runtime": 2.0445,
|
|
"eval_samples_per_second": 1534.866,
|
|
"eval_steps_per_second": 192.225,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 6.07640495300293,
|
|
"epoch": 1.5173978819969742,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004964566444640564,
|
|
"loss": 5.9111,
|
|
"mean_token_accuracy": 0.13448831364512442,
|
|
"num_tokens": 3326246.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 6.133991718292236,
|
|
"epoch": 1.5224407463439233,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004963861807939227,
|
|
"loss": 5.9101,
|
|
"mean_token_accuracy": 0.13363656848669053,
|
|
"num_tokens": 3336308.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 6.0454902172088625,
|
|
"epoch": 1.5274836106908725,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004963150290504321,
|
|
"loss": 5.8513,
|
|
"mean_token_accuracy": 0.13463763669133186,
|
|
"num_tokens": 3347596.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 6.061435747146606,
|
|
"epoch": 1.5325264750378214,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004962431894547259,
|
|
"loss": 5.8296,
|
|
"mean_token_accuracy": 0.13526435643434526,
|
|
"num_tokens": 3358370.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 6.218953037261963,
|
|
"epoch": 1.5375693393847705,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004961706622300834,
|
|
"loss": 5.8602,
|
|
"mean_token_accuracy": 0.12972610592842101,
|
|
"num_tokens": 3368630.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 6.079063796997071,
|
|
"epoch": 1.5426122037317196,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004960974476019206,
|
|
"loss": 5.8816,
|
|
"mean_token_accuracy": 0.13141481205821037,
|
|
"num_tokens": 3379952.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 6.05611252784729,
|
|
"epoch": 1.5476550680786687,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004960235457977905,
|
|
"loss": 5.8074,
|
|
"mean_token_accuracy": 0.13323391675949098,
|
|
"num_tokens": 3390097.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 6.045963382720947,
|
|
"epoch": 1.5526979324256178,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004959489570473814,
|
|
"loss": 5.8109,
|
|
"mean_token_accuracy": 0.13734844774007798,
|
|
"num_tokens": 3399778.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 6.1482421875,
|
|
"epoch": 1.557740796772567,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.000495873681582517,
|
|
"loss": 5.9663,
|
|
"mean_token_accuracy": 0.12791561409831048,
|
|
"num_tokens": 3410213.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 6.119000482559204,
|
|
"epoch": 1.5627836611195158,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004957977196371552,
|
|
"loss": 5.9637,
|
|
"mean_token_accuracy": 0.13379994928836822,
|
|
"num_tokens": 3422071.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 6.193398761749267,
|
|
"epoch": 1.567826525466465,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004957210714473873,
|
|
"loss": 5.9446,
|
|
"mean_token_accuracy": 0.1331654317677021,
|
|
"num_tokens": 3432102.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 6.056936597824096,
|
|
"epoch": 1.572869389813414,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004956437372514378,
|
|
"loss": 5.9172,
|
|
"mean_token_accuracy": 0.1301022320985794,
|
|
"num_tokens": 3442821.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 6.093140411376953,
|
|
"epoch": 1.577912254160363,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004955657172896631,
|
|
"loss": 5.8615,
|
|
"mean_token_accuracy": 0.13420210480690004,
|
|
"num_tokens": 3454378.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 6.064199829101563,
|
|
"epoch": 1.582955118507312,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004954870118045511,
|
|
"loss": 5.8829,
|
|
"mean_token_accuracy": 0.13961151838302613,
|
|
"num_tokens": 3463933.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 6.100225591659546,
|
|
"epoch": 1.5879979828542612,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004954076210407203,
|
|
"loss": 5.8286,
|
|
"mean_token_accuracy": 0.13067737594246864,
|
|
"num_tokens": 3473993.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 6.107269954681397,
|
|
"epoch": 1.5930408472012103,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004953275452449188,
|
|
"loss": 5.9742,
|
|
"mean_token_accuracy": 0.12542441859841347,
|
|
"num_tokens": 3484882.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 6.136404132843017,
|
|
"epoch": 1.5980837115481594,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004952467846660243,
|
|
"loss": 5.8912,
|
|
"mean_token_accuracy": 0.13101620897650718,
|
|
"num_tokens": 3496503.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 6.102597761154175,
|
|
"epoch": 1.6031265758951085,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004951653395550423,
|
|
"loss": 5.9766,
|
|
"mean_token_accuracy": 0.13298841714859008,
|
|
"num_tokens": 3508696.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 6.1873414516448975,
|
|
"epoch": 1.6081694402420574,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004950832101651062,
|
|
"loss": 5.8911,
|
|
"mean_token_accuracy": 0.13785310089588165,
|
|
"num_tokens": 3519542.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 6.101690292358398,
|
|
"epoch": 1.6132123045890066,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004950003967514762,
|
|
"loss": 5.8531,
|
|
"mean_token_accuracy": 0.1334747441112995,
|
|
"num_tokens": 3531576.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 6.099046945571899,
|
|
"epoch": 1.6182551689359557,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000494916899571538,
|
|
"loss": 5.8216,
|
|
"mean_token_accuracy": 0.13443656489253045,
|
|
"num_tokens": 3542816.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 6.117959117889404,
|
|
"epoch": 1.6232980332829046,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004948327188848029,
|
|
"loss": 5.9468,
|
|
"mean_token_accuracy": 0.12940212413668634,
|
|
"num_tokens": 3554100.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 6.150053024291992,
|
|
"epoch": 1.6283408976298537,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004947478549529065,
|
|
"loss": 5.8034,
|
|
"mean_token_accuracy": 0.1349252998828888,
|
|
"num_tokens": 3565048.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 6.134335947036743,
|
|
"epoch": 1.6333837619768028,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004946623080396077,
|
|
"loss": 5.8889,
|
|
"mean_token_accuracy": 0.1347687289118767,
|
|
"num_tokens": 3575587.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 6.073554849624633,
|
|
"epoch": 1.638426626323752,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004945760784107884,
|
|
"loss": 5.9495,
|
|
"mean_token_accuracy": 0.13131332620978356,
|
|
"num_tokens": 3587059.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 6.063474702835083,
|
|
"epoch": 1.643469490670701,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004944891663344522,
|
|
"loss": 5.7899,
|
|
"mean_token_accuracy": 0.13512271270155907,
|
|
"num_tokens": 3598492.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 6.031048393249511,
|
|
"epoch": 1.6485123550176501,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004944015720807239,
|
|
"loss": 5.8573,
|
|
"mean_token_accuracy": 0.13475991785526276,
|
|
"num_tokens": 3609513.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 6.127387857437133,
|
|
"epoch": 1.653555219364599,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004943132959218484,
|
|
"loss": 5.8857,
|
|
"mean_token_accuracy": 0.13433710113167763,
|
|
"num_tokens": 3621327.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 6.042216396331787,
|
|
"epoch": 1.6585980837115482,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00049422433813219,
|
|
"loss": 5.7909,
|
|
"mean_token_accuracy": 0.13622564375400542,
|
|
"num_tokens": 3632136.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 6.078888177871704,
|
|
"epoch": 1.6636409480584973,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004941346989882318,
|
|
"loss": 5.7725,
|
|
"mean_token_accuracy": 0.13878510296344757,
|
|
"num_tokens": 3642271.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 5.900990581512451,
|
|
"epoch": 1.6686838124054462,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004940443787685739,
|
|
"loss": 5.9042,
|
|
"mean_token_accuracy": 0.13379739746451377,
|
|
"num_tokens": 3652355.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 6.232107543945313,
|
|
"epoch": 1.6737266767523953,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004939533777539341,
|
|
"loss": 5.8782,
|
|
"mean_token_accuracy": 0.13564058914780616,
|
|
"num_tokens": 3663714.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 5.9709052562713625,
|
|
"epoch": 1.6787695410993444,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004938616962271453,
|
|
"loss": 5.8313,
|
|
"mean_token_accuracy": 0.12849392145872116,
|
|
"num_tokens": 3674675.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 6.053460216522216,
|
|
"epoch": 1.6838124054462935,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004937693344731561,
|
|
"loss": 5.9073,
|
|
"mean_token_accuracy": 0.13890698850154876,
|
|
"num_tokens": 3686403.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 6.068472099304199,
|
|
"epoch": 1.6888552697932426,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004936762927790286,
|
|
"loss": 5.8053,
|
|
"mean_token_accuracy": 0.14557132124900818,
|
|
"num_tokens": 3697259.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 6.067635250091553,
|
|
"epoch": 1.6938981341401917,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004935825714339389,
|
|
"loss": 5.8499,
|
|
"mean_token_accuracy": 0.1388585165143013,
|
|
"num_tokens": 3708536.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 6.171128368377685,
|
|
"epoch": 1.6989409984871406,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004934881707291748,
|
|
"loss": 5.8554,
|
|
"mean_token_accuracy": 0.1358127549290657,
|
|
"num_tokens": 3719181.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 5.9123929977417,
|
|
"epoch": 1.7039838628340898,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004933930909581362,
|
|
"loss": 5.7124,
|
|
"mean_token_accuracy": 0.13930310904979706,
|
|
"num_tokens": 3729325.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 6.060745573043823,
|
|
"epoch": 1.7090267271810389,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004932973324163328,
|
|
"loss": 5.8537,
|
|
"mean_token_accuracy": 0.13134554773569107,
|
|
"num_tokens": 3740462.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 6.171601963043213,
|
|
"epoch": 1.7140695915279878,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004932008954013846,
|
|
"loss": 5.9263,
|
|
"mean_token_accuracy": 0.13537270352244377,
|
|
"num_tokens": 3752523.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 5.921985626220703,
|
|
"epoch": 1.7191124558749369,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004931037802130201,
|
|
"loss": 5.7874,
|
|
"mean_token_accuracy": 0.1338386692106724,
|
|
"num_tokens": 3764120.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 6.072412490844727,
|
|
"epoch": 1.724155320221886,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004930059871530755,
|
|
"loss": 5.9185,
|
|
"mean_token_accuracy": 0.13412620201706887,
|
|
"num_tokens": 3776733.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 6.1733163356781,
|
|
"epoch": 1.7291981845688351,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004929075165254938,
|
|
"loss": 5.8943,
|
|
"mean_token_accuracy": 0.1337961360812187,
|
|
"num_tokens": 3787799.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 5.941089677810669,
|
|
"epoch": 1.7342410489157842,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004928083686363241,
|
|
"loss": 5.8108,
|
|
"mean_token_accuracy": 0.1348222739994526,
|
|
"num_tokens": 3797995.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 6.141709613800049,
|
|
"epoch": 1.7392839132627333,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004927085437937204,
|
|
"loss": 5.8929,
|
|
"mean_token_accuracy": 0.13604455217719078,
|
|
"num_tokens": 3809580.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 6.107912921905518,
|
|
"epoch": 1.7443267776096822,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004926080423079404,
|
|
"loss": 5.9223,
|
|
"mean_token_accuracy": 0.1334833487868309,
|
|
"num_tokens": 3820150.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 6.025250625610352,
|
|
"epoch": 1.7493696419566314,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000492506864491345,
|
|
"loss": 5.7724,
|
|
"mean_token_accuracy": 0.13755011111497878,
|
|
"num_tokens": 3830391.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 5.940911483764649,
|
|
"epoch": 1.7544125063035805,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004924050106583976,
|
|
"loss": 5.7949,
|
|
"mean_token_accuracy": 0.14076849222183227,
|
|
"num_tokens": 3842773.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 6.115080833435059,
|
|
"epoch": 1.7594553706505294,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004923024811256618,
|
|
"loss": 5.8311,
|
|
"mean_token_accuracy": 0.1350576788187027,
|
|
"num_tokens": 3853646.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 6.017044448852539,
|
|
"epoch": 1.7644982349974785,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000492199276211802,
|
|
"loss": 5.8024,
|
|
"mean_token_accuracy": 0.13398425951600074,
|
|
"num_tokens": 3865319.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 6.000295162200928,
|
|
"epoch": 1.7695410993444276,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004920953962375813,
|
|
"loss": 5.8484,
|
|
"mean_token_accuracy": 0.13326467648148538,
|
|
"num_tokens": 3877107.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 6.0226422309875485,
|
|
"epoch": 1.7745839636913767,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004919908415258612,
|
|
"loss": 5.8988,
|
|
"mean_token_accuracy": 0.1349693864583969,
|
|
"num_tokens": 3889334.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 6.117556428909301,
|
|
"epoch": 1.7796268280383258,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004918856124016002,
|
|
"loss": 5.7835,
|
|
"mean_token_accuracy": 0.14186740294098854,
|
|
"num_tokens": 3899482.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 5.953726100921631,
|
|
"epoch": 1.784669692385275,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004917797091918526,
|
|
"loss": 5.71,
|
|
"mean_token_accuracy": 0.1396735355257988,
|
|
"num_tokens": 3910216.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 6.064425754547119,
|
|
"epoch": 1.789712556732224,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004916731322257683,
|
|
"loss": 5.8681,
|
|
"mean_token_accuracy": 0.13618118837475776,
|
|
"num_tokens": 3922246.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 6.028489112854004,
|
|
"epoch": 1.794755421079173,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004915658818345908,
|
|
"loss": 5.7669,
|
|
"mean_token_accuracy": 0.14441163688898087,
|
|
"num_tokens": 3933688.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 6.023671102523804,
|
|
"epoch": 1.799798285426122,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004914579583516568,
|
|
"loss": 5.8717,
|
|
"mean_token_accuracy": 0.13164491429924965,
|
|
"num_tokens": 3944168.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 5.958587312698365,
|
|
"epoch": 1.804841149773071,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000491349362112395,
|
|
"loss": 5.7789,
|
|
"mean_token_accuracy": 0.14149321988224983,
|
|
"num_tokens": 3954346.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 6.089543962478638,
|
|
"epoch": 1.80988401412002,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.000491240093454325,
|
|
"loss": 5.8937,
|
|
"mean_token_accuracy": 0.13598302081227304,
|
|
"num_tokens": 3966375.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 6.050840950012207,
|
|
"epoch": 1.8149268784669692,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004911301527170563,
|
|
"loss": 5.8619,
|
|
"mean_token_accuracy": 0.13489308282732965,
|
|
"num_tokens": 3977030.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 5.947264814376831,
|
|
"epoch": 1.8199697428139183,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004910195402422873,
|
|
"loss": 5.7796,
|
|
"mean_token_accuracy": 0.13757651001214982,
|
|
"num_tokens": 3988064.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 6.084697341918945,
|
|
"epoch": 1.8250126071608674,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004909082563738042,
|
|
"loss": 5.8309,
|
|
"mean_token_accuracy": 0.13917500153183937,
|
|
"num_tokens": 3997994.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 6.1184464454650875,
|
|
"epoch": 1.8300554715078166,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004907963014574797,
|
|
"loss": 5.8819,
|
|
"mean_token_accuracy": 0.13045083805918695,
|
|
"num_tokens": 4008645.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 5.999236679077148,
|
|
"epoch": 1.8350983358547657,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004906836758412722,
|
|
"loss": 5.8269,
|
|
"mean_token_accuracy": 0.13780293762683868,
|
|
"num_tokens": 4020387.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 5.857476043701172,
|
|
"epoch": 1.8401412002017146,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004905703798752248,
|
|
"loss": 5.6984,
|
|
"mean_token_accuracy": 0.14144380390644073,
|
|
"num_tokens": 4032720.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 5.971011304855347,
|
|
"epoch": 1.8451840645486637,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.000490456413911464,
|
|
"loss": 5.7757,
|
|
"mean_token_accuracy": 0.13810631558299064,
|
|
"num_tokens": 4043438.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 5.9294352531433105,
|
|
"epoch": 1.8502269288956126,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004903417783041987,
|
|
"loss": 5.7543,
|
|
"mean_token_accuracy": 0.14420028328895568,
|
|
"num_tokens": 4054467.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 6.036222648620606,
|
|
"epoch": 1.8552697932425617,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004902264734097188,
|
|
"loss": 5.7498,
|
|
"mean_token_accuracy": 0.13947635442018508,
|
|
"num_tokens": 4066418.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 6.067141771316528,
|
|
"epoch": 1.8603126575895108,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004901104995863948,
|
|
"loss": 5.849,
|
|
"mean_token_accuracy": 0.13374825641512872,
|
|
"num_tokens": 4078770.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 6.000591516494751,
|
|
"epoch": 1.86535552193646,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004899938571946759,
|
|
"loss": 5.8825,
|
|
"mean_token_accuracy": 0.13524958193302156,
|
|
"num_tokens": 4089998.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 6.056681966781616,
|
|
"epoch": 1.870398386283409,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004898765465970894,
|
|
"loss": 5.7893,
|
|
"mean_token_accuracy": 0.14301273375749587,
|
|
"num_tokens": 4101409.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 6.052677965164184,
|
|
"epoch": 1.8754412506303582,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004897585681582393,
|
|
"loss": 5.9051,
|
|
"mean_token_accuracy": 0.13408755287528037,
|
|
"num_tokens": 4113598.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 5.9217283725738525,
|
|
"epoch": 1.8804841149773073,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004896399222448053,
|
|
"loss": 5.8482,
|
|
"mean_token_accuracy": 0.135797618329525,
|
|
"num_tokens": 4124639.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 6.125091075897217,
|
|
"epoch": 1.8855269793242562,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004895206092255417,
|
|
"loss": 5.9119,
|
|
"mean_token_accuracy": 0.12845465615391732,
|
|
"num_tokens": 4136280.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 6.0031355857849125,
|
|
"epoch": 1.8905698436712053,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004894006294712761,
|
|
"loss": 5.7546,
|
|
"mean_token_accuracy": 0.1370631605386734,
|
|
"num_tokens": 4148251.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 6.040983438491821,
|
|
"epoch": 1.8956127080181542,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004892799833549083,
|
|
"loss": 5.8332,
|
|
"mean_token_accuracy": 0.13612558841705322,
|
|
"num_tokens": 4159298.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 6.031946277618408,
|
|
"epoch": 1.9006555723651033,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004891586712514091,
|
|
"loss": 5.7585,
|
|
"mean_token_accuracy": 0.13925607949495317,
|
|
"num_tokens": 4169974.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 5.925895166397095,
|
|
"epoch": 1.9056984367120524,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004890366935378196,
|
|
"loss": 5.7572,
|
|
"mean_token_accuracy": 0.1423736333847046,
|
|
"num_tokens": 4181442.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 6.010127925872803,
|
|
"epoch": 1.9107413010590015,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004889140505932492,
|
|
"loss": 5.7996,
|
|
"mean_token_accuracy": 0.14177473187446593,
|
|
"num_tokens": 4191535.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 5.964219427108764,
|
|
"epoch": 1.9157841654059506,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004887907427988749,
|
|
"loss": 5.8233,
|
|
"mean_token_accuracy": 0.14068215638399123,
|
|
"num_tokens": 4201741.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 5.975893354415893,
|
|
"epoch": 1.9208270297528998,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004886667705379403,
|
|
"loss": 5.7198,
|
|
"mean_token_accuracy": 0.14402899593114854,
|
|
"num_tokens": 4211762.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 6.003858184814453,
|
|
"epoch": 1.9258698940998489,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004885421341957541,
|
|
"loss": 5.799,
|
|
"mean_token_accuracy": 0.13982706964015962,
|
|
"num_tokens": 4222322.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 6.066865730285644,
|
|
"epoch": 1.9309127584467978,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004884168341596888,
|
|
"loss": 5.8777,
|
|
"mean_token_accuracy": 0.14073347598314284,
|
|
"num_tokens": 4232166.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 5.997572183609009,
|
|
"epoch": 1.9359556227937469,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004882908708191797,
|
|
"loss": 5.834,
|
|
"mean_token_accuracy": 0.13255441561341286,
|
|
"num_tokens": 4243651.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 6.015108919143676,
|
|
"epoch": 1.9409984871406958,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.00048816424456572403,
|
|
"loss": 5.7598,
|
|
"mean_token_accuracy": 0.14530409276485443,
|
|
"num_tokens": 4255137.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 5.898231554031372,
|
|
"epoch": 1.946041351487645,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00048803695579287905,
|
|
"loss": 5.7681,
|
|
"mean_token_accuracy": 0.13556032925844191,
|
|
"num_tokens": 4264904.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 5.926236581802368,
|
|
"epoch": 1.951084215834594,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00048790900489626104,
|
|
"loss": 5.7307,
|
|
"mean_token_accuracy": 0.13858372569084168,
|
|
"num_tokens": 4275305.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 5.99683027267456,
|
|
"epoch": 1.9561270801815431,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004877803922735445,
|
|
"loss": 5.8136,
|
|
"mean_token_accuracy": 0.13879240304231644,
|
|
"num_tokens": 4287038.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 6.131782913208008,
|
|
"epoch": 1.9611699445284922,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004876511183244603,
|
|
"loss": 5.9209,
|
|
"mean_token_accuracy": 0.13195879608392716,
|
|
"num_tokens": 4298668.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 5.913279676437378,
|
|
"epoch": 1.9662128088754414,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00048752118345079496,
|
|
"loss": 5.6918,
|
|
"mean_token_accuracy": 0.14129177182912828,
|
|
"num_tokens": 4309429.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 5.979755687713623,
|
|
"epoch": 1.9712556732223905,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00048739058805638905,
|
|
"loss": 5.7812,
|
|
"mean_token_accuracy": 0.1406949557363987,
|
|
"num_tokens": 4321137.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 6.013188457489013,
|
|
"epoch": 1.9762985375693394,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000487259332547136,
|
|
"loss": 5.8151,
|
|
"mean_token_accuracy": 0.1299636922776699,
|
|
"num_tokens": 4331249.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 5.987004327774048,
|
|
"epoch": 1.9813414019162885,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000487127417330981,
|
|
"loss": 5.7964,
|
|
"mean_token_accuracy": 0.13558189198374748,
|
|
"num_tokens": 4342154.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 6.070882320404053,
|
|
"epoch": 1.9863842662632374,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004869948428179196,
|
|
"loss": 5.8384,
|
|
"mean_token_accuracy": 0.134127613902092,
|
|
"num_tokens": 4353906.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 5.964674758911133,
|
|
"epoch": 1.9914271306101865,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0004868616094199964,
|
|
"loss": 5.8851,
|
|
"mean_token_accuracy": 0.13231149092316627,
|
|
"num_tokens": 4366416.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 6.027207851409912,
|
|
"epoch": 1.9964699949571356,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000486727717551304,
|
|
"loss": 5.7887,
|
|
"mean_token_accuracy": 0.139586428552866,
|
|
"num_tokens": 4376743.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 5.891790284050836,
|
|
"epoch": 2.00100857286939,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004865931676279813,
|
|
"loss": 5.5739,
|
|
"mean_token_accuracy": 0.15117713643444908,
|
|
"num_tokens": 4385379.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 5.955478048324585,
|
|
"epoch": 2.006051437216339,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004864579600682125,
|
|
"loss": 5.5227,
|
|
"mean_token_accuracy": 0.14686036109924316,
|
|
"num_tokens": 4396872.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 5.963762331008911,
|
|
"epoch": 2.011094301563288,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00048632209529222607,
|
|
"loss": 5.4546,
|
|
"mean_token_accuracy": 0.15201186388731003,
|
|
"num_tokens": 4407075.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 5.865040445327759,
|
|
"epoch": 2.016137165910237,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00048618557372229253,
|
|
"loss": 5.4845,
|
|
"mean_token_accuracy": 0.15829388946294784,
|
|
"num_tokens": 4418020.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 2.016137165910237,
|
|
"eval_entropy": 5.68467633718146,
|
|
"eval_loss": 5.918817520141602,
|
|
"eval_mean_token_accuracy": 0.137464204229623,
|
|
"eval_num_tokens": 4418020.0,
|
|
"eval_runtime": 2.039,
|
|
"eval_samples_per_second": 1539.015,
|
|
"eval_steps_per_second": 192.745,
|
|
"step": 2000
|
|
}
|
|
],
|
|
"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": 1095357125652480.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|