ita-latn-10mb-after-ppt-Dp-10mb-ckpt500_seed3407/checkpoint-6000/trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 6.048411497730711,
  "eval_steps": 500,
  "global_step": 6000,
  "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
    },
    {
      "entropy": 5.917067813873291,
      "epoch": 2.0211800302571863,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004860483957827243,
      "loss": 5.5135,
      "mean_token_accuracy": 0.15428032279014586,
      "num_tokens": 4429740.0,
      "step": 2005
    },
    {
      "entropy": 5.99223747253418,
      "epoch": 2.026222894604135,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004859105618998735,
      "loss": 5.4874,
      "mean_token_accuracy": 0.14928828477859496,
      "num_tokens": 4440183.0,
      "step": 2010
    },
    {
      "entropy": 5.905515241622925,
      "epoch": 2.031265758951084,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004857720725021311,
      "loss": 5.4767,
      "mean_token_accuracy": 0.1511900842189789,
      "num_tokens": 4451750.0,
      "step": 2015
    },
    {
      "entropy": 5.7743007183074955,
      "epoch": 2.036308623298033,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048563292801992534,
      "loss": 5.4372,
      "mean_token_accuracy": 0.15540879219770432,
      "num_tokens": 4462071.0,
      "step": 2020
    },
    {
      "entropy": 5.968015956878662,
      "epoch": 2.0413514876449823,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004854931288857203,
      "loss": 5.6069,
      "mean_token_accuracy": 0.1447448194026947,
      "num_tokens": 4474241.0,
      "step": 2025
    },
    {
      "entropy": 5.934961032867432,
      "epoch": 2.0463943519919314,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00048535267553401505,
      "loss": 5.453,
      "mean_token_accuracy": 0.15439282059669496,
      "num_tokens": 4487083.0,
      "step": 2030
    },
    {
      "entropy": 5.784671497344971,
      "epoch": 2.0514372163388805,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048521156840134193,
      "loss": 5.5083,
      "mean_token_accuracy": 0.1550283372402191,
      "num_tokens": 4497181.0,
      "step": 2035
    },
    {
      "entropy": 5.953446102142334,
      "epoch": 2.0564800806858297,
      "grad_norm": 1.140625,
      "learning_rate": 0.000485069807926265,
      "loss": 5.4853,
      "mean_token_accuracy": 0.15532971620559693,
      "num_tokens": 4508250.0,
      "step": 2040
    },
    {
      "entropy": 5.854066467285156,
      "epoch": 2.0615229450327788,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004849273945493792,
      "loss": 5.4822,
      "mean_token_accuracy": 0.15595464184880256,
      "num_tokens": 4519778.0,
      "step": 2045
    },
    {
      "entropy": 5.857310056686401,
      "epoch": 2.066565809379728,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004847843287133084,
      "loss": 5.4681,
      "mean_token_accuracy": 0.15022315979003906,
      "num_tokens": 4530879.0,
      "step": 2050
    },
    {
      "entropy": 5.885866737365722,
      "epoch": 2.0716086737266766,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004846406108627048,
      "loss": 5.5349,
      "mean_token_accuracy": 0.15706447511911392,
      "num_tokens": 4542560.0,
      "step": 2055
    },
    {
      "entropy": 5.7986352920532225,
      "epoch": 2.0766515380736257,
      "grad_norm": 1.0,
      "learning_rate": 0.00048449624144424665,
      "loss": 5.425,
      "mean_token_accuracy": 0.16038900166749953,
      "num_tokens": 4552213.0,
      "step": 2060
    },
    {
      "entropy": 5.848188066482544,
      "epoch": 2.081694402420575,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00048435122090663744,
      "loss": 5.4514,
      "mean_token_accuracy": 0.15963771790266038,
      "num_tokens": 4562798.0,
      "step": 2065
    },
    {
      "entropy": 5.885587310791015,
      "epoch": 2.086737266767524,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048420554970060424,
      "loss": 5.4973,
      "mean_token_accuracy": 0.14645309895277023,
      "num_tokens": 4574534.0,
      "step": 2070
    },
    {
      "entropy": 5.855887317657471,
      "epoch": 2.091780131114473,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048405922827889677,
      "loss": 5.4657,
      "mean_token_accuracy": 0.1495537653565407,
      "num_tokens": 4586239.0,
      "step": 2075
    },
    {
      "entropy": 5.833091926574707,
      "epoch": 2.096822995461422,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048391225709628515,
      "loss": 5.5103,
      "mean_token_accuracy": 0.15426335334777833,
      "num_tokens": 4598018.0,
      "step": 2080
    },
    {
      "entropy": 5.871005439758301,
      "epoch": 2.1018658598083713,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004837646366095593,
      "loss": 5.4884,
      "mean_token_accuracy": 0.1490063801407814,
      "num_tokens": 4608985.0,
      "step": 2085
    },
    {
      "entropy": 5.857932424545288,
      "epoch": 2.1069087241553204,
      "grad_norm": 1.125,
      "learning_rate": 0.00048361636727752716,
      "loss": 5.542,
      "mean_token_accuracy": 0.1547345831990242,
      "num_tokens": 4621103.0,
      "step": 2090
    },
    {
      "entropy": 5.842451000213623,
      "epoch": 2.1119515885022695,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004834674495610131,
      "loss": 5.5248,
      "mean_token_accuracy": 0.15549753606319427,
      "num_tokens": 4632244.0,
      "step": 2095
    },
    {
      "entropy": 5.824394750595093,
      "epoch": 2.116994452849218,
      "grad_norm": 1.171875,
      "learning_rate": 0.00048331788392285693,
      "loss": 5.4396,
      "mean_token_accuracy": 0.15467981696128846,
      "num_tokens": 4642387.0,
      "step": 2100
    },
    {
      "entropy": 5.886133289337158,
      "epoch": 2.1220373171961673,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00048316767082791205,
      "loss": 5.5562,
      "mean_token_accuracy": 0.14972446486353874,
      "num_tokens": 4655181.0,
      "step": 2105
    },
    {
      "entropy": 5.8176854133605955,
      "epoch": 2.1270801815431164,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048301681074304427,
      "loss": 5.4698,
      "mean_token_accuracy": 0.15476909428834915,
      "num_tokens": 4665664.0,
      "step": 2110
    },
    {
      "entropy": 5.774665260314942,
      "epoch": 2.1321230458900655,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048286530413713016,
      "loss": 5.4367,
      "mean_token_accuracy": 0.1594224750995636,
      "num_tokens": 4677683.0,
      "step": 2115
    },
    {
      "entropy": 5.845004367828369,
      "epoch": 2.1371659102370146,
      "grad_norm": 1.171875,
      "learning_rate": 0.00048271315148105574,
      "loss": 5.4973,
      "mean_token_accuracy": 0.1480380341410637,
      "num_tokens": 4688891.0,
      "step": 2120
    },
    {
      "entropy": 5.824699211120605,
      "epoch": 2.1422087745839637,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000482560353247715,
      "loss": 5.5127,
      "mean_token_accuracy": 0.15076074451208116,
      "num_tokens": 4700765.0,
      "step": 2125
    },
    {
      "entropy": 5.857800912857056,
      "epoch": 2.147251638930913,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004824069099120084,
      "loss": 5.5329,
      "mean_token_accuracy": 0.15133563578128814,
      "num_tokens": 4712004.0,
      "step": 2130
    },
    {
      "entropy": 5.805648374557495,
      "epoch": 2.152294503277862,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004822528219508413,
      "loss": 5.4496,
      "mean_token_accuracy": 0.15349498838186265,
      "num_tokens": 4722427.0,
      "step": 2135
    },
    {
      "entropy": 5.870349168777466,
      "epoch": 2.157337367624811,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004820980898431227,
      "loss": 5.5316,
      "mean_token_accuracy": 0.1588563948869705,
      "num_tokens": 4733815.0,
      "step": 2140
    },
    {
      "entropy": 5.843318128585816,
      "epoch": 2.1623802319717598,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004819427140697635,
      "loss": 5.4885,
      "mean_token_accuracy": 0.1497521683573723,
      "num_tokens": 4745070.0,
      "step": 2145
    },
    {
      "entropy": 5.866540098190308,
      "epoch": 2.167423096318709,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004817866951136753,
      "loss": 5.5571,
      "mean_token_accuracy": 0.1433023065328598,
      "num_tokens": 4756066.0,
      "step": 2150
    },
    {
      "entropy": 5.882374477386475,
      "epoch": 2.172465960665658,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048163003345976847,
      "loss": 5.5551,
      "mean_token_accuracy": 0.1422162264585495,
      "num_tokens": 4767153.0,
      "step": 2155
    },
    {
      "entropy": 5.854587125778198,
      "epoch": 2.177508825012607,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004814727295949512,
      "loss": 5.4646,
      "mean_token_accuracy": 0.15525968819856645,
      "num_tokens": 4777554.0,
      "step": 2160
    },
    {
      "entropy": 5.80414571762085,
      "epoch": 2.1825516893595562,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00048131478400812735,
      "loss": 5.5127,
      "mean_token_accuracy": 0.15242572575807573,
      "num_tokens": 4788881.0,
      "step": 2165
    },
    {
      "entropy": 5.8979381084442135,
      "epoch": 2.1875945537065054,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004811561971901956,
      "loss": 5.4828,
      "mean_token_accuracy": 0.1481580525636673,
      "num_tokens": 4798680.0,
      "step": 2170
    },
    {
      "entropy": 5.853958797454834,
      "epoch": 2.1926374180534545,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004809969696340474,
      "loss": 5.4813,
      "mean_token_accuracy": 0.15251049250364304,
      "num_tokens": 4808096.0,
      "step": 2175
    },
    {
      "entropy": 5.752409648895264,
      "epoch": 2.1976802824004036,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048083710183456554,
      "loss": 5.4385,
      "mean_token_accuracy": 0.16206829100847245,
      "num_tokens": 4818778.0,
      "step": 2180
    },
    {
      "entropy": 5.757245588302612,
      "epoch": 2.2027231467473527,
      "grad_norm": 1.171875,
      "learning_rate": 0.00048067659428862306,
      "loss": 5.4378,
      "mean_token_accuracy": 0.15842392295598984,
      "num_tokens": 4828586.0,
      "step": 2185
    },
    {
      "entropy": 5.834274625778198,
      "epoch": 2.2077660110943014,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048051544749508096,
      "loss": 5.5268,
      "mean_token_accuracy": 0.1499437376856804,
      "num_tokens": 4839419.0,
      "step": 2190
    },
    {
      "entropy": 5.818333625793457,
      "epoch": 2.2128088754412505,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004803536619547872,
      "loss": 5.4635,
      "mean_token_accuracy": 0.15629658848047256,
      "num_tokens": 4849700.0,
      "step": 2195
    },
    {
      "entropy": 5.845297145843506,
      "epoch": 2.2178517397881996,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000480191238170575,
      "loss": 5.621,
      "mean_token_accuracy": 0.1466002345085144,
      "num_tokens": 4860261.0,
      "step": 2200
    },
    {
      "entropy": 5.872590589523315,
      "epoch": 2.2228946041351487,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004800281766472612,
      "loss": 5.5345,
      "mean_token_accuracy": 0.14465595185756683,
      "num_tokens": 4872277.0,
      "step": 2205
    },
    {
      "entropy": 5.823731279373169,
      "epoch": 2.227937468482098,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00047986447789164486,
      "loss": 5.5868,
      "mean_token_accuracy": 0.1541430413722992,
      "num_tokens": 4883512.0,
      "step": 2210
    },
    {
      "entropy": 5.848791027069092,
      "epoch": 2.232980332829047,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004797001424125054,
      "loss": 5.511,
      "mean_token_accuracy": 0.15099768787622453,
      "num_tokens": 4896197.0,
      "step": 2215
    },
    {
      "entropy": 5.8191258907318115,
      "epoch": 2.238023197175996,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047953517072060124,
      "loss": 5.4881,
      "mean_token_accuracy": 0.15332500040531158,
      "num_tokens": 4907397.0,
      "step": 2220
    },
    {
      "entropy": 5.8483093738555905,
      "epoch": 2.243066061522945,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004793695633286684,
      "loss": 5.5388,
      "mean_token_accuracy": 0.14989539384841918,
      "num_tokens": 4919008.0,
      "step": 2225
    },
    {
      "entropy": 5.777065944671631,
      "epoch": 2.2481089258698943,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047920332075141825,
      "loss": 5.4774,
      "mean_token_accuracy": 0.1533714011311531,
      "num_tokens": 4930877.0,
      "step": 2230
    },
    {
      "entropy": 5.786398124694824,
      "epoch": 2.253151790216843,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004790364435055367,
      "loss": 5.3835,
      "mean_token_accuracy": 0.1652776688337326,
      "num_tokens": 4942274.0,
      "step": 2235
    },
    {
      "entropy": 5.778122282028198,
      "epoch": 2.258194654563792,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000478868932109682,
      "loss": 5.4759,
      "mean_token_accuracy": 0.15380292609333993,
      "num_tokens": 4953740.0,
      "step": 2240
    },
    {
      "entropy": 5.838868284225464,
      "epoch": 2.263237518910741,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004787007870844835,
      "loss": 5.5114,
      "mean_token_accuracy": 0.15524505972862243,
      "num_tokens": 4964760.0,
      "step": 2245
    },
    {
      "entropy": 5.865630292892456,
      "epoch": 2.2682803832576903,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047853200895253976,
      "loss": 5.5726,
      "mean_token_accuracy": 0.1507643900811672,
      "num_tokens": 4976404.0,
      "step": 2250
    },
    {
      "entropy": 5.78554368019104,
      "epoch": 2.2733232476046394,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047836259823841716,
      "loss": 5.4598,
      "mean_token_accuracy": 0.147302483022213,
      "num_tokens": 4987552.0,
      "step": 2255
    },
    {
      "entropy": 5.808395290374756,
      "epoch": 2.2783661119515886,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047819255546864797,
      "loss": 5.4751,
      "mean_token_accuracy": 0.15528551936149598,
      "num_tokens": 4998912.0,
      "step": 2260
    },
    {
      "entropy": 5.849263525009155,
      "epoch": 2.2834089762985377,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004780218811717293,
      "loss": 5.4419,
      "mean_token_accuracy": 0.16054765433073043,
      "num_tokens": 5009556.0,
      "step": 2265
    },
    {
      "entropy": 5.760190820693969,
      "epoch": 2.288451840645487,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047785057587812057,
      "loss": 5.4893,
      "mean_token_accuracy": 0.15540411025285722,
      "num_tokens": 5021569.0,
      "step": 2270
    },
    {
      "entropy": 5.844359397888184,
      "epoch": 2.293494704992436,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047767864012024266,
      "loss": 5.5652,
      "mean_token_accuracy": 0.14730796068906785,
      "num_tokens": 5032994.0,
      "step": 2275
    },
    {
      "entropy": 5.8796374797821045,
      "epoch": 2.2985375693393846,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047750607443247585,
      "loss": 5.502,
      "mean_token_accuracy": 0.1529832512140274,
      "num_tokens": 5044681.0,
      "step": 2280
    },
    {
      "entropy": 5.717267274856567,
      "epoch": 2.3035804336863337,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00047733287935115837,
      "loss": 5.4152,
      "mean_token_accuracy": 0.16113044023513795,
      "num_tokens": 5055509.0,
      "step": 2285
    },
    {
      "entropy": 5.704999732971191,
      "epoch": 2.308623298033283,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004771590554145844,
      "loss": 5.4459,
      "mean_token_accuracy": 0.15664409399032592,
      "num_tokens": 5066903.0,
      "step": 2290
    },
    {
      "entropy": 5.803127956390381,
      "epoch": 2.313666162380232,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047698460316300285,
      "loss": 5.4902,
      "mean_token_accuracy": 0.15127339363098144,
      "num_tokens": 5078411.0,
      "step": 2295
    },
    {
      "entropy": 5.807288551330567,
      "epoch": 2.318709026727181,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004768095231386153,
      "loss": 5.5216,
      "mean_token_accuracy": 0.1488862954080105,
      "num_tokens": 5089811.0,
      "step": 2300
    },
    {
      "entropy": 5.853737449645996,
      "epoch": 2.32375189107413,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004766338158855744,
      "loss": 5.4969,
      "mean_token_accuracy": 0.15977130234241485,
      "num_tokens": 5101196.0,
      "step": 2305
    },
    {
      "entropy": 5.729493856430054,
      "epoch": 2.3287947554210793,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004764574819499826,
      "loss": 5.4854,
      "mean_token_accuracy": 0.15577066242694854,
      "num_tokens": 5112379.0,
      "step": 2310
    },
    {
      "entropy": 5.894827747344971,
      "epoch": 2.3338376197680284,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047628052187988955,
      "loss": 5.602,
      "mean_token_accuracy": 0.13939977288246155,
      "num_tokens": 5125065.0,
      "step": 2315
    },
    {
      "entropy": 5.8933769226074215,
      "epoch": 2.338880484114977,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004761029362252914,
      "loss": 5.5705,
      "mean_token_accuracy": 0.14920184016227722,
      "num_tokens": 5137471.0,
      "step": 2320
    },
    {
      "entropy": 5.864839887619018,
      "epoch": 2.343923348461926,
      "grad_norm": 1.125,
      "learning_rate": 0.00047592472553812856,
      "loss": 5.4976,
      "mean_token_accuracy": 0.15462417900562286,
      "num_tokens": 5147682.0,
      "step": 2325
    },
    {
      "entropy": 5.76884274482727,
      "epoch": 2.3489662128088753,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004757458903722837,
      "loss": 5.4753,
      "mean_token_accuracy": 0.14941407442092897,
      "num_tokens": 5156143.0,
      "step": 2330
    },
    {
      "entropy": 5.833264446258545,
      "epoch": 2.3540090771558244,
      "grad_norm": 0.953125,
      "learning_rate": 0.00047556643128358075,
      "loss": 5.5411,
      "mean_token_accuracy": 0.1472172513604164,
      "num_tokens": 5167462.0,
      "step": 2335
    },
    {
      "entropy": 5.740017509460449,
      "epoch": 2.3590519415027735,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047538634882978277,
      "loss": 5.4618,
      "mean_token_accuracy": 0.1543852999806404,
      "num_tokens": 5178030.0,
      "step": 2340
    },
    {
      "entropy": 5.723574829101563,
      "epoch": 2.3640948058497226,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047520564357059013,
      "loss": 5.4704,
      "mean_token_accuracy": 0.1547468364238739,
      "num_tokens": 5188296.0,
      "step": 2345
    },
    {
      "entropy": 5.731347990036011,
      "epoch": 2.3691376701966718,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004750243160676389,
      "loss": 5.4746,
      "mean_token_accuracy": 0.15414796620607377,
      "num_tokens": 5199058.0,
      "step": 2350
    },
    {
      "entropy": 5.811427736282349,
      "epoch": 2.374180534543621,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004748423668844992,
      "loss": 5.438,
      "mean_token_accuracy": 0.15424609631299974,
      "num_tokens": 5209314.0,
      "step": 2355
    },
    {
      "entropy": 5.8259844303131105,
      "epoch": 2.37922339889057,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004746597965866733,
      "loss": 5.5481,
      "mean_token_accuracy": 0.14829962104558944,
      "num_tokens": 5220248.0,
      "step": 2360
    },
    {
      "entropy": 5.845902299880981,
      "epoch": 2.384266263237519,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004744766057415938,
      "loss": 5.5331,
      "mean_token_accuracy": 0.14199754148721694,
      "num_tokens": 5230874.0,
      "step": 2365
    },
    {
      "entropy": 5.827665996551514,
      "epoch": 2.389309127584468,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004742927949186222,
      "loss": 5.4588,
      "mean_token_accuracy": 0.15653351396322251,
      "num_tokens": 5243328.0,
      "step": 2370
    },
    {
      "entropy": 5.690488004684449,
      "epoch": 2.394351991931417,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00047410836468904664,
      "loss": 5.4266,
      "mean_token_accuracy": 0.15579296946525573,
      "num_tokens": 5254357.0,
      "step": 2375
    },
    {
      "entropy": 5.7096113681793215,
      "epoch": 2.399394856278366,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047392331562608073,
      "loss": 5.4264,
      "mean_token_accuracy": 0.15632745772600173,
      "num_tokens": 5264492.0,
      "step": 2380
    },
    {
      "entropy": 5.685121965408325,
      "epoch": 2.404437720625315,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004737376483048609,
      "loss": 5.3884,
      "mean_token_accuracy": 0.15759580880403518,
      "num_tokens": 5274816.0,
      "step": 2385
    },
    {
      "entropy": 5.7668869972229,
      "epoch": 2.4094805849722642,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004735513633024458,
      "loss": 5.4849,
      "mean_token_accuracy": 0.15257013440132142,
      "num_tokens": 5286855.0,
      "step": 2390
    },
    {
      "entropy": 5.889518117904663,
      "epoch": 2.4145234493192134,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004733644611978134,
      "loss": 5.5875,
      "mean_token_accuracy": 0.1414315164089203,
      "num_tokens": 5299594.0,
      "step": 2395
    },
    {
      "entropy": 5.700799989700317,
      "epoch": 2.4195663136661625,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004731769425718597,
      "loss": 5.4203,
      "mean_token_accuracy": 0.15246255695819855,
      "num_tokens": 5310078.0,
      "step": 2400
    },
    {
      "entropy": 5.721229791641235,
      "epoch": 2.4246091780131116,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000472988808007397,
      "loss": 5.4443,
      "mean_token_accuracy": 0.15930624157190323,
      "num_tokens": 5319544.0,
      "step": 2405
    },
    {
      "entropy": 5.765945100784302,
      "epoch": 2.4296520423600603,
      "grad_norm": 0.87890625,
      "learning_rate": 0.00047280005808915176,
      "loss": 5.5046,
      "mean_token_accuracy": 0.15074443221092224,
      "num_tokens": 5331025.0,
      "step": 2410
    },
    {
      "entropy": 5.679812240600586,
      "epoch": 2.4346949067070094,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004726106934037631,
      "loss": 5.4956,
      "mean_token_accuracy": 0.15793420970439911,
      "num_tokens": 5343086.0,
      "step": 2415
    },
    {
      "entropy": 5.918830823898316,
      "epoch": 2.4397377710539585,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004724207145397809,
      "loss": 5.5399,
      "mean_token_accuracy": 0.16036049127578736,
      "num_tokens": 5353598.0,
      "step": 2420
    },
    {
      "entropy": 5.717051362991333,
      "epoch": 2.4447806354009076,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047223012208766384,
      "loss": 5.4207,
      "mean_token_accuracy": 0.15672159790992737,
      "num_tokens": 5363607.0,
      "step": 2425
    },
    {
      "entropy": 5.703886842727661,
      "epoch": 2.4498234997478567,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004720389166397777,
      "loss": 5.4787,
      "mean_token_accuracy": 0.1559496819972992,
      "num_tokens": 5375198.0,
      "step": 2430
    },
    {
      "entropy": 5.766331672668457,
      "epoch": 2.454866364094806,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047184709879039325,
      "loss": 5.4638,
      "mean_token_accuracy": 0.15198483914136887,
      "num_tokens": 5386058.0,
      "step": 2435
    },
    {
      "entropy": 5.817217540740967,
      "epoch": 2.459909228441755,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047165466913568497,
      "loss": 5.4904,
      "mean_token_accuracy": 0.15327959060668944,
      "num_tokens": 5395720.0,
      "step": 2440
    },
    {
      "entropy": 5.766813945770264,
      "epoch": 2.464952092788704,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004714616282737285,
      "loss": 5.5328,
      "mean_token_accuracy": 0.15218988060951233,
      "num_tokens": 5406252.0,
      "step": 2445
    },
    {
      "entropy": 5.761837148666382,
      "epoch": 2.469994957135653,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004712679768044995,
      "loss": 5.5356,
      "mean_token_accuracy": 0.15231966972351074,
      "num_tokens": 5418125.0,
      "step": 2450
    },
    {
      "entropy": 5.808177661895752,
      "epoch": 2.4750378214826023,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000471073715329871,
      "loss": 5.4955,
      "mean_token_accuracy": 0.15355192571878434,
      "num_tokens": 5428095.0,
      "step": 2455
    },
    {
      "entropy": 5.794961404800415,
      "epoch": 2.4800806858295514,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004708788444536123,
      "loss": 5.5117,
      "mean_token_accuracy": 0.14688692539930343,
      "num_tokens": 5438359.0,
      "step": 2460
    },
    {
      "entropy": 5.774765396118164,
      "epoch": 2.4851235501765,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004706833647813865,
      "loss": 5.4519,
      "mean_token_accuracy": 0.15822555422782897,
      "num_tokens": 5448559.0,
      "step": 2465
    },
    {
      "entropy": 5.693092679977417,
      "epoch": 2.4901664145234492,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00047048727692074893,
      "loss": 5.4598,
      "mean_token_accuracy": 0.1524144560098648,
      "num_tokens": 5459739.0,
      "step": 2470
    },
    {
      "entropy": 5.736509847640991,
      "epoch": 2.4952092788703983,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004702905814811452,
      "loss": 5.4335,
      "mean_token_accuracy": 0.1499810516834259,
      "num_tokens": 5471160.0,
      "step": 2475
    },
    {
      "entropy": 5.663192415237427,
      "epoch": 2.5002521432173475,
      "grad_norm": 0.921875,
      "learning_rate": 0.00047009327907390926,
      "loss": 5.4951,
      "mean_token_accuracy": 0.15223399847745894,
      "num_tokens": 5482344.0,
      "step": 2480
    },
    {
      "entropy": 5.836965608596802,
      "epoch": 2.5052950075642966,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004698953703122616,
      "loss": 5.4962,
      "mean_token_accuracy": 0.1480371192097664,
      "num_tokens": 5493111.0,
      "step": 2485
    },
    {
      "entropy": 5.642283582687378,
      "epoch": 2.5103378719112457,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046969685581130715,
      "loss": 5.4298,
      "mean_token_accuracy": 0.15816318541765212,
      "num_tokens": 5503877.0,
      "step": 2490
    },
    {
      "entropy": 5.816074657440185,
      "epoch": 2.515380736258195,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046949773618803353,
      "loss": 5.4471,
      "mean_token_accuracy": 0.15494221150875093,
      "num_tokens": 5514067.0,
      "step": 2495
    },
    {
      "entropy": 5.782722473144531,
      "epoch": 2.5204236006051435,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00046929801206130925,
      "loss": 5.4982,
      "mean_token_accuracy": 0.15230851769447326,
      "num_tokens": 5525534.0,
      "step": 2500
    },
    {
      "epoch": 2.5204236006051435,
      "eval_entropy": 5.604792223631881,
      "eval_loss": 5.833621025085449,
      "eval_mean_token_accuracy": 0.14200868150432602,
      "eval_num_tokens": 5525534.0,
      "eval_runtime": 2.0419,
      "eval_samples_per_second": 1536.829,
      "eval_steps_per_second": 192.471,
      "step": 2500
    },
    {
      "entropy": 5.749931716918946,
      "epoch": 2.5254664649520926,
      "grad_norm": 1.25,
      "learning_rate": 0.00046909768405188135,
      "loss": 5.4682,
      "mean_token_accuracy": 0.16120100468397142,
      "num_tokens": 5535525.0,
      "step": 2505
    },
    {
      "entropy": 5.702026128768921,
      "epoch": 2.5305093292990417,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046889675278237404,
      "loss": 5.459,
      "mean_token_accuracy": 0.15884576141834258,
      "num_tokens": 5546129.0,
      "step": 2510
    },
    {
      "entropy": 5.846855926513672,
      "epoch": 2.535552193645991,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004686952188772863,
      "loss": 5.6149,
      "mean_token_accuracy": 0.14743217825889587,
      "num_tokens": 5557548.0,
      "step": 2515
    },
    {
      "entropy": 5.85679144859314,
      "epoch": 2.54059505799294,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046849308296299005,
      "loss": 5.5121,
      "mean_token_accuracy": 0.15738593488931657,
      "num_tokens": 5567367.0,
      "step": 2520
    },
    {
      "entropy": 5.769592523574829,
      "epoch": 2.545637922339889,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004682903456677285,
      "loss": 5.411,
      "mean_token_accuracy": 0.15898290127515793,
      "num_tokens": 5577489.0,
      "step": 2525
    },
    {
      "entropy": 5.722674417495727,
      "epoch": 2.550680786686838,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046808700762161385,
      "loss": 5.4755,
      "mean_token_accuracy": 0.1436998128890991,
      "num_tokens": 5587877.0,
      "step": 2530
    },
    {
      "entropy": 5.718273115158081,
      "epoch": 2.5557236510337873,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004678830694566253,
      "loss": 5.4653,
      "mean_token_accuracy": 0.15911817401647568,
      "num_tokens": 5597863.0,
      "step": 2535
    },
    {
      "entropy": 5.773120594024658,
      "epoch": 2.5607665153807364,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004676785318066075,
      "loss": 5.4847,
      "mean_token_accuracy": 0.15134269148111343,
      "num_tokens": 5608141.0,
      "step": 2540
    },
    {
      "entropy": 5.718518304824829,
      "epoch": 2.5658093797276855,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004674733953072682,
      "loss": 5.5629,
      "mean_token_accuracy": 0.15045879259705544,
      "num_tokens": 5619660.0,
      "step": 2545
    },
    {
      "entropy": 5.894439649581909,
      "epoch": 2.5708522440746346,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004672676605961763,
      "loss": 5.538,
      "mean_token_accuracy": 0.14633310437202454,
      "num_tokens": 5630271.0,
      "step": 2550
    },
    {
      "entropy": 5.766078233718872,
      "epoch": 2.5758951084215833,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00046706132831276007,
      "loss": 5.5411,
      "mean_token_accuracy": 0.15482994318008422,
      "num_tokens": 5642594.0,
      "step": 2555
    },
    {
      "entropy": 5.713948917388916,
      "epoch": 2.5809379727685324,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046685439909830486,
      "loss": 5.4676,
      "mean_token_accuracy": 0.15393735021352767,
      "num_tokens": 5653361.0,
      "step": 2560
    },
    {
      "entropy": 5.690238904953003,
      "epoch": 2.5859808371154815,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004666468735959516,
      "loss": 5.3863,
      "mean_token_accuracy": 0.1587365075945854,
      "num_tokens": 5664167.0,
      "step": 2565
    },
    {
      "entropy": 5.730501413345337,
      "epoch": 2.5910237014624307,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004664387524506943,
      "loss": 5.4791,
      "mean_token_accuracy": 0.15582175850868224,
      "num_tokens": 5675903.0,
      "step": 2570
    },
    {
      "entropy": 5.746599721908569,
      "epoch": 2.59606656580938,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004662300363093784,
      "loss": 5.4612,
      "mean_token_accuracy": 0.15438395291566848,
      "num_tokens": 5686319.0,
      "step": 2575
    },
    {
      "entropy": 5.866937828063965,
      "epoch": 2.601109430156329,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004660207258206983,
      "loss": 5.557,
      "mean_token_accuracy": 0.14871260821819304,
      "num_tokens": 5696000.0,
      "step": 2580
    },
    {
      "entropy": 5.719299936294556,
      "epoch": 2.606152294503278,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00046581082163519585,
      "loss": 5.4662,
      "mean_token_accuracy": 0.1533816337585449,
      "num_tokens": 5708145.0,
      "step": 2585
    },
    {
      "entropy": 5.65442476272583,
      "epoch": 2.6111951588502267,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004656003244052582,
      "loss": 5.4203,
      "mean_token_accuracy": 0.16013493090867997,
      "num_tokens": 5718053.0,
      "step": 2590
    },
    {
      "entropy": 5.770934581756592,
      "epoch": 2.616238023197176,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004653892347851155,
      "loss": 5.5033,
      "mean_token_accuracy": 0.15183904767036438,
      "num_tokens": 5728723.0,
      "step": 2595
    },
    {
      "entropy": 5.6843973159790036,
      "epoch": 2.621280887544125,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004651775534308392,
      "loss": 5.4217,
      "mean_token_accuracy": 0.1600257158279419,
      "num_tokens": 5739414.0,
      "step": 2600
    },
    {
      "entropy": 5.790118026733398,
      "epoch": 2.626323751891074,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004649652810003398,
      "loss": 5.5691,
      "mean_token_accuracy": 0.14777129888534546,
      "num_tokens": 5751247.0,
      "step": 2605
    },
    {
      "entropy": 5.715647268295288,
      "epoch": 2.631366616238023,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00046475241815336504,
      "loss": 5.4329,
      "mean_token_accuracy": 0.1596490576863289,
      "num_tokens": 5761791.0,
      "step": 2610
    },
    {
      "entropy": 5.659279775619507,
      "epoch": 2.6364094805849723,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004645389655514974,
      "loss": 5.4166,
      "mean_token_accuracy": 0.15218251794576645,
      "num_tokens": 5773597.0,
      "step": 2615
    },
    {
      "entropy": 5.74836916923523,
      "epoch": 2.6414523449319214,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004643249238581525,
      "loss": 5.4435,
      "mean_token_accuracy": 0.15350425690412522,
      "num_tokens": 5782963.0,
      "step": 2620
    },
    {
      "entropy": 5.681864070892334,
      "epoch": 2.6464952092788705,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004641102937385769,
      "loss": 5.4655,
      "mean_token_accuracy": 0.15417479425668718,
      "num_tokens": 5793474.0,
      "step": 2625
    },
    {
      "entropy": 5.750307655334472,
      "epoch": 2.6515380736258196,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004638950758598461,
      "loss": 5.4604,
      "mean_token_accuracy": 0.15467366874217986,
      "num_tokens": 5804018.0,
      "step": 2630
    },
    {
      "entropy": 5.777842712402344,
      "epoch": 2.6565809379727687,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046367927089086203,
      "loss": 5.4862,
      "mean_token_accuracy": 0.1538143202662468,
      "num_tokens": 5815897.0,
      "step": 2635
    },
    {
      "entropy": 5.73519458770752,
      "epoch": 2.661623802319718,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046346287950235163,
      "loss": 5.5184,
      "mean_token_accuracy": 0.15442679822444916,
      "num_tokens": 5825883.0,
      "step": 2640
    },
    {
      "entropy": 5.790557098388672,
      "epoch": 2.6666666666666665,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004632459023668642,
      "loss": 5.4842,
      "mean_token_accuracy": 0.15851491093635559,
      "num_tokens": 5836668.0,
      "step": 2645
    },
    {
      "entropy": 5.824594211578369,
      "epoch": 2.6717095310136156,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004630283401587699,
      "loss": 5.6062,
      "mean_token_accuracy": 0.1482620917260647,
      "num_tokens": 5848497.0,
      "step": 2650
    },
    {
      "entropy": 5.6708885669708256,
      "epoch": 2.6767523953605648,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004628101935542568,
      "loss": 5.4177,
      "mean_token_accuracy": 0.16152112931013107,
      "num_tokens": 5859231.0,
      "step": 2655
    },
    {
      "entropy": 5.7441081523895265,
      "epoch": 2.681795259707514,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046259146323132986,
      "loss": 5.5636,
      "mean_token_accuracy": 0.1503380134701729,
      "num_tokens": 5871190.0,
      "step": 2660
    },
    {
      "entropy": 5.745948123931885,
      "epoch": 2.686838124054463,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046237214986980776,
      "loss": 5.4542,
      "mean_token_accuracy": 0.15525057166814804,
      "num_tokens": 5881788.0,
      "step": 2665
    },
    {
      "entropy": 5.73726978302002,
      "epoch": 2.691880988401412,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004621522541513216,
      "loss": 5.4179,
      "mean_token_accuracy": 0.15737532079219818,
      "num_tokens": 5891289.0,
      "step": 2670
    },
    {
      "entropy": 5.667320251464844,
      "epoch": 2.696923852748361,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004619317767593123,
      "loss": 5.4592,
      "mean_token_accuracy": 0.15221125036478042,
      "num_tokens": 5903594.0,
      "step": 2675
    },
    {
      "entropy": 5.780694246292114,
      "epoch": 2.70196671709531,
      "grad_norm": 0.96875,
      "learning_rate": 0.00046171071837902873,
      "loss": 5.5686,
      "mean_token_accuracy": 0.15084668695926667,
      "num_tokens": 5914592.0,
      "step": 2680
    },
    {
      "entropy": 5.66141676902771,
      "epoch": 2.707009581442259,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004614890796975254,
      "loss": 5.3945,
      "mean_token_accuracy": 0.15323598980903624,
      "num_tokens": 5927137.0,
      "step": 2685
    },
    {
      "entropy": 5.736548376083374,
      "epoch": 2.712052445789208,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004612668614036605,
      "loss": 5.5062,
      "mean_token_accuracy": 0.15607298016548157,
      "num_tokens": 5938730.0,
      "step": 2690
    },
    {
      "entropy": 5.748313045501709,
      "epoch": 2.7170953101361572,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004610440641880934,
      "loss": 5.5129,
      "mean_token_accuracy": 0.15209259688854218,
      "num_tokens": 5948912.0,
      "step": 2695
    },
    {
      "entropy": 5.643487167358399,
      "epoch": 2.7221381744831064,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046082068874328324,
      "loss": 5.4538,
      "mean_token_accuracy": 0.15674931257963182,
      "num_tokens": 5960253.0,
      "step": 2700
    },
    {
      "entropy": 5.782401037216187,
      "epoch": 2.7271810388300555,
      "grad_norm": 1.125,
      "learning_rate": 0.00046059673576348596,
      "loss": 5.5407,
      "mean_token_accuracy": 0.1502511218190193,
      "num_tokens": 5972615.0,
      "step": 2705
    },
    {
      "entropy": 5.8221601963043215,
      "epoch": 2.7322239031770046,
      "grad_norm": 1.0,
      "learning_rate": 0.00046037220594475264,
      "loss": 5.5235,
      "mean_token_accuracy": 0.15435368418693543,
      "num_tokens": 5985023.0,
      "step": 2710
    },
    {
      "entropy": 5.68713059425354,
      "epoch": 2.7372667675239537,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000460147099984927,
      "loss": 5.412,
      "mean_token_accuracy": 0.16228729784488677,
      "num_tokens": 5996445.0,
      "step": 2715
    },
    {
      "entropy": 5.6852774143219,
      "epoch": 2.742309631870903,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004599214185836438,
      "loss": 5.4575,
      "mean_token_accuracy": 0.15530781745910643,
      "num_tokens": 6007385.0,
      "step": 2720
    },
    {
      "entropy": 5.723224115371704,
      "epoch": 2.747352496217852,
      "grad_norm": 1.015625,
      "learning_rate": 0.00045969516244232574,
      "loss": 5.502,
      "mean_token_accuracy": 0.15536753088235855,
      "num_tokens": 6018174.0,
      "step": 2725
    },
    {
      "entropy": 5.66727843284607,
      "epoch": 2.752395360564801,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00045946833226418235,
      "loss": 5.4224,
      "mean_token_accuracy": 0.15106706768274308,
      "num_tokens": 6030049.0,
      "step": 2730
    },
    {
      "entropy": 5.689644384384155,
      "epoch": 2.7574382249117497,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004592409287542069,
      "loss": 5.4752,
      "mean_token_accuracy": 0.15385230332612992,
      "num_tokens": 6040155.0,
      "step": 2735
    },
    {
      "entropy": 5.641754293441773,
      "epoch": 2.762481089258699,
      "grad_norm": 1.125,
      "learning_rate": 0.0004590129526191748,
      "loss": 5.4713,
      "mean_token_accuracy": 0.1578552782535553,
      "num_tokens": 6050569.0,
      "step": 2740
    },
    {
      "entropy": 5.757769966125489,
      "epoch": 2.767523953605648,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004587844045676413,
      "loss": 5.406,
      "mean_token_accuracy": 0.15831348299980164,
      "num_tokens": 6060871.0,
      "step": 2745
    },
    {
      "entropy": 5.69254059791565,
      "epoch": 2.772566817952597,
      "grad_norm": 1.0,
      "learning_rate": 0.00045855528530993874,
      "loss": 5.4325,
      "mean_token_accuracy": 0.16034068167209625,
      "num_tokens": 6073045.0,
      "step": 2750
    },
    {
      "entropy": 5.713394403457642,
      "epoch": 2.777609682299546,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045832559555817525,
      "loss": 5.4809,
      "mean_token_accuracy": 0.15353266894817352,
      "num_tokens": 6084798.0,
      "step": 2755
    },
    {
      "entropy": 5.721564102172851,
      "epoch": 2.7826525466464953,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045809533602623185,
      "loss": 5.4029,
      "mean_token_accuracy": 0.1530379071831703,
      "num_tokens": 6095831.0,
      "step": 2760
    },
    {
      "entropy": 5.671160078048706,
      "epoch": 2.7876954109934444,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004578645074297603,
      "loss": 5.3846,
      "mean_token_accuracy": 0.1597864434123039,
      "num_tokens": 6106958.0,
      "step": 2765
    },
    {
      "entropy": 5.682317733764648,
      "epoch": 2.792738275340393,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045763311048618144,
      "loss": 5.4219,
      "mean_token_accuracy": 0.16095105409622193,
      "num_tokens": 6118165.0,
      "step": 2770
    },
    {
      "entropy": 5.596311283111572,
      "epoch": 2.797781139687342,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004574011459146822,
      "loss": 5.4184,
      "mean_token_accuracy": 0.163133005797863,
      "num_tokens": 6129297.0,
      "step": 2775
    },
    {
      "entropy": 5.718992757797241,
      "epoch": 2.8028240040342913,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00045716861443621386,
      "loss": 5.4535,
      "mean_token_accuracy": 0.1543310895562172,
      "num_tokens": 6140071.0,
      "step": 2780
    },
    {
      "entropy": 5.710288763046265,
      "epoch": 2.8078668683812404,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045693551677348967,
      "loss": 5.4461,
      "mean_token_accuracy": 0.15649698227643966,
      "num_tokens": 6150940.0,
      "step": 2785
    },
    {
      "entropy": 5.71898922920227,
      "epoch": 2.8129097327281896,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004567018536509823,
      "loss": 5.4294,
      "mean_token_accuracy": 0.15092435628175735,
      "num_tokens": 6161966.0,
      "step": 2790
    },
    {
      "entropy": 5.756949090957642,
      "epoch": 2.8179525970751387,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00045646762579492245,
      "loss": 5.4816,
      "mean_token_accuracy": 0.15153277218341826,
      "num_tokens": 6174455.0,
      "step": 2795
    },
    {
      "entropy": 5.679877471923828,
      "epoch": 2.822995461422088,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045623283393329546,
      "loss": 5.4191,
      "mean_token_accuracy": 0.1543664038181305,
      "num_tokens": 6184789.0,
      "step": 2800
    },
    {
      "entropy": 5.666750240325928,
      "epoch": 2.828038325769037,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045599747879584,
      "loss": 5.5097,
      "mean_token_accuracy": 0.1503747209906578,
      "num_tokens": 6195704.0,
      "step": 2805
    },
    {
      "entropy": 5.745576047897339,
      "epoch": 2.833081190115986,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004557615611140451,
      "loss": 5.3787,
      "mean_token_accuracy": 0.1610755890607834,
      "num_tokens": 6206290.0,
      "step": 2810
    },
    {
      "entropy": 5.727213716506958,
      "epoch": 2.838124054462935,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004555250816211485,
      "loss": 5.5041,
      "mean_token_accuracy": 0.14910117164254189,
      "num_tokens": 6216660.0,
      "step": 2815
    },
    {
      "entropy": 5.685876321792603,
      "epoch": 2.8431669188098843,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00045528804105213395,
      "loss": 5.4425,
      "mean_token_accuracy": 0.1630454868078232,
      "num_tokens": 6228005.0,
      "step": 2820
    },
    {
      "entropy": 5.686995935440064,
      "epoch": 2.848209783156833,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004550504401437289,
      "loss": 5.4042,
      "mean_token_accuracy": 0.1551094740629196,
      "num_tokens": 6239447.0,
      "step": 2825
    },
    {
      "entropy": 5.6741846084594725,
      "epoch": 2.853252647503782,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00045481227963440235,
      "loss": 5.5118,
      "mean_token_accuracy": 0.1445085808634758,
      "num_tokens": 6252312.0,
      "step": 2830
    },
    {
      "entropy": 5.715362453460694,
      "epoch": 2.858295511850731,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045457356026436295,
      "loss": 5.4484,
      "mean_token_accuracy": 0.15417277663946152,
      "num_tokens": 6264131.0,
      "step": 2835
    },
    {
      "entropy": 5.657751035690308,
      "epoch": 2.8633383761976803,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00045433428277555577,
      "loss": 5.3992,
      "mean_token_accuracy": 0.15836391001939773,
      "num_tokens": 6275161.0,
      "step": 2840
    },
    {
      "entropy": 5.6258327007293705,
      "epoch": 2.8683812405446294,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004540944479116609,
      "loss": 5.4533,
      "mean_token_accuracy": 0.15344715863466263,
      "num_tokens": 6285655.0,
      "step": 2845
    },
    {
      "entropy": 5.7377156734466555,
      "epoch": 2.8734241048915785,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004538540564180906,
      "loss": 5.5431,
      "mean_token_accuracy": 0.15740798711776732,
      "num_tokens": 6296532.0,
      "step": 2850
    },
    {
      "entropy": 5.845269393920899,
      "epoch": 2.8784669692385276,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004536131090419871,
      "loss": 5.5612,
      "mean_token_accuracy": 0.15288513749837876,
      "num_tokens": 6308229.0,
      "step": 2855
    },
    {
      "entropy": 5.691622352600097,
      "epoch": 2.8835098335854763,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004533716065322205,
      "loss": 5.4968,
      "mean_token_accuracy": 0.15628847032785415,
      "num_tokens": 6321257.0,
      "step": 2860
    },
    {
      "entropy": 5.747979593276978,
      "epoch": 2.8885526979324254,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004531295496393861,
      "loss": 5.4294,
      "mean_token_accuracy": 0.1566181883215904,
      "num_tokens": 6332040.0,
      "step": 2865
    },
    {
      "entropy": 5.609353876113891,
      "epoch": 2.8935955622793745,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004528869391158023,
      "loss": 5.3297,
      "mean_token_accuracy": 0.16526031941175462,
      "num_tokens": 6342925.0,
      "step": 2870
    },
    {
      "entropy": 5.692199850082398,
      "epoch": 2.8986384266263236,
      "grad_norm": 0.8984375,
      "learning_rate": 0.00045264377571550834,
      "loss": 5.4941,
      "mean_token_accuracy": 0.15457060635089875,
      "num_tokens": 6354263.0,
      "step": 2875
    },
    {
      "entropy": 5.7853111743927,
      "epoch": 2.9036812909732728,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004524000601942615,
      "loss": 5.4995,
      "mean_token_accuracy": 0.1555572748184204,
      "num_tokens": 6365497.0,
      "step": 2880
    },
    {
      "entropy": 5.747455787658692,
      "epoch": 2.908724155320222,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004521557933095354,
      "loss": 5.5383,
      "mean_token_accuracy": 0.14962867200374602,
      "num_tokens": 6376501.0,
      "step": 2885
    },
    {
      "entropy": 5.738136625289917,
      "epoch": 2.913767019667171,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000451910975820517,
      "loss": 5.4828,
      "mean_token_accuracy": 0.1548456773161888,
      "num_tokens": 6387632.0,
      "step": 2890
    },
    {
      "entropy": 5.628865480422974,
      "epoch": 2.91880988401412,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045166560848810476,
      "loss": 5.471,
      "mean_token_accuracy": 0.15752095729112625,
      "num_tokens": 6398813.0,
      "step": 2895
    },
    {
      "entropy": 5.785012340545654,
      "epoch": 2.9238527483610692,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045141969207490606,
      "loss": 5.3888,
      "mean_token_accuracy": 0.158993498980999,
      "num_tokens": 6409532.0,
      "step": 2900
    },
    {
      "entropy": 5.699265241622925,
      "epoch": 2.9288956127080183,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004511732273452347,
      "loss": 5.5195,
      "mean_token_accuracy": 0.14928267449140548,
      "num_tokens": 6420755.0,
      "step": 2905
    },
    {
      "entropy": 5.746183252334594,
      "epoch": 2.9339384770549675,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004509262150651089,
      "loss": 5.5535,
      "mean_token_accuracy": 0.15794235318899155,
      "num_tokens": 6430662.0,
      "step": 2910
    },
    {
      "entropy": 5.706696224212647,
      "epoch": 2.938981341401916,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045067865600224833,
      "loss": 5.4288,
      "mean_token_accuracy": 0.16091673374176024,
      "num_tokens": 6441230.0,
      "step": 2915
    },
    {
      "entropy": 5.776336336135865,
      "epoch": 2.9440242057488653,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004504305509260724,
      "loss": 5.5645,
      "mean_token_accuracy": 0.1445389173924923,
      "num_tokens": 6452574.0,
      "step": 2920
    },
    {
      "entropy": 5.690360641479492,
      "epoch": 2.9490670700958144,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004501819006076974,
      "loss": 5.4229,
      "mean_token_accuracy": 0.16237369179725647,
      "num_tokens": 6462562.0,
      "step": 2925
    },
    {
      "entropy": 5.652652740478516,
      "epoch": 2.9541099344427635,
      "grad_norm": 1.03125,
      "learning_rate": 0.00044993270581993413,
      "loss": 5.4418,
      "mean_token_accuracy": 0.15699686706066132,
      "num_tokens": 6473594.0,
      "step": 2930
    },
    {
      "entropy": 5.707343578338623,
      "epoch": 2.9591527987897126,
      "grad_norm": 1.0,
      "learning_rate": 0.00044968296733728587,
      "loss": 5.4492,
      "mean_token_accuracy": 0.15589323341846467,
      "num_tokens": 6484129.0,
      "step": 2935
    },
    {
      "entropy": 5.731469249725341,
      "epoch": 2.9641956631366617,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004494326859359456,
      "loss": 5.4302,
      "mean_token_accuracy": 0.1545371010899544,
      "num_tokens": 6495584.0,
      "step": 2940
    },
    {
      "entropy": 5.679045915603638,
      "epoch": 2.969238527483611,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044918186239379363,
      "loss": 5.5132,
      "mean_token_accuracy": 0.1568527489900589,
      "num_tokens": 6507872.0,
      "step": 2945
    },
    {
      "entropy": 5.684495162963867,
      "epoch": 2.9742813918305595,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004489304974903953,
      "loss": 5.3887,
      "mean_token_accuracy": 0.16150760352611543,
      "num_tokens": 6519172.0,
      "step": 2950
    },
    {
      "entropy": 5.625805950164795,
      "epoch": 2.9793242561775086,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004486785920069985,
      "loss": 5.4838,
      "mean_token_accuracy": 0.15884941071271896,
      "num_tokens": 6530151.0,
      "step": 2955
    },
    {
      "entropy": 5.805325746536255,
      "epoch": 2.9843671205244577,
      "grad_norm": 1.0,
      "learning_rate": 0.0004484261467265313,
      "loss": 5.4723,
      "mean_token_accuracy": 0.157195083796978,
      "num_tokens": 6541881.0,
      "step": 2960
    },
    {
      "entropy": 5.6935382843017575,
      "epoch": 2.989409984871407,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00044817316243359956,
      "loss": 5.4831,
      "mean_token_accuracy": 0.15258170664310455,
      "num_tokens": 6553390.0,
      "step": 2965
    },
    {
      "entropy": 5.648685073852539,
      "epoch": 2.994452849218356,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044791963991448427,
      "loss": 5.3176,
      "mean_token_accuracy": 0.16544754505157472,
      "num_tokens": 6563754.0,
      "step": 2970
    },
    {
      "entropy": 5.661775827407837,
      "epoch": 2.999495713565305,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00044766557995713915,
      "loss": 5.3993,
      "mean_token_accuracy": 0.15286386609077454,
      "num_tokens": 6574124.0,
      "step": 2975
    },
    {
      "entropy": 5.706845389472114,
      "epoch": 3.0040342914775593,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004474109833511885,
      "loss": 5.2217,
      "mean_token_accuracy": 0.16955596705277762,
      "num_tokens": 6582795.0,
      "step": 2980
    },
    {
      "entropy": 5.622677898406982,
      "epoch": 3.0090771558245084,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004471558508879244,
      "loss": 5.1452,
      "mean_token_accuracy": 0.1762877270579338,
      "num_tokens": 6593649.0,
      "step": 2985
    },
    {
      "entropy": 5.6546000957489015,
      "epoch": 3.0141200201714575,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004469001833603045,
      "loss": 5.1699,
      "mean_token_accuracy": 0.1765150845050812,
      "num_tokens": 6606370.0,
      "step": 2990
    },
    {
      "entropy": 5.576964712142944,
      "epoch": 3.0191628845184066,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004466439815629493,
      "loss": 5.099,
      "mean_token_accuracy": 0.17777346521615983,
      "num_tokens": 6616599.0,
      "step": 2995
    },
    {
      "entropy": 5.7396176815032955,
      "epoch": 3.0242057488653553,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044638724629214,
      "loss": 5.1765,
      "mean_token_accuracy": 0.17028990685939788,
      "num_tokens": 6627707.0,
      "step": 3000
    },
    {
      "epoch": 3.0242057488653553,
      "eval_entropy": 5.477556743694626,
      "eval_loss": 5.762594699859619,
      "eval_mean_token_accuracy": 0.1469519197485829,
      "eval_num_tokens": 6627707.0,
      "eval_runtime": 2.0848,
      "eval_samples_per_second": 1505.212,
      "eval_steps_per_second": 188.511,
      "step": 3000
    },
    {
      "entropy": 5.611760854721069,
      "epoch": 3.0292486132123044,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044612997834581557,
      "loss": 5.0985,
      "mean_token_accuracy": 0.17331636250019072,
      "num_tokens": 6639001.0,
      "step": 3005
    },
    {
      "entropy": 5.631704807281494,
      "epoch": 3.0342914775592535,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044587217852357086,
      "loss": 5.1754,
      "mean_token_accuracy": 0.16776107549667357,
      "num_tokens": 6649318.0,
      "step": 3010
    },
    {
      "entropy": 5.622303533554077,
      "epoch": 3.0393343419062027,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044561384762665363,
      "loss": 5.2094,
      "mean_token_accuracy": 0.1668701246380806,
      "num_tokens": 6660584.0,
      "step": 3015
    },
    {
      "entropy": 5.564846563339233,
      "epoch": 3.044377206253152,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004453549864579623,
      "loss": 5.0376,
      "mean_token_accuracy": 0.18073629438877106,
      "num_tokens": 6671361.0,
      "step": 3020
    },
    {
      "entropy": 5.5941565990447994,
      "epoch": 3.049420070600101,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004450955958220433,
      "loss": 5.2736,
      "mean_token_accuracy": 0.16895659565925597,
      "num_tokens": 6681896.0,
      "step": 3025
    },
    {
      "entropy": 5.659935808181762,
      "epoch": 3.05446293494705,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00044483567652508864,
      "loss": 5.1274,
      "mean_token_accuracy": 0.1717680051922798,
      "num_tokens": 6693360.0,
      "step": 3030
    },
    {
      "entropy": 5.600113582611084,
      "epoch": 3.059505799293999,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004445752293749337,
      "loss": 5.2024,
      "mean_token_accuracy": 0.16761997789144517,
      "num_tokens": 6704921.0,
      "step": 3035
    },
    {
      "entropy": 5.60845799446106,
      "epoch": 3.0645486636409482,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004443142551810541,
      "loss": 5.1565,
      "mean_token_accuracy": 0.17471077293157578,
      "num_tokens": 6715659.0,
      "step": 3040
    },
    {
      "entropy": 5.570842123031616,
      "epoch": 3.069591527987897,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044405275475456365,
      "loss": 5.0635,
      "mean_token_accuracy": 0.18722012341022493,
      "num_tokens": 6726431.0,
      "step": 3045
    },
    {
      "entropy": 5.573184156417847,
      "epoch": 3.074634392334846,
      "grad_norm": 1.0,
      "learning_rate": 0.0004437907289082116,
      "loss": 5.16,
      "mean_token_accuracy": 0.17500911206007003,
      "num_tokens": 6738002.0,
      "step": 3050
    },
    {
      "entropy": 5.627092695236206,
      "epoch": 3.079677256681795,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044352817845638053,
      "loss": 5.1485,
      "mean_token_accuracy": 0.17743653804063797,
      "num_tokens": 6748509.0,
      "step": 3055
    },
    {
      "entropy": 5.6118231296539305,
      "epoch": 3.0847201210287443,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004432651042150833,
      "loss": 5.1887,
      "mean_token_accuracy": 0.16978248804807664,
      "num_tokens": 6759258.0,
      "step": 3060
    },
    {
      "entropy": 5.578791809082031,
      "epoch": 3.0897629853756934,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044300150700196054,
      "loss": 5.177,
      "mean_token_accuracy": 0.1758012890815735,
      "num_tokens": 6771263.0,
      "step": 3065
    },
    {
      "entropy": 5.637490224838257,
      "epoch": 3.0948058497226425,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004427373876362786,
      "loss": 5.1532,
      "mean_token_accuracy": 0.17140825390815734,
      "num_tokens": 6779965.0,
      "step": 3070
    },
    {
      "entropy": 5.631720876693725,
      "epoch": 3.0998487140695916,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004424727469389265,
      "loss": 5.121,
      "mean_token_accuracy": 0.17350034415721893,
      "num_tokens": 6791521.0,
      "step": 3075
    },
    {
      "entropy": 5.645379304885864,
      "epoch": 3.1048915784165407,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004422075857324137,
      "loss": 5.2105,
      "mean_token_accuracy": 0.17218986302614211,
      "num_tokens": 6803537.0,
      "step": 3080
    },
    {
      "entropy": 5.647880697250367,
      "epoch": 3.10993444276349,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004419419048408673,
      "loss": 5.2151,
      "mean_token_accuracy": 0.16766090840101242,
      "num_tokens": 6813811.0,
      "step": 3085
    },
    {
      "entropy": 5.605061101913452,
      "epoch": 3.1149773071104385,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004416757050900297,
      "loss": 5.1232,
      "mean_token_accuracy": 0.1749053418636322,
      "num_tokens": 6825422.0,
      "step": 3090
    },
    {
      "entropy": 5.693388032913208,
      "epoch": 3.1200201714573876,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004414089873072557,
      "loss": 5.2764,
      "mean_token_accuracy": 0.16573522090911866,
      "num_tokens": 6837086.0,
      "step": 3095
    },
    {
      "entropy": 5.594672918319702,
      "epoch": 3.1250630358043368,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004411417523215105,
      "loss": 5.1127,
      "mean_token_accuracy": 0.17700614482164384,
      "num_tokens": 6847195.0,
      "step": 3100
    },
    {
      "entropy": 5.553398466110229,
      "epoch": 3.130105900151286,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004408740009633665,
      "loss": 5.1174,
      "mean_token_accuracy": 0.1782492682337761,
      "num_tokens": 6857492.0,
      "step": 3105
    },
    {
      "entropy": 5.618039608001709,
      "epoch": 3.135148764498235,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00044060573406500117,
      "loss": 5.3058,
      "mean_token_accuracy": 0.16275914758443832,
      "num_tokens": 6869190.0,
      "step": 3110
    },
    {
      "entropy": 5.651626443862915,
      "epoch": 3.140191628845184,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004403369524601941,
      "loss": 5.1078,
      "mean_token_accuracy": 0.17300194799900054,
      "num_tokens": 6880728.0,
      "step": 3115
    },
    {
      "entropy": 5.5053863525390625,
      "epoch": 3.145234493192133,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004400676569843249,
      "loss": 5.1337,
      "mean_token_accuracy": 0.16755828112363816,
      "num_tokens": 6892043.0,
      "step": 3120
    },
    {
      "entropy": 5.590396451950073,
      "epoch": 3.1502773575390823,
      "grad_norm": 1.28125,
      "learning_rate": 0.00043979784847436994,
      "loss": 5.1723,
      "mean_token_accuracy": 0.17638254165649414,
      "num_tokens": 6903913.0,
      "step": 3125
    },
    {
      "entropy": 5.5424778938293455,
      "epoch": 3.1553202218860315,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004395275277689004,
      "loss": 5.1301,
      "mean_token_accuracy": 0.1695476934313774,
      "num_tokens": 6913951.0,
      "step": 3130
    },
    {
      "entropy": 5.532778644561768,
      "epoch": 3.16036308623298,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004392566957080793,
      "loss": 5.1439,
      "mean_token_accuracy": 0.1718327820301056,
      "num_tokens": 6924762.0,
      "step": 3135
    },
    {
      "entropy": 5.557773685455322,
      "epoch": 3.1654059505799292,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043898535313365884,
      "loss": 5.139,
      "mean_token_accuracy": 0.17137170284986497,
      "num_tokens": 6935845.0,
      "step": 3140
    },
    {
      "entropy": 5.589786767959595,
      "epoch": 3.1704488149268784,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000438713500888978,
      "loss": 5.1672,
      "mean_token_accuracy": 0.1767355114221573,
      "num_tokens": 6947338.0,
      "step": 3145
    },
    {
      "entropy": 5.6020989418029785,
      "epoch": 3.1754916792738275,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004384411398189599,
      "loss": 5.1875,
      "mean_token_accuracy": 0.17276835441589355,
      "num_tokens": 6958084.0,
      "step": 3150
    },
    {
      "entropy": 5.574871301651001,
      "epoch": 3.1805345436207766,
      "grad_norm": 1.109375,
      "learning_rate": 0.000438168270770109,
      "loss": 5.1262,
      "mean_token_accuracy": 0.1767292559146881,
      "num_tokens": 6968063.0,
      "step": 3155
    },
    {
      "entropy": 5.599155235290527,
      "epoch": 3.1855774079677257,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004378948945905085,
      "loss": 5.1235,
      "mean_token_accuracy": 0.17755823731422424,
      "num_tokens": 6978424.0,
      "step": 3160
    },
    {
      "entropy": 5.506662321090698,
      "epoch": 3.190620272314675,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00043762101212981805,
      "loss": 5.1169,
      "mean_token_accuracy": 0.17420441955327987,
      "num_tokens": 6991067.0,
      "step": 3165
    },
    {
      "entropy": 5.538624382019043,
      "epoch": 3.195663136661624,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043734662423927054,
      "loss": 5.2022,
      "mean_token_accuracy": 0.17706032246351242,
      "num_tokens": 7000642.0,
      "step": 3170
    },
    {
      "entropy": 5.6724512577056885,
      "epoch": 3.200706001008573,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004370717317716699,
      "loss": 5.2334,
      "mean_token_accuracy": 0.16977518647909165,
      "num_tokens": 7011305.0,
      "step": 3175
    },
    {
      "entropy": 5.618666172027588,
      "epoch": 3.205748865355522,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004367963355813882,
      "loss": 5.1867,
      "mean_token_accuracy": 0.1626639649271965,
      "num_tokens": 7021452.0,
      "step": 3180
    },
    {
      "entropy": 5.57020058631897,
      "epoch": 3.210791729702471,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043652043652436323,
      "loss": 5.2418,
      "mean_token_accuracy": 0.16256314665079116,
      "num_tokens": 7033189.0,
      "step": 3185
    },
    {
      "entropy": 5.635352611541748,
      "epoch": 3.21583459404942,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004362440354580956,
      "loss": 5.2009,
      "mean_token_accuracy": 0.16863385289907457,
      "num_tokens": 7044153.0,
      "step": 3190
    },
    {
      "entropy": 5.592417240142822,
      "epoch": 3.220877458396369,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004359671332416462,
      "loss": 5.2386,
      "mean_token_accuracy": 0.16535451859235764,
      "num_tokens": 7056679.0,
      "step": 3195
    },
    {
      "entropy": 5.622775030136109,
      "epoch": 3.225920322743318,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004356897307356335,
      "loss": 5.1083,
      "mean_token_accuracy": 0.18078520894050598,
      "num_tokens": 7067428.0,
      "step": 3200
    },
    {
      "entropy": 5.592712020874023,
      "epoch": 3.2309631870902673,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004354118288022309,
      "loss": 5.2212,
      "mean_token_accuracy": 0.16834693253040314,
      "num_tokens": 7079196.0,
      "step": 3205
    },
    {
      "entropy": 5.603639268875122,
      "epoch": 3.2360060514372164,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043513342830516405,
      "loss": 5.1174,
      "mean_token_accuracy": 0.17468537837266923,
      "num_tokens": 7088936.0,
      "step": 3210
    },
    {
      "entropy": 5.571869564056397,
      "epoch": 3.2410489157841655,
      "grad_norm": 0.9296875,
      "learning_rate": 0.000434854530109708,
      "loss": 5.2178,
      "mean_token_accuracy": 0.17067920714616774,
      "num_tokens": 7101504.0,
      "step": 3215
    },
    {
      "entropy": 5.613890600204468,
      "epoch": 3.2460917801311147,
      "grad_norm": 1.0,
      "learning_rate": 0.00043457513508268487,
      "loss": 5.1964,
      "mean_token_accuracy": 0.17143479585647584,
      "num_tokens": 7114121.0,
      "step": 3220
    },
    {
      "entropy": 5.485863590240479,
      "epoch": 3.2511346444780633,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043429524409246083,
      "loss": 5.095,
      "mean_token_accuracy": 0.16754829734563828,
      "num_tokens": 7123321.0,
      "step": 3225
    },
    {
      "entropy": 5.544781732559204,
      "epoch": 3.2561775088250124,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004340148580089436,
      "loss": 5.1669,
      "mean_token_accuracy": 0.17092314809560777,
      "num_tokens": 7134145.0,
      "step": 3230
    },
    {
      "entropy": 5.664975547790528,
      "epoch": 3.2612203731719616,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004337339777035794,
      "loss": 5.2123,
      "mean_token_accuracy": 0.16639080941677092,
      "num_tokens": 7144727.0,
      "step": 3235
    },
    {
      "entropy": 5.517119407653809,
      "epoch": 3.2662632375189107,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000433452604049351,
      "loss": 5.1089,
      "mean_token_accuracy": 0.18070477694272996,
      "num_tokens": 7155531.0,
      "step": 3240
    },
    {
      "entropy": 5.501289176940918,
      "epoch": 3.27130610186586,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00043317073792077394,
      "loss": 5.1553,
      "mean_token_accuracy": 0.17724716365337373,
      "num_tokens": 7167234.0,
      "step": 3245
    },
    {
      "entropy": 5.6310209274292,
      "epoch": 3.276348966212809,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004328883801938947,
      "loss": 5.2372,
      "mean_token_accuracy": 0.16700577288866042,
      "num_tokens": 7178556.0,
      "step": 3250
    },
    {
      "entropy": 5.4585095882415775,
      "epoch": 3.281391830559758,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043260553174628776,
      "loss": 5.0999,
      "mean_token_accuracy": 0.17160716503858567,
      "num_tokens": 7189074.0,
      "step": 3255
    },
    {
      "entropy": 5.627786779403687,
      "epoch": 3.286434694906707,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004323221934570524,
      "loss": 5.2384,
      "mean_token_accuracy": 0.171539244055748,
      "num_tokens": 7200567.0,
      "step": 3260
    },
    {
      "entropy": 5.647009181976318,
      "epoch": 3.2914775592536563,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00043203836620681074,
      "loss": 5.2776,
      "mean_token_accuracy": 0.15886227488517762,
      "num_tokens": 7211012.0,
      "step": 3265
    },
    {
      "entropy": 5.502143430709839,
      "epoch": 3.2965204236006054,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00043175405087770435,
      "loss": 5.1611,
      "mean_token_accuracy": 0.17646982818841933,
      "num_tokens": 7222271.0,
      "step": 3270
    },
    {
      "entropy": 5.578966903686523,
      "epoch": 3.301563287947554,
      "grad_norm": 0.9375,
      "learning_rate": 0.00043146924835339186,
      "loss": 5.1118,
      "mean_token_accuracy": 0.1821632131934166,
      "num_tokens": 7235067.0,
      "step": 3275
    },
    {
      "entropy": 5.56513614654541,
      "epoch": 3.306606152294503,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043118395951904605,
      "loss": 5.1873,
      "mean_token_accuracy": 0.16946637332439424,
      "num_tokens": 7246159.0,
      "step": 3280
    },
    {
      "entropy": 5.685435056686401,
      "epoch": 3.3116490166414523,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004308981852613512,
      "loss": 5.2328,
      "mean_token_accuracy": 0.16842542439699174,
      "num_tokens": 7255944.0,
      "step": 3285
    },
    {
      "entropy": 5.615934753417969,
      "epoch": 3.3166918809884014,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004306119264685003,
      "loss": 5.1596,
      "mean_token_accuracy": 0.17425203770399095,
      "num_tokens": 7267273.0,
      "step": 3290
    },
    {
      "entropy": 5.505468893051147,
      "epoch": 3.3217347453353505,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004303251840301923,
      "loss": 5.1598,
      "mean_token_accuracy": 0.1717524603009224,
      "num_tokens": 7277622.0,
      "step": 3295
    },
    {
      "entropy": 5.5486010074615475,
      "epoch": 3.3267776096822996,
      "grad_norm": 0.921875,
      "learning_rate": 0.00043003795883762933,
      "loss": 5.21,
      "mean_token_accuracy": 0.1685906544327736,
      "num_tokens": 7288851.0,
      "step": 3300
    },
    {
      "entropy": 5.638274908065796,
      "epoch": 3.3318204740292487,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00042975025178351386,
      "loss": 5.263,
      "mean_token_accuracy": 0.16811128109693527,
      "num_tokens": 7302175.0,
      "step": 3305
    },
    {
      "entropy": 5.570336866378784,
      "epoch": 3.336863338376198,
      "grad_norm": 1.1875,
      "learning_rate": 0.000429462063762046,
      "loss": 5.1257,
      "mean_token_accuracy": 0.18063922226428986,
      "num_tokens": 7312693.0,
      "step": 3310
    },
    {
      "entropy": 5.529784250259399,
      "epoch": 3.3419062027231465,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042917339566892095,
      "loss": 5.1621,
      "mean_token_accuracy": 0.16770990788936616,
      "num_tokens": 7323790.0,
      "step": 3315
    },
    {
      "entropy": 5.5149060726165775,
      "epoch": 3.3469490670700957,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004288842484013257,
      "loss": 5.1403,
      "mean_token_accuracy": 0.175401271879673,
      "num_tokens": 7333797.0,
      "step": 3320
    },
    {
      "entropy": 5.579447603225708,
      "epoch": 3.3519919314170448,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004285946228579365,
      "loss": 5.1839,
      "mean_token_accuracy": 0.17431568056344987,
      "num_tokens": 7344674.0,
      "step": 3325
    },
    {
      "entropy": 5.5620870113372805,
      "epoch": 3.357034795763994,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004283045199389164,
      "loss": 5.2067,
      "mean_token_accuracy": 0.1729896619915962,
      "num_tokens": 7355978.0,
      "step": 3330
    },
    {
      "entropy": 5.653823900222778,
      "epoch": 3.362077660110943,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042801394054591197,
      "loss": 5.2803,
      "mean_token_accuracy": 0.16814187318086624,
      "num_tokens": 7367054.0,
      "step": 3335
    },
    {
      "entropy": 5.639859628677368,
      "epoch": 3.367120524457892,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042772288558205057,
      "loss": 5.2519,
      "mean_token_accuracy": 0.16838383972644805,
      "num_tokens": 7379588.0,
      "step": 3340
    },
    {
      "entropy": 5.505940055847168,
      "epoch": 3.3721633888048412,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004274313559519377,
      "loss": 5.2058,
      "mean_token_accuracy": 0.17158799320459367,
      "num_tokens": 7392411.0,
      "step": 3345
    },
    {
      "entropy": 5.5562413215637205,
      "epoch": 3.3772062531517903,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004271393525616544,
      "loss": 5.1062,
      "mean_token_accuracy": 0.17242553681135178,
      "num_tokens": 7402801.0,
      "step": 3350
    },
    {
      "entropy": 5.531540632247925,
      "epoch": 3.3822491174987395,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042684687631875377,
      "loss": 5.1732,
      "mean_token_accuracy": 0.16918187886476516,
      "num_tokens": 7413084.0,
      "step": 3355
    },
    {
      "entropy": 5.556858777999878,
      "epoch": 3.3872919818456886,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042655392813225864,
      "loss": 5.1198,
      "mean_token_accuracy": 0.17909411042928697,
      "num_tokens": 7424004.0,
      "step": 3360
    },
    {
      "entropy": 5.493006563186645,
      "epoch": 3.3923348461926373,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042626050891265884,
      "loss": 5.1326,
      "mean_token_accuracy": 0.18097762614488602,
      "num_tokens": 7434461.0,
      "step": 3365
    },
    {
      "entropy": 5.605647373199463,
      "epoch": 3.3973777105395864,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004259666195719081,
      "loss": 5.2107,
      "mean_token_accuracy": 0.17090678513050078,
      "num_tokens": 7445552.0,
      "step": 3370
    },
    {
      "entropy": 5.56715497970581,
      "epoch": 3.4024205748865355,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042567226102342134,
      "loss": 5.181,
      "mean_token_accuracy": 0.16866763383150102,
      "num_tokens": 7457073.0,
      "step": 3375
    },
    {
      "entropy": 5.535942697525025,
      "epoch": 3.4074634392334846,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004253774341820715,
      "loss": 5.1758,
      "mean_token_accuracy": 0.1721722275018692,
      "num_tokens": 7468197.0,
      "step": 3380
    },
    {
      "entropy": 5.61024489402771,
      "epoch": 3.4125063035804337,
      "grad_norm": 1.0,
      "learning_rate": 0.0004250821399641874,
      "loss": 5.1386,
      "mean_token_accuracy": 0.17660127729177474,
      "num_tokens": 7480751.0,
      "step": 3385
    },
    {
      "entropy": 5.517545080184936,
      "epoch": 3.417549167927383,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004247863792875502,
      "loss": 5.1683,
      "mean_token_accuracy": 0.17240165024995804,
      "num_tokens": 7492259.0,
      "step": 3390
    },
    {
      "entropy": 5.587699842453003,
      "epoch": 3.422592032274332,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042449015307139076,
      "loss": 5.2434,
      "mean_token_accuracy": 0.1663287788629532,
      "num_tokens": 7504317.0,
      "step": 3395
    },
    {
      "entropy": 5.5454551696777346,
      "epoch": 3.427634896621281,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042419346223638725,
      "loss": 5.1463,
      "mean_token_accuracy": 0.16699292212724687,
      "num_tokens": 7516039.0,
      "step": 3400
    },
    {
      "entropy": 5.573681116104126,
      "epoch": 3.4326777609682297,
      "grad_norm": 1.125,
      "learning_rate": 0.0004238963077046616,
      "loss": 5.2059,
      "mean_token_accuracy": 0.1723529264330864,
      "num_tokens": 7526095.0,
      "step": 3405
    },
    {
      "entropy": 5.520991468429566,
      "epoch": 3.437720625315179,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004235986903997767,
      "loss": 5.1675,
      "mean_token_accuracy": 0.1683811292052269,
      "num_tokens": 7537320.0,
      "step": 3410
    },
    {
      "entropy": 5.54693570137024,
      "epoch": 3.442763489662128,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042330061124673437,
      "loss": 5.227,
      "mean_token_accuracy": 0.17466289550065994,
      "num_tokens": 7548100.0,
      "step": 3415
    },
    {
      "entropy": 5.616562747955323,
      "epoch": 3.447806354009077,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042300207117197116,
      "loss": 5.2368,
      "mean_token_accuracy": 0.1656766802072525,
      "num_tokens": 7560483.0,
      "step": 3420
    },
    {
      "entropy": 5.576569223403931,
      "epoch": 3.452849218356026,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042270307110335657,
      "loss": 5.1466,
      "mean_token_accuracy": 0.17578542977571487,
      "num_tokens": 7571256.0,
      "step": 3425
    },
    {
      "entropy": 5.581924104690552,
      "epoch": 3.4578920827029753,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004224036119701899,
      "loss": 5.1363,
      "mean_token_accuracy": 0.17344688922166823,
      "num_tokens": 7581716.0,
      "step": 3430
    },
    {
      "entropy": 5.568325090408325,
      "epoch": 3.4629349470499244,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042210369470319694,
      "loss": 5.1716,
      "mean_token_accuracy": 0.17784251570701598,
      "num_tokens": 7591932.0,
      "step": 3435
    },
    {
      "entropy": 5.54473614692688,
      "epoch": 3.4679778113968736,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004218033202345274,
      "loss": 5.1547,
      "mean_token_accuracy": 0.17579526603221893,
      "num_tokens": 7602842.0,
      "step": 3440
    },
    {
      "entropy": 5.5082536220550535,
      "epoch": 3.4730206757438227,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004215024894977522,
      "loss": 5.1284,
      "mean_token_accuracy": 0.170816670358181,
      "num_tokens": 7614075.0,
      "step": 3445
    },
    {
      "entropy": 5.619481372833252,
      "epoch": 3.478063540090772,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004212012034278601,
      "loss": 5.3188,
      "mean_token_accuracy": 0.16399993300437926,
      "num_tokens": 7625642.0,
      "step": 3450
    },
    {
      "entropy": 5.6069879055023195,
      "epoch": 3.4831064044377205,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042089946296125513,
      "loss": 5.167,
      "mean_token_accuracy": 0.17243748605251313,
      "num_tokens": 7636447.0,
      "step": 3455
    },
    {
      "entropy": 5.5001214981079105,
      "epoch": 3.4881492687846696,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042059726903575373,
      "loss": 5.0947,
      "mean_token_accuracy": 0.1768268719315529,
      "num_tokens": 7648684.0,
      "step": 3460
    },
    {
      "entropy": 5.489742374420166,
      "epoch": 3.4931921331316187,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004202946225905815,
      "loss": 5.1754,
      "mean_token_accuracy": 0.1704980343580246,
      "num_tokens": 7660864.0,
      "step": 3465
    },
    {
      "entropy": 5.514662170410157,
      "epoch": 3.498234997478568,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004199915245663706,
      "loss": 5.1872,
      "mean_token_accuracy": 0.17248573452234267,
      "num_tokens": 7671939.0,
      "step": 3470
    },
    {
      "entropy": 5.699912452697754,
      "epoch": 3.503277861825517,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00041968797590515673,
      "loss": 5.2061,
      "mean_token_accuracy": 0.17173860222101212,
      "num_tokens": 7681536.0,
      "step": 3475
    },
    {
      "entropy": 5.601624536514282,
      "epoch": 3.508320726172466,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004193839775503761,
      "loss": 5.1414,
      "mean_token_accuracy": 0.17879641503095628,
      "num_tokens": 7693674.0,
      "step": 3480
    },
    {
      "entropy": 5.47628345489502,
      "epoch": 3.513363590519415,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041907953044686253,
      "loss": 5.1825,
      "mean_token_accuracy": 0.17948997914791107,
      "num_tokens": 7704533.0,
      "step": 3485
    },
    {
      "entropy": 5.518052768707276,
      "epoch": 3.518406454866364,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004187746355408448,
      "loss": 5.2263,
      "mean_token_accuracy": 0.16977648884058,
      "num_tokens": 7715448.0,
      "step": 3490
    },
    {
      "entropy": 5.579763174057007,
      "epoch": 3.523449319213313,
      "grad_norm": 1.125,
      "learning_rate": 0.0004184692937799431,
      "loss": 5.2023,
      "mean_token_accuracy": 0.17356830686330796,
      "num_tokens": 7726808.0,
      "step": 3495
    },
    {
      "entropy": 5.729559993743896,
      "epoch": 3.528492183560262,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041816350611316693,
      "loss": 5.2001,
      "mean_token_accuracy": 0.16943881660699844,
      "num_tokens": 7737899.0,
      "step": 3500
    },
    {
      "epoch": 3.528492183560262,
      "eval_entropy": 5.410774234597009,
      "eval_loss": 5.73056697845459,
      "eval_mean_token_accuracy": 0.14811429887555028,
      "eval_num_tokens": 7737899.0,
      "eval_runtime": 2.0683,
      "eval_samples_per_second": 1517.181,
      "eval_steps_per_second": 190.01,
      "step": 3500
    },
    {
      "entropy": 5.527662754058838,
      "epoch": 3.533535047907211,
      "grad_norm": 0.984375,
      "learning_rate": 0.00041785727349091117,
      "loss": 5.1386,
      "mean_token_accuracy": 0.17433379888534545,
      "num_tokens": 7749268.0,
      "step": 3505
    },
    {
      "entropy": 5.563727283477784,
      "epoch": 3.5385779122541603,
      "grad_norm": 1.140625,
      "learning_rate": 0.000417550596864954,
      "loss": 5.2129,
      "mean_token_accuracy": 0.1663740873336792,
      "num_tokens": 7760312.0,
      "step": 3510
    },
    {
      "entropy": 5.581764173507691,
      "epoch": 3.5436207766011094,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041724347718845344,
      "loss": 5.2045,
      "mean_token_accuracy": 0.1739761710166931,
      "num_tokens": 7771130.0,
      "step": 3515
    },
    {
      "entropy": 5.568504285812378,
      "epoch": 3.5486636409480585,
      "grad_norm": 1.328125,
      "learning_rate": 0.00041693591541594455,
      "loss": 5.1698,
      "mean_token_accuracy": 0.1687783792614937,
      "num_tokens": 7783068.0,
      "step": 3520
    },
    {
      "entropy": 5.549349689483643,
      "epoch": 3.5537065052950076,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004166279125033363,
      "loss": 5.2754,
      "mean_token_accuracy": 0.16857239007949829,
      "num_tokens": 7794314.0,
      "step": 3525
    },
    {
      "entropy": 5.653335380554199,
      "epoch": 3.5587493696419568,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041631946940790893,
      "loss": 5.2218,
      "mean_token_accuracy": 0.17252931296825408,
      "num_tokens": 7806888.0,
      "step": 3530
    },
    {
      "entropy": 5.5294476509094235,
      "epoch": 3.563792233988906,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041601058708831077,
      "loss": 5.2149,
      "mean_token_accuracy": 0.1707366332411766,
      "num_tokens": 7818690.0,
      "step": 3535
    },
    {
      "entropy": 5.567249727249146,
      "epoch": 3.568835098335855,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000415701266504555,
      "loss": 5.1458,
      "mean_token_accuracy": 0.16551820188760757,
      "num_tokens": 7829321.0,
      "step": 3540
    },
    {
      "entropy": 5.538458251953125,
      "epoch": 3.5738779626828037,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004153915086180173,
      "loss": 5.1947,
      "mean_token_accuracy": 0.17398821860551833,
      "num_tokens": 7840223.0,
      "step": 3545
    },
    {
      "entropy": 5.51594386100769,
      "epoch": 3.578920827029753,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004150813143914323,
      "loss": 5.1898,
      "mean_token_accuracy": 0.16978464126586915,
      "num_tokens": 7850999.0,
      "step": 3550
    },
    {
      "entropy": 5.628743028640747,
      "epoch": 3.583963691376702,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004147706847888908,
      "loss": 5.2538,
      "mean_token_accuracy": 0.1682332292199135,
      "num_tokens": 7861935.0,
      "step": 3555
    },
    {
      "entropy": 5.507300281524659,
      "epoch": 3.589006555723651,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004144596207758367,
      "loss": 5.2402,
      "mean_token_accuracy": 0.17258633375167848,
      "num_tokens": 7872960.0,
      "step": 3560
    },
    {
      "entropy": 5.569037914276123,
      "epoch": 3.5940494200706,
      "grad_norm": 1.125,
      "learning_rate": 0.0004141481233190642,
      "loss": 5.2699,
      "mean_token_accuracy": 0.1672738164663315,
      "num_tokens": 7883881.0,
      "step": 3565
    },
    {
      "entropy": 5.613865613937378,
      "epoch": 3.5990922844175492,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004138361933867145,
      "loss": 5.2578,
      "mean_token_accuracy": 0.1652541786432266,
      "num_tokens": 7895106.0,
      "step": 3570
    },
    {
      "entropy": 5.512177133560181,
      "epoch": 3.6041351487644984,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004135238319482731,
      "loss": 5.1541,
      "mean_token_accuracy": 0.17173427790403367,
      "num_tokens": 7906020.0,
      "step": 3575
    },
    {
      "entropy": 5.540937423706055,
      "epoch": 3.609178013111447,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004132110399745666,
      "loss": 5.2514,
      "mean_token_accuracy": 0.1678391680121422,
      "num_tokens": 7918959.0,
      "step": 3580
    },
    {
      "entropy": 5.630393028259277,
      "epoch": 3.614220877458396,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004128978184377596,
      "loss": 5.1901,
      "mean_token_accuracy": 0.16656388640403746,
      "num_tokens": 7929804.0,
      "step": 3585
    },
    {
      "entropy": 5.53603572845459,
      "epoch": 3.6192637418053453,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004125841683113518,
      "loss": 5.1958,
      "mean_token_accuracy": 0.16942005455493928,
      "num_tokens": 7941558.0,
      "step": 3590
    },
    {
      "entropy": 5.504640531539917,
      "epoch": 3.6243066061522944,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041227009057017523,
      "loss": 5.2362,
      "mean_token_accuracy": 0.1669396474957466,
      "num_tokens": 7951737.0,
      "step": 3595
    },
    {
      "entropy": 5.477955484390259,
      "epoch": 3.6293494704992435,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004119555861903906,
      "loss": 5.1491,
      "mean_token_accuracy": 0.1855486288666725,
      "num_tokens": 7963164.0,
      "step": 3600
    },
    {
      "entropy": 5.591642713546753,
      "epoch": 3.6343923348461926,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004116406561494847,
      "loss": 5.206,
      "mean_token_accuracy": 0.17106180042028427,
      "num_tokens": 7973275.0,
      "step": 3605
    },
    {
      "entropy": 5.56058440208435,
      "epoch": 3.6394351991931417,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041132530142626756,
      "loss": 5.1788,
      "mean_token_accuracy": 0.17878760546445846,
      "num_tokens": 7984568.0,
      "step": 3610
    },
    {
      "entropy": 5.566657400131225,
      "epoch": 3.644478063540091,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041100952300086884,
      "loss": 5.1969,
      "mean_token_accuracy": 0.16652931421995162,
      "num_tokens": 7994263.0,
      "step": 3615
    },
    {
      "entropy": 5.579933023452758,
      "epoch": 3.64952092788704,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041069332185473514,
      "loss": 5.1776,
      "mean_token_accuracy": 0.17230696678161622,
      "num_tokens": 8004881.0,
      "step": 3620
    },
    {
      "entropy": 5.534660816192627,
      "epoch": 3.654563792233989,
      "grad_norm": 1.234375,
      "learning_rate": 0.000410376698970627,
      "loss": 5.172,
      "mean_token_accuracy": 0.1734602451324463,
      "num_tokens": 8015955.0,
      "step": 3625
    },
    {
      "entropy": 5.525668287277222,
      "epoch": 3.659606656580938,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00041005965533261554,
      "loss": 5.169,
      "mean_token_accuracy": 0.17173525989055632,
      "num_tokens": 8027010.0,
      "step": 3630
    },
    {
      "entropy": 5.475025987625122,
      "epoch": 3.6646495209278873,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040974219192607973,
      "loss": 5.1299,
      "mean_token_accuracy": 0.1725467100739479,
      "num_tokens": 8038564.0,
      "step": 3635
    },
    {
      "entropy": 5.547386026382446,
      "epoch": 3.669692385274836,
      "grad_norm": 1.046875,
      "learning_rate": 0.00040942430973770323,
      "loss": 5.1876,
      "mean_token_accuracy": 0.17370582222938538,
      "num_tokens": 8049975.0,
      "step": 3640
    },
    {
      "entropy": 5.579010152816773,
      "epoch": 3.674735249621785,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004091060097554712,
      "loss": 5.211,
      "mean_token_accuracy": 0.17175272703170777,
      "num_tokens": 8061042.0,
      "step": 3645
    },
    {
      "entropy": 5.5676249980926515,
      "epoch": 3.679778113968734,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004087872929686673,
      "loss": 5.1471,
      "mean_token_accuracy": 0.17469496577978133,
      "num_tokens": 8071777.0,
      "step": 3650
    },
    {
      "entropy": 5.457219123840332,
      "epoch": 3.6848209783156833,
      "grad_norm": 1.0,
      "learning_rate": 0.00040846816036787056,
      "loss": 5.225,
      "mean_token_accuracy": 0.17314725369215012,
      "num_tokens": 8082387.0,
      "step": 3655
    },
    {
      "entropy": 5.623361539840698,
      "epoch": 3.6898638426626325,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004081486129449527,
      "loss": 5.1434,
      "mean_token_accuracy": 0.17689687311649321,
      "num_tokens": 8091818.0,
      "step": 3660
    },
    {
      "entropy": 5.609028339385986,
      "epoch": 3.6949067070095816,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00040782865169307423,
      "loss": 5.199,
      "mean_token_accuracy": 0.1698301613330841,
      "num_tokens": 8103377.0,
      "step": 3665
    },
    {
      "entropy": 5.546071767807007,
      "epoch": 3.6999495713565302,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004075082776066822,
      "loss": 5.2442,
      "mean_token_accuracy": 0.1719037488102913,
      "num_tokens": 8114723.0,
      "step": 3670
    },
    {
      "entropy": 5.483772945404053,
      "epoch": 3.7049924357034794,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004071874916815068,
      "loss": 5.0908,
      "mean_token_accuracy": 0.17202049344778061,
      "num_tokens": 8124530.0,
      "step": 3675
    },
    {
      "entropy": 5.445228815078735,
      "epoch": 3.7100353000504285,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040686629491455787,
      "loss": 5.1713,
      "mean_token_accuracy": 0.17136170864105224,
      "num_tokens": 8134894.0,
      "step": 3680
    },
    {
      "entropy": 5.535150051116943,
      "epoch": 3.7150781643973776,
      "grad_norm": 1.21875,
      "learning_rate": 0.00040654468830412247,
      "loss": 5.1438,
      "mean_token_accuracy": 0.1749508187174797,
      "num_tokens": 8143499.0,
      "step": 3685
    },
    {
      "entropy": 5.500586891174317,
      "epoch": 3.7201210287443267,
      "grad_norm": 1.046875,
      "learning_rate": 0.00040622267284976134,
      "loss": 5.1882,
      "mean_token_accuracy": 0.1722453713417053,
      "num_tokens": 8154556.0,
      "step": 3690
    },
    {
      "entropy": 5.547470283508301,
      "epoch": 3.725163893091276,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040590024955230596,
      "loss": 5.2759,
      "mean_token_accuracy": 0.16633043438196182,
      "num_tokens": 8166047.0,
      "step": 3695
    },
    {
      "entropy": 5.585486316680909,
      "epoch": 3.730206757438225,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004055774194138554,
      "loss": 5.193,
      "mean_token_accuracy": 0.1715107947587967,
      "num_tokens": 8177068.0,
      "step": 3700
    },
    {
      "entropy": 5.4966779232025145,
      "epoch": 3.735249621785174,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004052541834377731,
      "loss": 5.1798,
      "mean_token_accuracy": 0.17475247979164124,
      "num_tokens": 8187316.0,
      "step": 3705
    },
    {
      "entropy": 5.552641677856445,
      "epoch": 3.740292486132123,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040493054262868397,
      "loss": 5.2944,
      "mean_token_accuracy": 0.1712053671479225,
      "num_tokens": 8198642.0,
      "step": 3710
    },
    {
      "entropy": 5.5439847946167,
      "epoch": 3.7453353504790723,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040460649799247116,
      "loss": 5.1425,
      "mean_token_accuracy": 0.172932006418705,
      "num_tokens": 8208476.0,
      "step": 3715
    },
    {
      "entropy": 5.520895195007324,
      "epoch": 3.7503782148260214,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004042820505362728,
      "loss": 5.2276,
      "mean_token_accuracy": 0.1698770135641098,
      "num_tokens": 8219590.0,
      "step": 3720
    },
    {
      "entropy": 5.488117837905884,
      "epoch": 3.7554210791729705,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004039572012684792,
      "loss": 5.1448,
      "mean_token_accuracy": 0.1791684791445732,
      "num_tokens": 8230707.0,
      "step": 3725
    },
    {
      "entropy": 5.606340312957764,
      "epoch": 3.760463943519919,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00040363195119872916,
      "loss": 5.2015,
      "mean_token_accuracy": 0.16538176387548448,
      "num_tokens": 8241774.0,
      "step": 3730
    },
    {
      "entropy": 5.616064262390137,
      "epoch": 3.7655068078668683,
      "grad_norm": 1.0625,
      "learning_rate": 0.00040330630133790757,
      "loss": 5.3181,
      "mean_token_accuracy": 0.16288650184869766,
      "num_tokens": 8254314.0,
      "step": 3735
    },
    {
      "entropy": 5.552481555938721,
      "epoch": 3.7705496722138174,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040298025269814165,
      "loss": 5.1656,
      "mean_token_accuracy": 0.17419897466897966,
      "num_tokens": 8265650.0,
      "step": 3740
    },
    {
      "entropy": 5.503885316848755,
      "epoch": 3.7755925365607665,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004026538062927981,
      "loss": 5.1287,
      "mean_token_accuracy": 0.17540400475263596,
      "num_tokens": 8276808.0,
      "step": 3745
    },
    {
      "entropy": 5.454111766815186,
      "epoch": 3.7806354009077157,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040232696313647985,
      "loss": 5.0806,
      "mean_token_accuracy": 0.1725289911031723,
      "num_tokens": 8287867.0,
      "step": 3750
    },
    {
      "entropy": 5.5325781345367435,
      "epoch": 3.7856782652546648,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004019997242450231,
      "loss": 5.1261,
      "mean_token_accuracy": 0.1780746951699257,
      "num_tokens": 8298707.0,
      "step": 3755
    },
    {
      "entropy": 5.454637432098389,
      "epoch": 3.7907211296016134,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004016720906354937,
      "loss": 5.0933,
      "mean_token_accuracy": 0.17663703709840775,
      "num_tokens": 8309363.0,
      "step": 3760
    },
    {
      "entropy": 5.550456285476685,
      "epoch": 3.7957639939485626,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004013440633261845,
      "loss": 5.2596,
      "mean_token_accuracy": 0.17098919153213502,
      "num_tokens": 8319376.0,
      "step": 3765
    },
    {
      "entropy": 5.522191858291626,
      "epoch": 3.8008068582955117,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00040101564333661213,
      "loss": 5.169,
      "mean_token_accuracy": 0.17651111632585526,
      "num_tokens": 8330842.0,
      "step": 3770
    },
    {
      "entropy": 5.5011437892913815,
      "epoch": 3.805849722642461,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040068683168751316,
      "loss": 5.1215,
      "mean_token_accuracy": 0.180177204310894,
      "num_tokens": 8341588.0,
      "step": 3775
    },
    {
      "entropy": 5.451708507537842,
      "epoch": 3.81089258698941,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00040035762940084205,
      "loss": 5.1685,
      "mean_token_accuracy": 0.17241425365209578,
      "num_tokens": 8351664.0,
      "step": 3780
    },
    {
      "entropy": 5.607029628753662,
      "epoch": 3.815935451336359,
      "grad_norm": 1.03125,
      "learning_rate": 0.000400028037499767,
      "loss": 5.2478,
      "mean_token_accuracy": 0.164016430079937,
      "num_tokens": 8362803.0,
      "step": 3785
    },
    {
      "entropy": 5.493664979934692,
      "epoch": 3.820978315683308,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003996980570086673,
      "loss": 5.175,
      "mean_token_accuracy": 0.1727006509900093,
      "num_tokens": 8373791.0,
      "step": 3790
    },
    {
      "entropy": 5.549453020095825,
      "epoch": 3.8260211800302573,
      "grad_norm": 1.078125,
      "learning_rate": 0.00039936768895312987,
      "loss": 5.172,
      "mean_token_accuracy": 0.1739651754498482,
      "num_tokens": 8385738.0,
      "step": 3795
    },
    {
      "entropy": 5.537290287017822,
      "epoch": 3.8310640443772064,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00039903693435994626,
      "loss": 5.1774,
      "mean_token_accuracy": 0.1732986092567444,
      "num_tokens": 8396924.0,
      "step": 3800
    },
    {
      "entropy": 5.531874084472657,
      "epoch": 3.8361069087241555,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00039870579425710937,
      "loss": 5.2513,
      "mean_token_accuracy": 0.16677893623709678,
      "num_tokens": 8409040.0,
      "step": 3805
    },
    {
      "entropy": 5.437445211410522,
      "epoch": 3.8411497730711046,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003983742696738103,
      "loss": 5.0969,
      "mean_token_accuracy": 0.180625282227993,
      "num_tokens": 8419630.0,
      "step": 3810
    },
    {
      "entropy": 5.589530992507934,
      "epoch": 3.8461926374180537,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003980423616404351,
      "loss": 5.2383,
      "mean_token_accuracy": 0.17389712631702423,
      "num_tokens": 8430036.0,
      "step": 3815
    },
    {
      "entropy": 5.54722228050232,
      "epoch": 3.8512355017650024,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003977100711885616,
      "loss": 5.1301,
      "mean_token_accuracy": 0.1754632204771042,
      "num_tokens": 8439829.0,
      "step": 3820
    },
    {
      "entropy": 5.453087949752808,
      "epoch": 3.8562783661119515,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039737739935095614,
      "loss": 5.1345,
      "mean_token_accuracy": 0.1706774950027466,
      "num_tokens": 8450492.0,
      "step": 3825
    },
    {
      "entropy": 5.502632713317871,
      "epoch": 3.8613212304589006,
      "grad_norm": 0.953125,
      "learning_rate": 0.0003970443471615705,
      "loss": 5.2081,
      "mean_token_accuracy": 0.1714776113629341,
      "num_tokens": 8462200.0,
      "step": 3830
    },
    {
      "entropy": 5.5232398986816404,
      "epoch": 3.8663640948058497,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003967109156555386,
      "loss": 5.1592,
      "mean_token_accuracy": 0.17164637446403502,
      "num_tokens": 8473393.0,
      "step": 3835
    },
    {
      "entropy": 5.4516627311706545,
      "epoch": 3.871406959152799,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00039637710586917324,
      "loss": 5.1469,
      "mean_token_accuracy": 0.17415427267551423,
      "num_tokens": 8484646.0,
      "step": 3840
    },
    {
      "entropy": 5.546061944961548,
      "epoch": 3.876449823499748,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039604291883996287,
      "loss": 5.1894,
      "mean_token_accuracy": 0.17310727387666702,
      "num_tokens": 8495371.0,
      "step": 3845
    },
    {
      "entropy": 5.492070913314819,
      "epoch": 3.8814926878466967,
      "grad_norm": 1.078125,
      "learning_rate": 0.00039570835560656854,
      "loss": 5.2057,
      "mean_token_accuracy": 0.17188188880681993,
      "num_tokens": 8507976.0,
      "step": 3850
    },
    {
      "entropy": 5.521370792388916,
      "epoch": 3.8865355521936458,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003953734172088205,
      "loss": 5.1875,
      "mean_token_accuracy": 0.17714476585388184,
      "num_tokens": 8519120.0,
      "step": 3855
    },
    {
      "entropy": 5.544731330871582,
      "epoch": 3.891578416540595,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039503810468771494,
      "loss": 5.1772,
      "mean_token_accuracy": 0.17596625685691833,
      "num_tokens": 8529727.0,
      "step": 3860
    },
    {
      "entropy": 5.470485258102417,
      "epoch": 3.896621280887544,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003947024190854111,
      "loss": 5.1433,
      "mean_token_accuracy": 0.17271711826324462,
      "num_tokens": 8540202.0,
      "step": 3865
    },
    {
      "entropy": 5.553111171722412,
      "epoch": 3.901664145234493,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039436636144522737,
      "loss": 5.1463,
      "mean_token_accuracy": 0.17616454958915712,
      "num_tokens": 8550810.0,
      "step": 3870
    },
    {
      "entropy": 5.466653299331665,
      "epoch": 3.9067070095814422,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039402993281163875,
      "loss": 5.1485,
      "mean_token_accuracy": 0.17874222993850708,
      "num_tokens": 8561120.0,
      "step": 3875
    },
    {
      "entropy": 5.521090650558472,
      "epoch": 3.9117498739283914,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003936931342302732,
      "loss": 5.1957,
      "mean_token_accuracy": 0.17193490862846375,
      "num_tokens": 8572153.0,
      "step": 3880
    },
    {
      "entropy": 5.499665260314941,
      "epoch": 3.9167927382753405,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003933559667479084,
      "loss": 5.1694,
      "mean_token_accuracy": 0.1719837799668312,
      "num_tokens": 8583740.0,
      "step": 3885
    },
    {
      "entropy": 5.550255823135376,
      "epoch": 3.9218356026222896,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039301843141246876,
      "loss": 5.1746,
      "mean_token_accuracy": 0.17574147135019302,
      "num_tokens": 8596017.0,
      "step": 3890
    },
    {
      "entropy": 5.529006099700927,
      "epoch": 3.9268784669692387,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039268052927302185,
      "loss": 5.1726,
      "mean_token_accuracy": 0.17430050820112228,
      "num_tokens": 8608121.0,
      "step": 3895
    },
    {
      "entropy": 5.544325113296509,
      "epoch": 3.931921331316188,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0003923422613797753,
      "loss": 5.2122,
      "mean_token_accuracy": 0.17328143417835234,
      "num_tokens": 8621446.0,
      "step": 3900
    },
    {
      "entropy": 5.535483932495117,
      "epoch": 3.936964195663137,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003920036287840734,
      "loss": 5.2076,
      "mean_token_accuracy": 0.17391748279333114,
      "num_tokens": 8632088.0,
      "step": 3905
    },
    {
      "entropy": 5.499102592468262,
      "epoch": 3.9420070600100856,
      "grad_norm": 1.125,
      "learning_rate": 0.00039166463253839434,
      "loss": 5.0767,
      "mean_token_accuracy": 0.18794226497411728,
      "num_tokens": 8642546.0,
      "step": 3910
    },
    {
      "entropy": 5.4973818302154545,
      "epoch": 3.9470499243570347,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003913252736963462,
      "loss": 5.1805,
      "mean_token_accuracy": 0.17101371586322783,
      "num_tokens": 8652869.0,
      "step": 3915
    },
    {
      "entropy": 5.510268974304199,
      "epoch": 3.952092788703984,
      "grad_norm": 1.109375,
      "learning_rate": 0.000390985553312664,
      "loss": 5.1208,
      "mean_token_accuracy": 0.1707632213830948,
      "num_tokens": 8662834.0,
      "step": 3920
    },
    {
      "entropy": 5.546622085571289,
      "epoch": 3.957135653050933,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003906454724432065,
      "loss": 5.2409,
      "mean_token_accuracy": 0.17106539458036424,
      "num_tokens": 8673372.0,
      "step": 3925
    },
    {
      "entropy": 5.503659582138061,
      "epoch": 3.962178517397882,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00039030503214495296,
      "loss": 5.1776,
      "mean_token_accuracy": 0.17401908636093139,
      "num_tokens": 8683660.0,
      "step": 3930
    },
    {
      "entropy": 5.540773248672485,
      "epoch": 3.967221381744831,
      "grad_norm": 1.125,
      "learning_rate": 0.0003899642334759996,
      "loss": 5.2035,
      "mean_token_accuracy": 0.17064203023910524,
      "num_tokens": 8694044.0,
      "step": 3935
    },
    {
      "entropy": 5.559845685958862,
      "epoch": 3.97226424609178,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003896230774955567,
      "loss": 5.2165,
      "mean_token_accuracy": 0.1667879894375801,
      "num_tokens": 8705014.0,
      "step": 3940
    },
    {
      "entropy": 5.5432085514068605,
      "epoch": 3.977307110438729,
      "grad_norm": 1.21875,
      "learning_rate": 0.00038928156526394476,
      "loss": 5.1556,
      "mean_token_accuracy": 0.17010375410318374,
      "num_tokens": 8716007.0,
      "step": 3945
    },
    {
      "entropy": 5.53149642944336,
      "epoch": 3.982349974785678,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003889396978425918,
      "loss": 5.1835,
      "mean_token_accuracy": 0.1785310074687004,
      "num_tokens": 8728390.0,
      "step": 3950
    },
    {
      "entropy": 5.491573667526245,
      "epoch": 3.987392839132627,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003885974762940295,
      "loss": 5.1408,
      "mean_token_accuracy": 0.17592601627111434,
      "num_tokens": 8738806.0,
      "step": 3955
    },
    {
      "entropy": 5.526895141601562,
      "epoch": 3.9924357034795763,
      "grad_norm": 0.96875,
      "learning_rate": 0.0003882549016818905,
      "loss": 5.1857,
      "mean_token_accuracy": 0.17386045008897782,
      "num_tokens": 8749667.0,
      "step": 3960
    },
    {
      "entropy": 5.477924108505249,
      "epoch": 3.9974785678265254,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00038791197507090456,
      "loss": 5.1966,
      "mean_token_accuracy": 0.17171989381313324,
      "num_tokens": 8761380.0,
      "step": 3965
    },
    {
      "entropy": 5.490013811323378,
      "epoch": 4.00201714573878,
      "grad_norm": 1.09375,
      "learning_rate": 0.00038756869752689527,
      "loss": 5.0361,
      "mean_token_accuracy": 0.18232868280675676,
      "num_tokens": 8770921.0,
      "step": 3970
    },
    {
      "entropy": 5.447642278671265,
      "epoch": 4.007060010085729,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003872250701167775,
      "loss": 4.8406,
      "mean_token_accuracy": 0.18878782093524932,
      "num_tokens": 8782706.0,
      "step": 3975
    },
    {
      "entropy": 5.490310335159302,
      "epoch": 4.012102874432678,
      "grad_norm": 1.015625,
      "learning_rate": 0.00038688109390855297,
      "loss": 4.9533,
      "mean_token_accuracy": 0.19089592546224593,
      "num_tokens": 8794600.0,
      "step": 3980
    },
    {
      "entropy": 5.46259651184082,
      "epoch": 4.017145738779627,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003865367699713077,
      "loss": 4.8377,
      "mean_token_accuracy": 0.19187896102666854,
      "num_tokens": 8804149.0,
      "step": 3985
    },
    {
      "entropy": 5.486433506011963,
      "epoch": 4.022188603126576,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00038619209937520836,
      "loss": 4.8538,
      "mean_token_accuracy": 0.19788114428520204,
      "num_tokens": 8814723.0,
      "step": 3990
    },
    {
      "entropy": 5.495340728759766,
      "epoch": 4.027231467473525,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00038584708319149924,
      "loss": 4.9445,
      "mean_token_accuracy": 0.18946300446987152,
      "num_tokens": 8827258.0,
      "step": 3995
    },
    {
      "entropy": 5.46904182434082,
      "epoch": 4.032274331820474,
      "grad_norm": 1.125,
      "learning_rate": 0.00038550172249249856,
      "loss": 4.9115,
      "mean_token_accuracy": 0.19089395105838775,
      "num_tokens": 8837528.0,
      "step": 4000
    },
    {
      "epoch": 4.032274331820474,
      "eval_entropy": 5.28786424098124,
      "eval_loss": 5.69589376449585,
      "eval_mean_token_accuracy": 0.15155033272413806,
      "eval_num_tokens": 8837528.0,
      "eval_runtime": 2.0496,
      "eval_samples_per_second": 1531.023,
      "eval_steps_per_second": 191.744,
      "step": 4000
    },
    {
      "entropy": 5.462073087692261,
      "epoch": 4.0373171961674235,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003851560183515953,
      "loss": 4.905,
      "mean_token_accuracy": 0.191329924762249,
      "num_tokens": 8848154.0,
      "step": 4005
    },
    {
      "entropy": 5.434517478942871,
      "epoch": 4.042360060514373,
      "grad_norm": 1.234375,
      "learning_rate": 0.000384809971843246,
      "loss": 4.8629,
      "mean_token_accuracy": 0.19430924504995345,
      "num_tokens": 8858396.0,
      "step": 4010
    },
    {
      "entropy": 5.510921335220337,
      "epoch": 4.047402924861321,
      "grad_norm": 1.265625,
      "learning_rate": 0.000384463584042971,
      "loss": 4.8883,
      "mean_token_accuracy": 0.198105388879776,
      "num_tokens": 8868590.0,
      "step": 4015
    },
    {
      "entropy": 5.45649561882019,
      "epoch": 4.05244578920827,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00038411685602735175,
      "loss": 4.8956,
      "mean_token_accuracy": 0.1927472785115242,
      "num_tokens": 8878955.0,
      "step": 4020
    },
    {
      "entropy": 5.416157865524292,
      "epoch": 4.057488653555219,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038376978887402687,
      "loss": 4.862,
      "mean_token_accuracy": 0.20044808089733124,
      "num_tokens": 8889402.0,
      "step": 4025
    },
    {
      "entropy": 5.44850344657898,
      "epoch": 4.062531517902168,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003834223836616891,
      "loss": 4.8987,
      "mean_token_accuracy": 0.189875091612339,
      "num_tokens": 8901433.0,
      "step": 4030
    },
    {
      "entropy": 5.401690292358398,
      "epoch": 4.067574382249117,
      "grad_norm": 1.15625,
      "learning_rate": 0.00038307464147008176,
      "loss": 4.9243,
      "mean_token_accuracy": 0.190950845181942,
      "num_tokens": 8912895.0,
      "step": 4035
    },
    {
      "entropy": 5.460486841201782,
      "epoch": 4.072617246596066,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003827265633799957,
      "loss": 4.9087,
      "mean_token_accuracy": 0.19388046413660048,
      "num_tokens": 8923529.0,
      "step": 4040
    },
    {
      "entropy": 5.45203423500061,
      "epoch": 4.0776601109430155,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003823781504732657,
      "loss": 4.8463,
      "mean_token_accuracy": 0.19833841025829316,
      "num_tokens": 8933279.0,
      "step": 4045
    },
    {
      "entropy": 5.494933128356934,
      "epoch": 4.082702975289965,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003820294038327669,
      "loss": 4.94,
      "mean_token_accuracy": 0.1916810765862465,
      "num_tokens": 8944396.0,
      "step": 4050
    },
    {
      "entropy": 5.4086400985717775,
      "epoch": 4.087745839636914,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003816803245424121,
      "loss": 4.7503,
      "mean_token_accuracy": 0.2029646098613739,
      "num_tokens": 8955535.0,
      "step": 4055
    },
    {
      "entropy": 5.4522851467132565,
      "epoch": 4.092788703983863,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003813309136871478,
      "loss": 4.9305,
      "mean_token_accuracy": 0.18798318654298782,
      "num_tokens": 8965859.0,
      "step": 4060
    },
    {
      "entropy": 5.469114685058594,
      "epoch": 4.097831568330812,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038098117235295083,
      "loss": 4.9591,
      "mean_token_accuracy": 0.18637694120407106,
      "num_tokens": 8976580.0,
      "step": 4065
    },
    {
      "entropy": 5.43038239479065,
      "epoch": 4.102874432677761,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003806311016268254,
      "loss": 4.9128,
      "mean_token_accuracy": 0.19565491378307343,
      "num_tokens": 8988154.0,
      "step": 4070
    },
    {
      "entropy": 5.456332445144653,
      "epoch": 4.10791729702471,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003802807025967995,
      "loss": 4.939,
      "mean_token_accuracy": 0.18973323851823806,
      "num_tokens": 8999758.0,
      "step": 4075
    },
    {
      "entropy": 5.419324207305908,
      "epoch": 4.112960161371659,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003799299763519213,
      "loss": 4.8878,
      "mean_token_accuracy": 0.19338995814323426,
      "num_tokens": 9010371.0,
      "step": 4080
    },
    {
      "entropy": 5.4459387302398685,
      "epoch": 4.118003025718608,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003795789239822561,
      "loss": 4.88,
      "mean_token_accuracy": 0.1987368032336235,
      "num_tokens": 9021224.0,
      "step": 4085
    },
    {
      "entropy": 5.428819370269776,
      "epoch": 4.1230458900655576,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003792275465788826,
      "loss": 4.9326,
      "mean_token_accuracy": 0.1912503197789192,
      "num_tokens": 9033569.0,
      "step": 4090
    },
    {
      "entropy": 5.458049774169922,
      "epoch": 4.128088754412507,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00037887584523389013,
      "loss": 4.9572,
      "mean_token_accuracy": 0.19111103862524031,
      "num_tokens": 9046166.0,
      "step": 4095
    },
    {
      "entropy": 5.427979326248169,
      "epoch": 4.133131618759456,
      "grad_norm": 1.125,
      "learning_rate": 0.00037852382104037454,
      "loss": 4.9464,
      "mean_token_accuracy": 0.1934666782617569,
      "num_tokens": 9056210.0,
      "step": 4100
    },
    {
      "entropy": 5.456173086166382,
      "epoch": 4.138174483106404,
      "grad_norm": 1.109375,
      "learning_rate": 0.00037817147509243497,
      "loss": 4.8916,
      "mean_token_accuracy": 0.18922187387943268,
      "num_tokens": 9067473.0,
      "step": 4105
    },
    {
      "entropy": 5.483760356903076,
      "epoch": 4.143217347453353,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003778188084851709,
      "loss": 4.9435,
      "mean_token_accuracy": 0.19201175570487977,
      "num_tokens": 9077744.0,
      "step": 4110
    },
    {
      "entropy": 5.456737422943116,
      "epoch": 4.148260211800302,
      "grad_norm": 1.15625,
      "learning_rate": 0.00037746582231467816,
      "loss": 4.95,
      "mean_token_accuracy": 0.18718351870775224,
      "num_tokens": 9088988.0,
      "step": 4115
    },
    {
      "entropy": 5.45961275100708,
      "epoch": 4.153303076147251,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003771125176780461,
      "loss": 4.9691,
      "mean_token_accuracy": 0.18171961307525636,
      "num_tokens": 9100584.0,
      "step": 4120
    },
    {
      "entropy": 5.445645046234131,
      "epoch": 4.1583459404942005,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00037675889567335345,
      "loss": 4.9367,
      "mean_token_accuracy": 0.184403595328331,
      "num_tokens": 9112442.0,
      "step": 4125
    },
    {
      "entropy": 5.452325391769409,
      "epoch": 4.16338880484115,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00037640495739966573,
      "loss": 4.9891,
      "mean_token_accuracy": 0.18297390639781952,
      "num_tokens": 9122903.0,
      "step": 4130
    },
    {
      "entropy": 5.53175539970398,
      "epoch": 4.168431669188099,
      "grad_norm": 1.140625,
      "learning_rate": 0.00037605070395703124,
      "loss": 5.0003,
      "mean_token_accuracy": 0.18721186965703965,
      "num_tokens": 9134301.0,
      "step": 4135
    },
    {
      "entropy": 5.595280170440674,
      "epoch": 4.173474533535048,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00037569613644647783,
      "loss": 4.9934,
      "mean_token_accuracy": 0.18326281011104584,
      "num_tokens": 9144503.0,
      "step": 4140
    },
    {
      "entropy": 5.372558832168579,
      "epoch": 4.178517397881997,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003753412559700094,
      "loss": 4.8128,
      "mean_token_accuracy": 0.1915489464998245,
      "num_tokens": 9154173.0,
      "step": 4145
    },
    {
      "entropy": 5.47000036239624,
      "epoch": 4.183560262228946,
      "grad_norm": 1.28125,
      "learning_rate": 0.00037498606363060285,
      "loss": 4.97,
      "mean_token_accuracy": 0.18873539119958876,
      "num_tokens": 9163772.0,
      "step": 4150
    },
    {
      "entropy": 5.480148839950561,
      "epoch": 4.188603126575895,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003746305605322041,
      "loss": 4.9838,
      "mean_token_accuracy": 0.18592649549245835,
      "num_tokens": 9174618.0,
      "step": 4155
    },
    {
      "entropy": 5.554883766174316,
      "epoch": 4.193645990922844,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00037427474777972477,
      "loss": 5.0381,
      "mean_token_accuracy": 0.18470816016197206,
      "num_tokens": 9185199.0,
      "step": 4160
    },
    {
      "entropy": 5.4969367504119875,
      "epoch": 4.198688855269793,
      "grad_norm": 0.96875,
      "learning_rate": 0.00037391862647903943,
      "loss": 4.9341,
      "mean_token_accuracy": 0.18754270523786545,
      "num_tokens": 9197329.0,
      "step": 4165
    },
    {
      "entropy": 5.4007669448852536,
      "epoch": 4.2037317196167425,
      "grad_norm": 1.140625,
      "learning_rate": 0.00037356219773698106,
      "loss": 4.9124,
      "mean_token_accuracy": 0.1893045112490654,
      "num_tokens": 9207516.0,
      "step": 4170
    },
    {
      "entropy": 5.4253232955932615,
      "epoch": 4.208774583963692,
      "grad_norm": 1.15625,
      "learning_rate": 0.00037320546266133854,
      "loss": 4.9827,
      "mean_token_accuracy": 0.17612583488225936,
      "num_tokens": 9219598.0,
      "step": 4175
    },
    {
      "entropy": 5.411268997192383,
      "epoch": 4.213817448310641,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003728484223608526,
      "loss": 4.8635,
      "mean_token_accuracy": 0.19200971871614456,
      "num_tokens": 9229675.0,
      "step": 4180
    },
    {
      "entropy": 5.383372116088867,
      "epoch": 4.21886031265759,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00037249107794521286,
      "loss": 4.9246,
      "mean_token_accuracy": 0.1911775916814804,
      "num_tokens": 9240308.0,
      "step": 4185
    },
    {
      "entropy": 5.377345848083496,
      "epoch": 4.223903177004539,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00037213343052505394,
      "loss": 4.8889,
      "mean_token_accuracy": 0.19183996617794036,
      "num_tokens": 9250299.0,
      "step": 4190
    },
    {
      "entropy": 5.43930835723877,
      "epoch": 4.228946041351487,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003717754812119523,
      "loss": 4.9551,
      "mean_token_accuracy": 0.1842781648039818,
      "num_tokens": 9262659.0,
      "step": 4195
    },
    {
      "entropy": 5.4641939163208,
      "epoch": 4.233988905698436,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003714172311184226,
      "loss": 4.8958,
      "mean_token_accuracy": 0.19679191559553147,
      "num_tokens": 9274430.0,
      "step": 4200
    },
    {
      "entropy": 5.434781312942505,
      "epoch": 4.2390317700453854,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003710586813579144,
      "loss": 4.9931,
      "mean_token_accuracy": 0.18488436937332153,
      "num_tokens": 9285484.0,
      "step": 4205
    },
    {
      "entropy": 5.4645696640014645,
      "epoch": 4.244074634392335,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00037069983304480875,
      "loss": 4.9115,
      "mean_token_accuracy": 0.19475334137678146,
      "num_tokens": 9296253.0,
      "step": 4210
    },
    {
      "entropy": 5.418675279617309,
      "epoch": 4.249117498739284,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003703406872944144,
      "loss": 4.9487,
      "mean_token_accuracy": 0.19164616465568543,
      "num_tokens": 9308491.0,
      "step": 4215
    },
    {
      "entropy": 5.496989727020264,
      "epoch": 4.254160363086233,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003699812452229645,
      "loss": 4.9287,
      "mean_token_accuracy": 0.18727146536111833,
      "num_tokens": 9319722.0,
      "step": 4220
    },
    {
      "entropy": 5.385287666320801,
      "epoch": 4.259203227433182,
      "grad_norm": 1.296875,
      "learning_rate": 0.00036962150794761354,
      "loss": 4.8471,
      "mean_token_accuracy": 0.2009499654173851,
      "num_tokens": 9329324.0,
      "step": 4225
    },
    {
      "entropy": 5.389019870758057,
      "epoch": 4.264246091780131,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003692614765864331,
      "loss": 4.9527,
      "mean_token_accuracy": 0.18734048753976823,
      "num_tokens": 9340060.0,
      "step": 4230
    },
    {
      "entropy": 5.435498523712158,
      "epoch": 4.26928895612708,
      "grad_norm": 1.140625,
      "learning_rate": 0.00036890115225840904,
      "loss": 4.9647,
      "mean_token_accuracy": 0.18994056880474092,
      "num_tokens": 9351071.0,
      "step": 4235
    },
    {
      "entropy": 5.498436164855957,
      "epoch": 4.274331820474029,
      "grad_norm": 1.25,
      "learning_rate": 0.0003685405360834376,
      "loss": 5.01,
      "mean_token_accuracy": 0.1808197483420372,
      "num_tokens": 9361910.0,
      "step": 4240
    },
    {
      "entropy": 5.487107276916504,
      "epoch": 4.279374684820978,
      "grad_norm": 1.296875,
      "learning_rate": 0.00036817962918232234,
      "loss": 4.9327,
      "mean_token_accuracy": 0.19063286185264589,
      "num_tokens": 9371964.0,
      "step": 4245
    },
    {
      "entropy": 5.422836875915527,
      "epoch": 4.2844175491679275,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00036781843267677,
      "loss": 4.9213,
      "mean_token_accuracy": 0.19161167442798616,
      "num_tokens": 9381922.0,
      "step": 4250
    },
    {
      "entropy": 5.39989275932312,
      "epoch": 4.289460413514877,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003674569476893878,
      "loss": 4.8587,
      "mean_token_accuracy": 0.2021702840924263,
      "num_tokens": 9393256.0,
      "step": 4255
    },
    {
      "entropy": 5.380013179779053,
      "epoch": 4.294503277861826,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003670951753436794,
      "loss": 4.9026,
      "mean_token_accuracy": 0.19248400777578353,
      "num_tokens": 9403703.0,
      "step": 4260
    },
    {
      "entropy": 5.500961351394653,
      "epoch": 4.299546142208775,
      "grad_norm": 1.203125,
      "learning_rate": 0.00036673311676404163,
      "loss": 4.9738,
      "mean_token_accuracy": 0.19227057546377183,
      "num_tokens": 9414454.0,
      "step": 4265
    },
    {
      "entropy": 5.360117864608765,
      "epoch": 4.304589006555724,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003663707730757608,
      "loss": 4.8901,
      "mean_token_accuracy": 0.19245599955320358,
      "num_tokens": 9425566.0,
      "step": 4270
    },
    {
      "entropy": 5.412634181976318,
      "epoch": 4.309631870902673,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00036600814540500956,
      "loss": 5.0211,
      "mean_token_accuracy": 0.1835702806711197,
      "num_tokens": 9435555.0,
      "step": 4275
    },
    {
      "entropy": 5.554074096679687,
      "epoch": 4.314674735249622,
      "grad_norm": 1.203125,
      "learning_rate": 0.000365645234878843,
      "loss": 5.0301,
      "mean_token_accuracy": 0.17809366285800934,
      "num_tokens": 9447784.0,
      "step": 4280
    },
    {
      "entropy": 5.459975814819336,
      "epoch": 4.31971759959657,
      "grad_norm": 1.265625,
      "learning_rate": 0.00036528204262519545,
      "loss": 4.9313,
      "mean_token_accuracy": 0.18740205019712447,
      "num_tokens": 9458243.0,
      "step": 4285
    },
    {
      "entropy": 5.425280332565308,
      "epoch": 4.3247604639435195,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003649185697728768,
      "loss": 4.9322,
      "mean_token_accuracy": 0.18356121331453323,
      "num_tokens": 9470258.0,
      "step": 4290
    },
    {
      "entropy": 5.445868730545044,
      "epoch": 4.329803328290469,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00036455481745156906,
      "loss": 4.9322,
      "mean_token_accuracy": 0.19494981765747071,
      "num_tokens": 9480787.0,
      "step": 4295
    },
    {
      "entropy": 5.440693378448486,
      "epoch": 4.334846192637418,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003641907867918228,
      "loss": 4.921,
      "mean_token_accuracy": 0.1914638862013817,
      "num_tokens": 9492459.0,
      "step": 4300
    },
    {
      "entropy": 5.508681201934815,
      "epoch": 4.339889056984367,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003638264789250538,
      "loss": 5.0177,
      "mean_token_accuracy": 0.1854952722787857,
      "num_tokens": 9504600.0,
      "step": 4305
    },
    {
      "entropy": 5.482989549636841,
      "epoch": 4.344931921331316,
      "grad_norm": 1.484375,
      "learning_rate": 0.00036346189498353914,
      "loss": 5.024,
      "mean_token_accuracy": 0.1867216095328331,
      "num_tokens": 9515853.0,
      "step": 4310
    },
    {
      "entropy": 5.46557240486145,
      "epoch": 4.349974785678265,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003630970361004143,
      "loss": 4.9852,
      "mean_token_accuracy": 0.18903328180313111,
      "num_tokens": 9528109.0,
      "step": 4315
    },
    {
      "entropy": 5.4458990573883055,
      "epoch": 4.355017650025214,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003627319034096691,
      "loss": 4.9491,
      "mean_token_accuracy": 0.18713343888521194,
      "num_tokens": 9539795.0,
      "step": 4320
    },
    {
      "entropy": 5.452094459533692,
      "epoch": 4.360060514372163,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003623664980461443,
      "loss": 4.9656,
      "mean_token_accuracy": 0.1839947983622551,
      "num_tokens": 9551020.0,
      "step": 4325
    },
    {
      "entropy": 5.463205575942993,
      "epoch": 4.3651033787191125,
      "grad_norm": 1.34375,
      "learning_rate": 0.00036200082114552806,
      "loss": 4.9622,
      "mean_token_accuracy": 0.190327425301075,
      "num_tokens": 9562540.0,
      "step": 4330
    },
    {
      "entropy": 5.49603157043457,
      "epoch": 4.370146243066062,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00036163487384435294,
      "loss": 4.9805,
      "mean_token_accuracy": 0.18498869240283966,
      "num_tokens": 9574122.0,
      "step": 4335
    },
    {
      "entropy": 5.445807695388794,
      "epoch": 4.375189107413011,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003612686572799914,
      "loss": 4.9757,
      "mean_token_accuracy": 0.1899045467376709,
      "num_tokens": 9584781.0,
      "step": 4340
    },
    {
      "entropy": 5.53016996383667,
      "epoch": 4.38023197175996,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003609021725906529,
      "loss": 5.0002,
      "mean_token_accuracy": 0.19231095612049104,
      "num_tokens": 9595230.0,
      "step": 4345
    },
    {
      "entropy": 5.497966289520264,
      "epoch": 4.385274836106909,
      "grad_norm": 1.984375,
      "learning_rate": 0.00036053542091538045,
      "loss": 4.9279,
      "mean_token_accuracy": 0.19131026268005372,
      "num_tokens": 9605814.0,
      "step": 4350
    },
    {
      "entropy": 5.3784668922424315,
      "epoch": 4.390317700453858,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00036016840339404664,
      "loss": 4.9158,
      "mean_token_accuracy": 0.19189533293247224,
      "num_tokens": 9617900.0,
      "step": 4355
    },
    {
      "entropy": 5.418680334091187,
      "epoch": 4.395360564800807,
      "grad_norm": 1.125,
      "learning_rate": 0.0003598011211673505,
      "loss": 4.9508,
      "mean_token_accuracy": 0.1888875335454941,
      "num_tokens": 9628987.0,
      "step": 4360
    },
    {
      "entropy": 5.45054874420166,
      "epoch": 4.400403429147756,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003594335753768135,
      "loss": 4.9168,
      "mean_token_accuracy": 0.19449739903211594,
      "num_tokens": 9638419.0,
      "step": 4365
    },
    {
      "entropy": 5.406636047363281,
      "epoch": 4.405446293494705,
      "grad_norm": 1.25,
      "learning_rate": 0.0003590657671647766,
      "loss": 4.9426,
      "mean_token_accuracy": 0.1835890844464302,
      "num_tokens": 9648975.0,
      "step": 4370
    },
    {
      "entropy": 5.414427900314331,
      "epoch": 4.4104891578416545,
      "grad_norm": 1.234375,
      "learning_rate": 0.00035869769767439606,
      "loss": 4.9027,
      "mean_token_accuracy": 0.19657850861549378,
      "num_tokens": 9659935.0,
      "step": 4375
    },
    {
      "entropy": 5.521138048171997,
      "epoch": 4.415532022188603,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003583293680496404,
      "loss": 5.0325,
      "mean_token_accuracy": 0.1821470484137535,
      "num_tokens": 9671269.0,
      "step": 4380
    },
    {
      "entropy": 5.4268065929412845,
      "epoch": 4.420574886535552,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003579607794352867,
      "loss": 4.9344,
      "mean_token_accuracy": 0.18964253813028337,
      "num_tokens": 9682627.0,
      "step": 4385
    },
    {
      "entropy": 5.373068904876709,
      "epoch": 4.425617750882501,
      "grad_norm": 1.28125,
      "learning_rate": 0.00035759193297691674,
      "loss": 4.8748,
      "mean_token_accuracy": 0.18809968382120132,
      "num_tokens": 9693482.0,
      "step": 4390
    },
    {
      "entropy": 5.4125419616699215,
      "epoch": 4.43066061522945,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00035722282982091373,
      "loss": 4.9653,
      "mean_token_accuracy": 0.19016257524490357,
      "num_tokens": 9705279.0,
      "step": 4395
    },
    {
      "entropy": 5.464582633972168,
      "epoch": 4.435703479576399,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003568534711144591,
      "loss": 4.9737,
      "mean_token_accuracy": 0.18579922914505004,
      "num_tokens": 9716695.0,
      "step": 4400
    },
    {
      "entropy": 5.410967016220093,
      "epoch": 4.440746343923348,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000356483858005528,
      "loss": 4.9055,
      "mean_token_accuracy": 0.19376523196697235,
      "num_tokens": 9727960.0,
      "step": 4405
    },
    {
      "entropy": 5.398622655868531,
      "epoch": 4.445789208270297,
      "grad_norm": 1.140625,
      "learning_rate": 0.00035611399164288654,
      "loss": 4.9838,
      "mean_token_accuracy": 0.18477021008729935,
      "num_tokens": 9738341.0,
      "step": 4410
    },
    {
      "entropy": 5.425491428375244,
      "epoch": 4.450832072617247,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003557438731760879,
      "loss": 4.9792,
      "mean_token_accuracy": 0.1845449224114418,
      "num_tokens": 9748261.0,
      "step": 4415
    },
    {
      "entropy": 5.485393524169922,
      "epoch": 4.455874936964196,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003553735037554689,
      "loss": 4.9788,
      "mean_token_accuracy": 0.18714309930801393,
      "num_tokens": 9759109.0,
      "step": 4420
    },
    {
      "entropy": 5.460876893997193,
      "epoch": 4.460917801311145,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003550028845321461,
      "loss": 4.921,
      "mean_token_accuracy": 0.1924830511212349,
      "num_tokens": 9770431.0,
      "step": 4425
    },
    {
      "entropy": 5.413283014297486,
      "epoch": 4.465960665658094,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003546320166580127,
      "loss": 4.9825,
      "mean_token_accuracy": 0.1831095799803734,
      "num_tokens": 9783459.0,
      "step": 4430
    },
    {
      "entropy": 5.460632276535034,
      "epoch": 4.471003530005043,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00035426090128573464,
      "loss": 4.9083,
      "mean_token_accuracy": 0.19544671326875687,
      "num_tokens": 9793961.0,
      "step": 4435
    },
    {
      "entropy": 5.438150930404663,
      "epoch": 4.476046394351992,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00035388953956874686,
      "loss": 4.9672,
      "mean_token_accuracy": 0.18613309860229493,
      "num_tokens": 9803964.0,
      "step": 4440
    },
    {
      "entropy": 5.419139814376831,
      "epoch": 4.481089258698941,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003535179326612504,
      "loss": 4.8909,
      "mean_token_accuracy": 0.1951744630932808,
      "num_tokens": 9816176.0,
      "step": 4445
    },
    {
      "entropy": 5.346704769134521,
      "epoch": 4.48613212304589,
      "grad_norm": 1.25,
      "learning_rate": 0.00035314608171820787,
      "loss": 4.8713,
      "mean_token_accuracy": 0.19878727942705154,
      "num_tokens": 9826649.0,
      "step": 4450
    },
    {
      "entropy": 5.475081491470337,
      "epoch": 4.4911749873928395,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00035277398789534075,
      "loss": 5.0229,
      "mean_token_accuracy": 0.18393101841211318,
      "num_tokens": 9838396.0,
      "step": 4455
    },
    {
      "entropy": 5.518260955810547,
      "epoch": 4.496217851739789,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003524016523491252,
      "loss": 4.9729,
      "mean_token_accuracy": 0.18730095028877258,
      "num_tokens": 9848863.0,
      "step": 4460
    },
    {
      "entropy": 5.357536458969117,
      "epoch": 4.501260716086737,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00035202907623678865,
      "loss": 4.8922,
      "mean_token_accuracy": 0.1869596868753433,
      "num_tokens": 9860673.0,
      "step": 4465
    },
    {
      "entropy": 5.409299659729004,
      "epoch": 4.506303580433686,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003516562607163062,
      "loss": 4.9645,
      "mean_token_accuracy": 0.1865547701716423,
      "num_tokens": 9871246.0,
      "step": 4470
    },
    {
      "entropy": 5.46815505027771,
      "epoch": 4.511346444780635,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003512832069463971,
      "loss": 5.0109,
      "mean_token_accuracy": 0.1803091734647751,
      "num_tokens": 9883352.0,
      "step": 4475
    },
    {
      "entropy": 5.445433473587036,
      "epoch": 4.516389309127584,
      "grad_norm": 1.3125,
      "learning_rate": 0.00035090991608652125,
      "loss": 5.0148,
      "mean_token_accuracy": 0.18997027575969697,
      "num_tokens": 9894797.0,
      "step": 4480
    },
    {
      "entropy": 5.44483757019043,
      "epoch": 4.521432173474533,
      "grad_norm": 1.1875,
      "learning_rate": 0.00035053638929687504,
      "loss": 4.9461,
      "mean_token_accuracy": 0.18382203578948975,
      "num_tokens": 9905618.0,
      "step": 4485
    },
    {
      "entropy": 5.367643260955811,
      "epoch": 4.526475037821482,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003501626277383884,
      "loss": 4.9569,
      "mean_token_accuracy": 0.1849820211529732,
      "num_tokens": 9916817.0,
      "step": 4490
    },
    {
      "entropy": 5.436738920211792,
      "epoch": 4.5315179021684315,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000349788632572721,
      "loss": 4.949,
      "mean_token_accuracy": 0.18787739872932435,
      "num_tokens": 9929052.0,
      "step": 4495
    },
    {
      "entropy": 5.435274410247803,
      "epoch": 4.536560766515381,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003494144049622582,
      "loss": 4.8384,
      "mean_token_accuracy": 0.19309654235839843,
      "num_tokens": 9939513.0,
      "step": 4500
    },
    {
      "epoch": 4.536560766515381,
      "eval_entropy": 5.294977138363981,
      "eval_loss": 5.679702281951904,
      "eval_mean_token_accuracy": 0.15328879043952806,
      "eval_num_tokens": 9939513.0,
      "eval_runtime": 2.2396,
      "eval_samples_per_second": 1401.154,
      "eval_steps_per_second": 175.479,
      "step": 4500
    },
    {
      "entropy": 5.358317470550537,
      "epoch": 4.54160363086233,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003490399460701083,
      "loss": 4.911,
      "mean_token_accuracy": 0.1925751104950905,
      "num_tokens": 9950149.0,
      "step": 4505
    },
    {
      "entropy": 5.5022735595703125,
      "epoch": 4.546646495209279,
      "grad_norm": 1.375,
      "learning_rate": 0.000348665257060098,
      "loss": 4.9664,
      "mean_token_accuracy": 0.18934469372034074,
      "num_tokens": 9961696.0,
      "step": 4510
    },
    {
      "entropy": 5.462430191040039,
      "epoch": 4.551689359556228,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003482903390967695,
      "loss": 4.9925,
      "mean_token_accuracy": 0.1850539818406105,
      "num_tokens": 9971446.0,
      "step": 4515
    },
    {
      "entropy": 5.423494434356689,
      "epoch": 4.556732223903177,
      "grad_norm": 1.15625,
      "learning_rate": 0.00034791519334537635,
      "loss": 4.9119,
      "mean_token_accuracy": 0.19724505096673967,
      "num_tokens": 9982992.0,
      "step": 4520
    },
    {
      "entropy": 5.391956424713134,
      "epoch": 4.561775088250126,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003475398209718804,
      "loss": 4.8666,
      "mean_token_accuracy": 0.1952480763196945,
      "num_tokens": 9993443.0,
      "step": 4525
    },
    {
      "entropy": 5.397252750396729,
      "epoch": 4.566817952597075,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003471642231429475,
      "loss": 4.9905,
      "mean_token_accuracy": 0.1915498659014702,
      "num_tokens": 10004531.0,
      "step": 4530
    },
    {
      "entropy": 5.453794050216675,
      "epoch": 4.5718608169440245,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003467884010259443,
      "loss": 4.9232,
      "mean_token_accuracy": 0.19060313999652861,
      "num_tokens": 10016455.0,
      "step": 4535
    },
    {
      "entropy": 5.494798374176026,
      "epoch": 4.576903681290974,
      "grad_norm": 1.125,
      "learning_rate": 0.00034641235578893476,
      "loss": 4.8812,
      "mean_token_accuracy": 0.19779495298862457,
      "num_tokens": 10028067.0,
      "step": 4540
    },
    {
      "entropy": 5.483316278457641,
      "epoch": 4.581946545637923,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003460360886006761,
      "loss": 5.047,
      "mean_token_accuracy": 0.1896531656384468,
      "num_tokens": 10038615.0,
      "step": 4545
    },
    {
      "entropy": 5.428381443023682,
      "epoch": 4.586989409984872,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00034565960063061534,
      "loss": 4.9863,
      "mean_token_accuracy": 0.1903046950697899,
      "num_tokens": 10050022.0,
      "step": 4550
    },
    {
      "entropy": 5.351579761505127,
      "epoch": 4.592032274331821,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003452828930488859,
      "loss": 4.9246,
      "mean_token_accuracy": 0.18745099902153015,
      "num_tokens": 10062565.0,
      "step": 4555
    },
    {
      "entropy": 5.544126701354981,
      "epoch": 4.597075138678769,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00034490596702630355,
      "loss": 5.0781,
      "mean_token_accuracy": 0.18337109684944153,
      "num_tokens": 10075553.0,
      "step": 4560
    },
    {
      "entropy": 5.504072093963623,
      "epoch": 4.602118003025718,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003445288237343632,
      "loss": 4.9734,
      "mean_token_accuracy": 0.18662497252225876,
      "num_tokens": 10086144.0,
      "step": 4565
    },
    {
      "entropy": 5.401547574996949,
      "epoch": 4.607160867372667,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00034415146434523463,
      "loss": 4.9723,
      "mean_token_accuracy": 0.18603927940130233,
      "num_tokens": 10097218.0,
      "step": 4570
    },
    {
      "entropy": 5.355431795120239,
      "epoch": 4.6122037317196165,
      "grad_norm": 1.265625,
      "learning_rate": 0.00034377389003175984,
      "loss": 4.9598,
      "mean_token_accuracy": 0.17987670749425888,
      "num_tokens": 10107614.0,
      "step": 4575
    },
    {
      "entropy": 5.496437644958496,
      "epoch": 4.617246596066566,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003433961019674483,
      "loss": 5.0034,
      "mean_token_accuracy": 0.1837390646338463,
      "num_tokens": 10118976.0,
      "step": 4580
    },
    {
      "entropy": 5.497903442382812,
      "epoch": 4.622289460413515,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003430181013264741,
      "loss": 4.9553,
      "mean_token_accuracy": 0.1883614182472229,
      "num_tokens": 10129914.0,
      "step": 4585
    },
    {
      "entropy": 5.47073884010315,
      "epoch": 4.627332324760464,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003426398892836719,
      "loss": 4.9677,
      "mean_token_accuracy": 0.18729270100593567,
      "num_tokens": 10139490.0,
      "step": 4590
    },
    {
      "entropy": 5.4461322784423825,
      "epoch": 4.632375189107413,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003422614670145336,
      "loss": 5.0042,
      "mean_token_accuracy": 0.191692017018795,
      "num_tokens": 10149874.0,
      "step": 4595
    },
    {
      "entropy": 5.436661720275879,
      "epoch": 4.637418053454362,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003418828356952041,
      "loss": 5.0216,
      "mean_token_accuracy": 0.1882268726825714,
      "num_tokens": 10161373.0,
      "step": 4600
    },
    {
      "entropy": 5.41108283996582,
      "epoch": 4.642460917801311,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00034150399650247846,
      "loss": 4.9484,
      "mean_token_accuracy": 0.19099740386009217,
      "num_tokens": 10172044.0,
      "step": 4605
    },
    {
      "entropy": 5.413085460662842,
      "epoch": 4.64750378214826,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00034112495061379764,
      "loss": 4.958,
      "mean_token_accuracy": 0.1898003786802292,
      "num_tokens": 10183127.0,
      "step": 4610
    },
    {
      "entropy": 5.4633465766906735,
      "epoch": 4.652546646495209,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00034074569920724483,
      "loss": 4.9854,
      "mean_token_accuracy": 0.1862984448671341,
      "num_tokens": 10194366.0,
      "step": 4615
    },
    {
      "entropy": 5.459962701797485,
      "epoch": 4.6575895108421586,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003403662434615423,
      "loss": 4.9715,
      "mean_token_accuracy": 0.18344075083732606,
      "num_tokens": 10205655.0,
      "step": 4620
    },
    {
      "entropy": 5.445805454254151,
      "epoch": 4.662632375189108,
      "grad_norm": 1.40625,
      "learning_rate": 0.00033998658455604736,
      "loss": 4.9418,
      "mean_token_accuracy": 0.19574547410011292,
      "num_tokens": 10215596.0,
      "step": 4625
    },
    {
      "entropy": 5.365432977676392,
      "epoch": 4.667675239536057,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00033960672367074827,
      "loss": 4.9446,
      "mean_token_accuracy": 0.19192101806402206,
      "num_tokens": 10225298.0,
      "step": 4630
    },
    {
      "entropy": 5.548795557022094,
      "epoch": 4.672718103883006,
      "grad_norm": 1.125,
      "learning_rate": 0.0003392266619862619,
      "loss": 5.0535,
      "mean_token_accuracy": 0.18375632762908936,
      "num_tokens": 10236979.0,
      "step": 4635
    },
    {
      "entropy": 5.453441333770752,
      "epoch": 4.677760968229954,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003388464006838284,
      "loss": 4.9085,
      "mean_token_accuracy": 0.18680524230003356,
      "num_tokens": 10248109.0,
      "step": 4640
    },
    {
      "entropy": 5.405192136764526,
      "epoch": 4.682803832576903,
      "grad_norm": 1.328125,
      "learning_rate": 0.00033846594094530874,
      "loss": 4.9901,
      "mean_token_accuracy": 0.19211216717958451,
      "num_tokens": 10258931.0,
      "step": 4645
    },
    {
      "entropy": 5.411888790130615,
      "epoch": 4.687846696923852,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003380852839531806,
      "loss": 5.031,
      "mean_token_accuracy": 0.18861302137374877,
      "num_tokens": 10272521.0,
      "step": 4650
    },
    {
      "entropy": 5.433878374099732,
      "epoch": 4.6928895612708015,
      "grad_norm": 1.21875,
      "learning_rate": 0.00033770443089053476,
      "loss": 4.9248,
      "mean_token_accuracy": 0.18855932354927063,
      "num_tokens": 10283481.0,
      "step": 4655
    },
    {
      "entropy": 5.432025480270386,
      "epoch": 4.697932425617751,
      "grad_norm": 1.15625,
      "learning_rate": 0.00033732338294107116,
      "loss": 4.9759,
      "mean_token_accuracy": 0.1873525321483612,
      "num_tokens": 10294258.0,
      "step": 4660
    },
    {
      "entropy": 5.475202655792236,
      "epoch": 4.7029752899647,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00033694214128909567,
      "loss": 4.9627,
      "mean_token_accuracy": 0.1921197310090065,
      "num_tokens": 10304464.0,
      "step": 4665
    },
    {
      "entropy": 5.439158201217651,
      "epoch": 4.708018154311649,
      "grad_norm": 1.25,
      "learning_rate": 0.00033656070711951633,
      "loss": 4.9673,
      "mean_token_accuracy": 0.18413261622190474,
      "num_tokens": 10316343.0,
      "step": 4670
    },
    {
      "entropy": 5.4633959293365475,
      "epoch": 4.713061018658598,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003361790816178391,
      "loss": 5.0364,
      "mean_token_accuracy": 0.18760841637849807,
      "num_tokens": 10327057.0,
      "step": 4675
    },
    {
      "entropy": 5.473525762557983,
      "epoch": 4.718103883005547,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003357972659701649,
      "loss": 5.0091,
      "mean_token_accuracy": 0.1868601068854332,
      "num_tokens": 10338483.0,
      "step": 4680
    },
    {
      "entropy": 5.445736265182495,
      "epoch": 4.723146747352496,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00033541526136318564,
      "loss": 4.9241,
      "mean_token_accuracy": 0.19348323345184326,
      "num_tokens": 10349522.0,
      "step": 4685
    },
    {
      "entropy": 5.396135997772217,
      "epoch": 4.728189611699445,
      "grad_norm": 0.953125,
      "learning_rate": 0.0003350330689841804,
      "loss": 4.9227,
      "mean_token_accuracy": 0.18947552144527435,
      "num_tokens": 10361036.0,
      "step": 4690
    },
    {
      "entropy": 5.463840341567993,
      "epoch": 4.733232476046394,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003346506900210118,
      "loss": 5.053,
      "mean_token_accuracy": 0.18605293333530426,
      "num_tokens": 10372673.0,
      "step": 4695
    },
    {
      "entropy": 5.457823753356934,
      "epoch": 4.7382753403933435,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003342681256621227,
      "loss": 5.0761,
      "mean_token_accuracy": 0.18717198371887206,
      "num_tokens": 10384100.0,
      "step": 4700
    },
    {
      "entropy": 5.461978578567505,
      "epoch": 4.743318204740293,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003338853770965319,
      "loss": 4.9693,
      "mean_token_accuracy": 0.18431959450244903,
      "num_tokens": 10395503.0,
      "step": 4705
    },
    {
      "entropy": 5.427508687973022,
      "epoch": 4.748361069087242,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00033350244551383064,
      "loss": 4.9225,
      "mean_token_accuracy": 0.19294063448905946,
      "num_tokens": 10406147.0,
      "step": 4710
    },
    {
      "entropy": 5.410728693008423,
      "epoch": 4.753403933434191,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003331193321041793,
      "loss": 4.8714,
      "mean_token_accuracy": 0.1944505140185356,
      "num_tokens": 10416659.0,
      "step": 4715
    },
    {
      "entropy": 5.431481075286865,
      "epoch": 4.75844679778114,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00033273603805830296,
      "loss": 4.9736,
      "mean_token_accuracy": 0.18503912538290024,
      "num_tokens": 10427765.0,
      "step": 4720
    },
    {
      "entropy": 5.408118677139282,
      "epoch": 4.763489662128089,
      "grad_norm": 1.203125,
      "learning_rate": 0.00033235256456748847,
      "loss": 4.9105,
      "mean_token_accuracy": 0.19530655741691588,
      "num_tokens": 10439376.0,
      "step": 4725
    },
    {
      "entropy": 5.394193983078003,
      "epoch": 4.768532526475038,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003319689128235804,
      "loss": 4.993,
      "mean_token_accuracy": 0.1818249374628067,
      "num_tokens": 10449232.0,
      "step": 4730
    },
    {
      "entropy": 5.352348327636719,
      "epoch": 4.773575390821987,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00033158508401897707,
      "loss": 4.9152,
      "mean_token_accuracy": 0.1893765151500702,
      "num_tokens": 10459258.0,
      "step": 4735
    },
    {
      "entropy": 5.429558038711548,
      "epoch": 4.778618255168936,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003312010793466273,
      "loss": 4.985,
      "mean_token_accuracy": 0.18403921276330948,
      "num_tokens": 10469909.0,
      "step": 4740
    },
    {
      "entropy": 5.471266555786133,
      "epoch": 4.783661119515885,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003308169000000265,
      "loss": 4.9269,
      "mean_token_accuracy": 0.1910149171948433,
      "num_tokens": 10479848.0,
      "step": 4745
    },
    {
      "entropy": 5.450354051589966,
      "epoch": 4.788703983862834,
      "grad_norm": 1.125,
      "learning_rate": 0.0003304325471732129,
      "loss": 4.9751,
      "mean_token_accuracy": 0.18426334261894226,
      "num_tokens": 10490352.0,
      "step": 4750
    },
    {
      "entropy": 5.373760604858399,
      "epoch": 4.793746848209783,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00033004802206076396,
      "loss": 4.8842,
      "mean_token_accuracy": 0.19841017127037047,
      "num_tokens": 10501905.0,
      "step": 4755
    },
    {
      "entropy": 5.424583530426025,
      "epoch": 4.798789712556732,
      "grad_norm": 1.125,
      "learning_rate": 0.00032966332585779263,
      "loss": 5.0382,
      "mean_token_accuracy": 0.18531617969274522,
      "num_tokens": 10512491.0,
      "step": 4760
    },
    {
      "entropy": 5.400948858261108,
      "epoch": 4.803832576903681,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00032927845975994356,
      "loss": 4.8301,
      "mean_token_accuracy": 0.20473436117172242,
      "num_tokens": 10522686.0,
      "step": 4765
    },
    {
      "entropy": 5.369522285461426,
      "epoch": 4.80887544125063,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00032889342496338953,
      "loss": 4.9439,
      "mean_token_accuracy": 0.1866819977760315,
      "num_tokens": 10534240.0,
      "step": 4770
    },
    {
      "entropy": 5.434429931640625,
      "epoch": 4.813918305597579,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003285082226648273,
      "loss": 5.0481,
      "mean_token_accuracy": 0.1853307381272316,
      "num_tokens": 10545096.0,
      "step": 4775
    },
    {
      "entropy": 5.540076541900635,
      "epoch": 4.8189611699445285,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003281228540614749,
      "loss": 5.0373,
      "mean_token_accuracy": 0.18597427755594254,
      "num_tokens": 10557076.0,
      "step": 4780
    },
    {
      "entropy": 5.357098960876465,
      "epoch": 4.824004034291478,
      "grad_norm": 1.203125,
      "learning_rate": 0.00032773732035106657,
      "loss": 4.8505,
      "mean_token_accuracy": 0.19171000868082047,
      "num_tokens": 10567635.0,
      "step": 4785
    },
    {
      "entropy": 5.456152677536011,
      "epoch": 4.829046898638427,
      "grad_norm": 1.171875,
      "learning_rate": 0.00032735162273185006,
      "loss": 5.0627,
      "mean_token_accuracy": 0.17826583087444306,
      "num_tokens": 10579488.0,
      "step": 4790
    },
    {
      "entropy": 5.418813037872314,
      "epoch": 4.834089762985376,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003269657624025824,
      "loss": 4.9361,
      "mean_token_accuracy": 0.1880239725112915,
      "num_tokens": 10591092.0,
      "step": 4795
    },
    {
      "entropy": 5.428745460510254,
      "epoch": 4.839132627332325,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003265797405625265,
      "loss": 4.9873,
      "mean_token_accuracy": 0.18475101441144942,
      "num_tokens": 10603165.0,
      "step": 4800
    },
    {
      "entropy": 5.363924837112426,
      "epoch": 4.844175491679274,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003261935584114471,
      "loss": 4.8862,
      "mean_token_accuracy": 0.2026914030313492,
      "num_tokens": 10615149.0,
      "step": 4805
    },
    {
      "entropy": 5.503559160232544,
      "epoch": 4.849218356026223,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003258072171496072,
      "loss": 5.0445,
      "mean_token_accuracy": 0.18933772742748262,
      "num_tokens": 10626786.0,
      "step": 4810
    },
    {
      "entropy": 5.456537437438965,
      "epoch": 4.854261220373172,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003254207179777644,
      "loss": 4.9407,
      "mean_token_accuracy": 0.18621895462274551,
      "num_tokens": 10637107.0,
      "step": 4815
    },
    {
      "entropy": 5.359054851531982,
      "epoch": 4.8593040847201205,
      "grad_norm": 1.125,
      "learning_rate": 0.000325034062097167,
      "loss": 4.9384,
      "mean_token_accuracy": 0.192344531416893,
      "num_tokens": 10648198.0,
      "step": 4820
    },
    {
      "entropy": 5.492145442962647,
      "epoch": 4.86434694906707,
      "grad_norm": 1.234375,
      "learning_rate": 0.00032464725070955046,
      "loss": 5.0903,
      "mean_token_accuracy": 0.18057768195867538,
      "num_tokens": 10659867.0,
      "step": 4825
    },
    {
      "entropy": 5.49399585723877,
      "epoch": 4.869389813414019,
      "grad_norm": 1.0625,
      "learning_rate": 0.00032426028501713347,
      "loss": 4.9969,
      "mean_token_accuracy": 0.18500359803438188,
      "num_tokens": 10672117.0,
      "step": 4830
    },
    {
      "entropy": 5.42034945487976,
      "epoch": 4.874432677760968,
      "grad_norm": 1.1875,
      "learning_rate": 0.00032387316622261425,
      "loss": 4.9103,
      "mean_token_accuracy": 0.19433372765779494,
      "num_tokens": 10682014.0,
      "step": 4835
    },
    {
      "entropy": 5.456826162338257,
      "epoch": 4.879475542107917,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00032348589552916703,
      "loss": 5.0684,
      "mean_token_accuracy": 0.1784452974796295,
      "num_tokens": 10693330.0,
      "step": 4840
    },
    {
      "entropy": 5.4022890567779545,
      "epoch": 4.884518406454866,
      "grad_norm": 1.25,
      "learning_rate": 0.0003230984741404381,
      "loss": 4.9023,
      "mean_token_accuracy": 0.19363944828510285,
      "num_tokens": 10703772.0,
      "step": 4845
    },
    {
      "entropy": 5.4632611751556395,
      "epoch": 4.889561270801815,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003227109032605419,
      "loss": 5.0003,
      "mean_token_accuracy": 0.18492232710123063,
      "num_tokens": 10714088.0,
      "step": 4850
    },
    {
      "entropy": 5.343895721435547,
      "epoch": 4.894604135148764,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003223231840940578,
      "loss": 4.8021,
      "mean_token_accuracy": 0.2004563570022583,
      "num_tokens": 10724701.0,
      "step": 4855
    },
    {
      "entropy": 5.389372539520264,
      "epoch": 4.8996469994957135,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00032193531784602595,
      "loss": 4.9355,
      "mean_token_accuracy": 0.1912728264927864,
      "num_tokens": 10735361.0,
      "step": 4860
    },
    {
      "entropy": 5.464755010604859,
      "epoch": 4.904689863842663,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00032154730572194346,
      "loss": 4.9352,
      "mean_token_accuracy": 0.18611158877611161,
      "num_tokens": 10745909.0,
      "step": 4865
    },
    {
      "entropy": 5.424462032318115,
      "epoch": 4.909732728189612,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000321159148927761,
      "loss": 5.0567,
      "mean_token_accuracy": 0.1866586148738861,
      "num_tokens": 10759305.0,
      "step": 4870
    },
    {
      "entropy": 5.391438817977905,
      "epoch": 4.914775592536561,
      "grad_norm": 1.15625,
      "learning_rate": 0.000320770848669879,
      "loss": 5.0068,
      "mean_token_accuracy": 0.19083456695079803,
      "num_tokens": 10770451.0,
      "step": 4875
    },
    {
      "entropy": 5.451483631134034,
      "epoch": 4.91981845688351,
      "grad_norm": 1.171875,
      "learning_rate": 0.00032038240615514345,
      "loss": 4.9314,
      "mean_token_accuracy": 0.1890146479010582,
      "num_tokens": 10782376.0,
      "step": 4880
    },
    {
      "entropy": 5.417233324050903,
      "epoch": 4.924861321230459,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003199938225908426,
      "loss": 5.0326,
      "mean_token_accuracy": 0.18465255796909333,
      "num_tokens": 10794252.0,
      "step": 4885
    },
    {
      "entropy": 5.442014122009278,
      "epoch": 4.929904185577408,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003196050991847033,
      "loss": 4.999,
      "mean_token_accuracy": 0.1788456544280052,
      "num_tokens": 10804117.0,
      "step": 4890
    },
    {
      "entropy": 5.430909061431885,
      "epoch": 4.934947049924357,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003192162371448868,
      "loss": 4.9353,
      "mean_token_accuracy": 0.19325827658176423,
      "num_tokens": 10815798.0,
      "step": 4895
    },
    {
      "entropy": 5.343411636352539,
      "epoch": 4.939989914271306,
      "grad_norm": 1.125,
      "learning_rate": 0.00031882723767998517,
      "loss": 4.9474,
      "mean_token_accuracy": 0.19216787219047546,
      "num_tokens": 10827135.0,
      "step": 4900
    },
    {
      "entropy": 5.471387577056885,
      "epoch": 4.9450327786182555,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00031843810199901795,
      "loss": 5.047,
      "mean_token_accuracy": 0.18531995862722397,
      "num_tokens": 10838988.0,
      "step": 4905
    },
    {
      "entropy": 5.429068899154663,
      "epoch": 4.950075642965205,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003180488313114277,
      "loss": 4.8728,
      "mean_token_accuracy": 0.18997331708669662,
      "num_tokens": 10848223.0,
      "step": 4910
    },
    {
      "entropy": 5.438881158828735,
      "epoch": 4.955118507312154,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003176594268270766,
      "loss": 4.9202,
      "mean_token_accuracy": 0.18719177097082138,
      "num_tokens": 10859052.0,
      "step": 4915
    },
    {
      "entropy": 5.408689260482788,
      "epoch": 4.960161371659103,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00031726988975624293,
      "loss": 4.9617,
      "mean_token_accuracy": 0.1857636407017708,
      "num_tokens": 10869108.0,
      "step": 4920
    },
    {
      "entropy": 5.439595890045166,
      "epoch": 4.965204236006051,
      "grad_norm": 1.140625,
      "learning_rate": 0.00031688022130961675,
      "loss": 4.9805,
      "mean_token_accuracy": 0.18924712538719177,
      "num_tokens": 10880691.0,
      "step": 4925
    },
    {
      "entropy": 5.3867377758026125,
      "epoch": 4.970247100353,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003164904226982967,
      "loss": 4.8872,
      "mean_token_accuracy": 0.18952036201953887,
      "num_tokens": 10892206.0,
      "step": 4930
    },
    {
      "entropy": 5.410093784332275,
      "epoch": 4.975289964699949,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00031610049513378574,
      "loss": 4.8446,
      "mean_token_accuracy": 0.2034765362739563,
      "num_tokens": 10902130.0,
      "step": 4935
    },
    {
      "entropy": 5.436312246322632,
      "epoch": 4.9803328290468984,
      "grad_norm": 1.109375,
      "learning_rate": 0.00031571043982798786,
      "loss": 4.985,
      "mean_token_accuracy": 0.18430037647485734,
      "num_tokens": 10913424.0,
      "step": 4940
    },
    {
      "entropy": 5.418987035751343,
      "epoch": 4.985375693393848,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003153202579932039,
      "loss": 5.0488,
      "mean_token_accuracy": 0.1842774897813797,
      "num_tokens": 10925718.0,
      "step": 4945
    },
    {
      "entropy": 5.401063776016235,
      "epoch": 4.990418557740797,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00031492995084212794,
      "loss": 4.9193,
      "mean_token_accuracy": 0.19016022384166717,
      "num_tokens": 10937102.0,
      "step": 4950
    },
    {
      "entropy": 5.401037645339966,
      "epoch": 4.995461422087746,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003145395195878437,
      "loss": 4.9224,
      "mean_token_accuracy": 0.1923469364643097,
      "num_tokens": 10948679.0,
      "step": 4955
    },
    {
      "entropy": 5.371797879536946,
      "epoch": 5.0,
      "grad_norm": 1.9609375,
      "learning_rate": 0.00031414896544382047,
      "loss": 4.8612,
      "mean_token_accuracy": 0.19405212667253283,
      "num_tokens": 10958200.0,
      "step": 4960
    },
    {
      "entropy": 5.422413110733032,
      "epoch": 5.005042864346949,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003137582896239095,
      "loss": 4.717,
      "mean_token_accuracy": 0.20915528684854506,
      "num_tokens": 10969775.0,
      "step": 4965
    },
    {
      "entropy": 5.3845212936401365,
      "epoch": 5.010085728693898,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003133674933423404,
      "loss": 4.6992,
      "mean_token_accuracy": 0.20794562995433807,
      "num_tokens": 10980423.0,
      "step": 4970
    },
    {
      "entropy": 5.353721332550049,
      "epoch": 5.015128593040847,
      "grad_norm": 1.109375,
      "learning_rate": 0.00031297657781371687,
      "loss": 4.7218,
      "mean_token_accuracy": 0.20617383122444152,
      "num_tokens": 10992716.0,
      "step": 4975
    },
    {
      "entropy": 5.4330377101898195,
      "epoch": 5.0201714573877965,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003125855442530135,
      "loss": 4.791,
      "mean_token_accuracy": 0.20222671031951905,
      "num_tokens": 11004597.0,
      "step": 4980
    },
    {
      "entropy": 5.439330244064331,
      "epoch": 5.025214321734746,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00031219439387557155,
      "loss": 4.7598,
      "mean_token_accuracy": 0.20851201862096785,
      "num_tokens": 11015423.0,
      "step": 4985
    },
    {
      "entropy": 5.4698091506958,
      "epoch": 5.030257186081695,
      "grad_norm": 1.109375,
      "learning_rate": 0.00031180312789709535,
      "loss": 4.7951,
      "mean_token_accuracy": 0.21242404133081436,
      "num_tokens": 11028191.0,
      "step": 4990
    },
    {
      "entropy": 5.393818044662476,
      "epoch": 5.035300050428644,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003114117475336487,
      "loss": 4.7968,
      "mean_token_accuracy": 0.20695600509643555,
      "num_tokens": 11040690.0,
      "step": 4995
    },
    {
      "entropy": 5.331779718399048,
      "epoch": 5.040342914775593,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003110202540016509,
      "loss": 4.6669,
      "mean_token_accuracy": 0.20674961805343628,
      "num_tokens": 11052580.0,
      "step": 5000
    },
    {
      "epoch": 5.040342914775593,
      "eval_entropy": 5.22243711845262,
      "eval_loss": 5.670998573303223,
      "eval_mean_token_accuracy": 0.15447733949386436,
      "eval_num_tokens": 11052580.0,
      "eval_runtime": 2.0811,
      "eval_samples_per_second": 1507.866,
      "eval_steps_per_second": 188.844,
      "step": 5000
    },
    {
      "entropy": 5.398833417892456,
      "epoch": 5.045385779122542,
      "grad_norm": 1.234375,
      "learning_rate": 0.00031062864851787267,
      "loss": 4.7135,
      "mean_token_accuracy": 0.2079475700855255,
      "num_tokens": 11063249.0,
      "step": 5005
    },
    {
      "entropy": 5.379339790344238,
      "epoch": 5.05042864346949,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00031023693229943294,
      "loss": 4.6809,
      "mean_token_accuracy": 0.20460328906774522,
      "num_tokens": 11073738.0,
      "step": 5010
    },
    {
      "entropy": 5.322503900527954,
      "epoch": 5.055471507816439,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003098451065637948,
      "loss": 4.7025,
      "mean_token_accuracy": 0.2008847638964653,
      "num_tokens": 11084555.0,
      "step": 5015
    },
    {
      "entropy": 5.3878766059875485,
      "epoch": 5.0605143721633885,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00030945317252876163,
      "loss": 4.7338,
      "mean_token_accuracy": 0.21230010241270064,
      "num_tokens": 11095947.0,
      "step": 5020
    },
    {
      "entropy": 5.40286021232605,
      "epoch": 5.065557236510338,
      "grad_norm": 1.171875,
      "learning_rate": 0.00030906113141247353,
      "loss": 4.6668,
      "mean_token_accuracy": 0.21439242959022523,
      "num_tokens": 11106332.0,
      "step": 5025
    },
    {
      "entropy": 5.417520332336426,
      "epoch": 5.070600100857287,
      "grad_norm": 1.203125,
      "learning_rate": 0.00030866898443340324,
      "loss": 4.7915,
      "mean_token_accuracy": 0.21110500395298004,
      "num_tokens": 11118396.0,
      "step": 5030
    },
    {
      "entropy": 5.396037197113037,
      "epoch": 5.075642965204236,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003082767328103527,
      "loss": 4.6943,
      "mean_token_accuracy": 0.2072602927684784,
      "num_tokens": 11128702.0,
      "step": 5035
    },
    {
      "entropy": 5.3590575695037845,
      "epoch": 5.080685829551185,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00030788437776244895,
      "loss": 4.6874,
      "mean_token_accuracy": 0.20847920179367066,
      "num_tokens": 11138004.0,
      "step": 5040
    },
    {
      "entropy": 5.346593618392944,
      "epoch": 5.085728693898134,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003074919205091405,
      "loss": 4.7456,
      "mean_token_accuracy": 0.21073976457118987,
      "num_tokens": 11149320.0,
      "step": 5045
    },
    {
      "entropy": 5.444488859176635,
      "epoch": 5.090771558245083,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003070993622701936,
      "loss": 4.7372,
      "mean_token_accuracy": 0.200305213034153,
      "num_tokens": 11160942.0,
      "step": 5050
    },
    {
      "entropy": 5.413809299468994,
      "epoch": 5.095814422592032,
      "grad_norm": 1.1875,
      "learning_rate": 0.00030670670426568837,
      "loss": 4.6988,
      "mean_token_accuracy": 0.20708912163972854,
      "num_tokens": 11172109.0,
      "step": 5055
    },
    {
      "entropy": 5.365613651275635,
      "epoch": 5.100857286938981,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003063139477160147,
      "loss": 4.7326,
      "mean_token_accuracy": 0.20575842559337615,
      "num_tokens": 11182455.0,
      "step": 5060
    },
    {
      "entropy": 5.340777730941772,
      "epoch": 5.1059001512859306,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00030592109384186935,
      "loss": 4.7978,
      "mean_token_accuracy": 0.20523615479469298,
      "num_tokens": 11194612.0,
      "step": 5065
    },
    {
      "entropy": 5.377616548538208,
      "epoch": 5.11094301563288,
      "grad_norm": 1.09375,
      "learning_rate": 0.000305528143864251,
      "loss": 4.7304,
      "mean_token_accuracy": 0.20711070150136948,
      "num_tokens": 11205889.0,
      "step": 5070
    },
    {
      "entropy": 5.332159757614136,
      "epoch": 5.115985879979829,
      "grad_norm": 1.3125,
      "learning_rate": 0.00030513509900445725,
      "loss": 4.671,
      "mean_token_accuracy": 0.21728027611970901,
      "num_tokens": 11217386.0,
      "step": 5075
    },
    {
      "entropy": 5.384092855453491,
      "epoch": 5.121028744326778,
      "grad_norm": 1.046875,
      "learning_rate": 0.00030474196048408065,
      "loss": 4.767,
      "mean_token_accuracy": 0.20545441061258315,
      "num_tokens": 11229963.0,
      "step": 5080
    },
    {
      "entropy": 5.341325330734253,
      "epoch": 5.126071608673727,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003043487295250049,
      "loss": 4.7201,
      "mean_token_accuracy": 0.20790380537509917,
      "num_tokens": 11240959.0,
      "step": 5085
    },
    {
      "entropy": 5.316199779510498,
      "epoch": 5.131114473020676,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003039554073494006,
      "loss": 4.6374,
      "mean_token_accuracy": 0.21337012499570845,
      "num_tokens": 11251971.0,
      "step": 5090
    },
    {
      "entropy": 5.35986852645874,
      "epoch": 5.136157337367624,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003035619951797224,
      "loss": 4.7401,
      "mean_token_accuracy": 0.20273467153310776,
      "num_tokens": 11261933.0,
      "step": 5095
    },
    {
      "entropy": 5.419374179840088,
      "epoch": 5.1412002017145735,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003031684942387043,
      "loss": 4.8118,
      "mean_token_accuracy": 0.20186106860637665,
      "num_tokens": 11272579.0,
      "step": 5100
    },
    {
      "entropy": 5.480310869216919,
      "epoch": 5.146243066061523,
      "grad_norm": 1.3125,
      "learning_rate": 0.00030277490574935633,
      "loss": 4.7991,
      "mean_token_accuracy": 0.20652221739292145,
      "num_tokens": 11282675.0,
      "step": 5105
    },
    {
      "entropy": 5.408874130249023,
      "epoch": 5.151285930408472,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003023812309349605,
      "loss": 4.7621,
      "mean_token_accuracy": 0.20043923407793046,
      "num_tokens": 11294837.0,
      "step": 5110
    },
    {
      "entropy": 5.439652156829834,
      "epoch": 5.156328794755421,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003019874710190674,
      "loss": 4.8519,
      "mean_token_accuracy": 0.1969693899154663,
      "num_tokens": 11305723.0,
      "step": 5115
    },
    {
      "entropy": 5.473764705657959,
      "epoch": 5.16137165910237,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003015936272254918,
      "loss": 4.813,
      "mean_token_accuracy": 0.20029695183038712,
      "num_tokens": 11316034.0,
      "step": 5120
    },
    {
      "entropy": 5.340233659744262,
      "epoch": 5.166414523449319,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0003011997007783094,
      "loss": 4.7093,
      "mean_token_accuracy": 0.20725106447935104,
      "num_tokens": 11325515.0,
      "step": 5125
    },
    {
      "entropy": 5.420378255844116,
      "epoch": 5.171457387796268,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003008056929018525,
      "loss": 4.7602,
      "mean_token_accuracy": 0.20451808869838714,
      "num_tokens": 11336874.0,
      "step": 5130
    },
    {
      "entropy": 5.36290602684021,
      "epoch": 5.176500252143217,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00030041160482070686,
      "loss": 4.7991,
      "mean_token_accuracy": 0.19886385798454284,
      "num_tokens": 11347749.0,
      "step": 5135
    },
    {
      "entropy": 5.339568281173706,
      "epoch": 5.181543116490166,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00030001743775970714,
      "loss": 4.7579,
      "mean_token_accuracy": 0.19968337416648865,
      "num_tokens": 11358865.0,
      "step": 5140
    },
    {
      "entropy": 5.3331952571868895,
      "epoch": 5.1865859808371155,
      "grad_norm": 1.5625,
      "learning_rate": 0.00029962319294393383,
      "loss": 4.7554,
      "mean_token_accuracy": 0.20657864958047867,
      "num_tokens": 11369699.0,
      "step": 5145
    },
    {
      "entropy": 5.37179741859436,
      "epoch": 5.191628845184065,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002992288715987088,
      "loss": 4.7374,
      "mean_token_accuracy": 0.20746450871229172,
      "num_tokens": 11381532.0,
      "step": 5150
    },
    {
      "entropy": 5.459340953826905,
      "epoch": 5.196671709531014,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002988344749495919,
      "loss": 4.7782,
      "mean_token_accuracy": 0.20839528888463973,
      "num_tokens": 11392373.0,
      "step": 5155
    },
    {
      "entropy": 5.412649297714234,
      "epoch": 5.201714573877963,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00029844000422237714,
      "loss": 4.7858,
      "mean_token_accuracy": 0.2075645238161087,
      "num_tokens": 11403732.0,
      "step": 5160
    },
    {
      "entropy": 5.349086618423462,
      "epoch": 5.206757438224912,
      "grad_norm": 1.265625,
      "learning_rate": 0.00029804546064308854,
      "loss": 4.7052,
      "mean_token_accuracy": 0.21083900928497315,
      "num_tokens": 11415344.0,
      "step": 5165
    },
    {
      "entropy": 5.347791910171509,
      "epoch": 5.211800302571861,
      "grad_norm": 1.1875,
      "learning_rate": 0.00029765084543797657,
      "loss": 4.7328,
      "mean_token_accuracy": 0.20074511915445328,
      "num_tokens": 11427564.0,
      "step": 5170
    },
    {
      "entropy": 5.448832750320435,
      "epoch": 5.21684316691881,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002972561598335145,
      "loss": 4.7725,
      "mean_token_accuracy": 0.2104944944381714,
      "num_tokens": 11439119.0,
      "step": 5175
    },
    {
      "entropy": 5.37475004196167,
      "epoch": 5.221886031265759,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002968614050563943,
      "loss": 4.7867,
      "mean_token_accuracy": 0.20467415004968642,
      "num_tokens": 11450460.0,
      "step": 5180
    },
    {
      "entropy": 5.386923503875733,
      "epoch": 5.2269288956127085,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002964665823335229,
      "loss": 4.8117,
      "mean_token_accuracy": 0.2003346025943756,
      "num_tokens": 11461414.0,
      "step": 5185
    },
    {
      "entropy": 5.398661994934082,
      "epoch": 5.231971759959657,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00029607169289201836,
      "loss": 4.7882,
      "mean_token_accuracy": 0.20694567859172822,
      "num_tokens": 11472071.0,
      "step": 5190
    },
    {
      "entropy": 5.390883541107177,
      "epoch": 5.237014624306606,
      "grad_norm": 1.28125,
      "learning_rate": 0.00029567673795920625,
      "loss": 4.7703,
      "mean_token_accuracy": 0.20460604429244994,
      "num_tokens": 11483098.0,
      "step": 5195
    },
    {
      "entropy": 5.392841434478759,
      "epoch": 5.242057488653555,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00029528171876261555,
      "loss": 4.7389,
      "mean_token_accuracy": 0.20840124487876893,
      "num_tokens": 11494339.0,
      "step": 5200
    },
    {
      "entropy": 5.43208875656128,
      "epoch": 5.247100353000504,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002948866365299751,
      "loss": 4.8096,
      "mean_token_accuracy": 0.2030286654829979,
      "num_tokens": 11504539.0,
      "step": 5205
    },
    {
      "entropy": 5.4345057010650635,
      "epoch": 5.252143217347453,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002944914924892096,
      "loss": 4.8223,
      "mean_token_accuracy": 0.20265262871980666,
      "num_tokens": 11515298.0,
      "step": 5210
    },
    {
      "entropy": 5.339143705368042,
      "epoch": 5.257186081694402,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0002940962878684358,
      "loss": 4.6593,
      "mean_token_accuracy": 0.21598684191703796,
      "num_tokens": 11524766.0,
      "step": 5215
    },
    {
      "entropy": 5.357718992233276,
      "epoch": 5.262228946041351,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002937010238959588,
      "loss": 4.7992,
      "mean_token_accuracy": 0.19846050292253495,
      "num_tokens": 11536603.0,
      "step": 5220
    },
    {
      "entropy": 5.4047308444976805,
      "epoch": 5.2672718103883005,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002933057018002681,
      "loss": 4.8294,
      "mean_token_accuracy": 0.20696419924497605,
      "num_tokens": 11548535.0,
      "step": 5225
    },
    {
      "entropy": 5.3881267547607425,
      "epoch": 5.27231467473525,
      "grad_norm": 1.28125,
      "learning_rate": 0.000292910322810034,
      "loss": 4.7231,
      "mean_token_accuracy": 0.20546931624412537,
      "num_tokens": 11559044.0,
      "step": 5230
    },
    {
      "entropy": 5.322073459625244,
      "epoch": 5.277357539082199,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00029251488815410344,
      "loss": 4.7573,
      "mean_token_accuracy": 0.20776389092206954,
      "num_tokens": 11571276.0,
      "step": 5235
    },
    {
      "entropy": 5.3177143096923825,
      "epoch": 5.282400403429148,
      "grad_norm": 1.28125,
      "learning_rate": 0.00029211939906149644,
      "loss": 4.7398,
      "mean_token_accuracy": 0.20512422174215317,
      "num_tokens": 11581935.0,
      "step": 5240
    },
    {
      "entropy": 5.389452838897705,
      "epoch": 5.287443267776097,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00029172385676140225,
      "loss": 4.8627,
      "mean_token_accuracy": 0.19998458921909332,
      "num_tokens": 11594188.0,
      "step": 5245
    },
    {
      "entropy": 5.44022912979126,
      "epoch": 5.292486132123046,
      "grad_norm": 1.328125,
      "learning_rate": 0.00029132826248317537,
      "loss": 4.8428,
      "mean_token_accuracy": 0.1989250361919403,
      "num_tokens": 11606025.0,
      "step": 5250
    },
    {
      "entropy": 5.403619003295899,
      "epoch": 5.297528996469995,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002909326174563322,
      "loss": 4.7863,
      "mean_token_accuracy": 0.2013598710298538,
      "num_tokens": 11617660.0,
      "step": 5255
    },
    {
      "entropy": 5.39440884590149,
      "epoch": 5.302571860816944,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00029053692291054636,
      "loss": 4.7609,
      "mean_token_accuracy": 0.20394814908504486,
      "num_tokens": 11628137.0,
      "step": 5260
    },
    {
      "entropy": 5.391919231414795,
      "epoch": 5.307614725163893,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00029014118007564564,
      "loss": 4.8103,
      "mean_token_accuracy": 0.19782521426677704,
      "num_tokens": 11639123.0,
      "step": 5265
    },
    {
      "entropy": 5.398278713226318,
      "epoch": 5.3126575895108425,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002897453901816081,
      "loss": 4.7644,
      "mean_token_accuracy": 0.21019506454467773,
      "num_tokens": 11650998.0,
      "step": 5270
    },
    {
      "entropy": 5.318338537216187,
      "epoch": 5.317700453857791,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00028934955445855773,
      "loss": 4.7463,
      "mean_token_accuracy": 0.20723402947187425,
      "num_tokens": 11660543.0,
      "step": 5275
    },
    {
      "entropy": 5.412715768814087,
      "epoch": 5.32274331820474,
      "grad_norm": 1.4375,
      "learning_rate": 0.00028895367413676095,
      "loss": 4.7503,
      "mean_token_accuracy": 0.20325214862823487,
      "num_tokens": 11671258.0,
      "step": 5280
    },
    {
      "entropy": 5.383661890029908,
      "epoch": 5.327786182551689,
      "grad_norm": 1.25,
      "learning_rate": 0.0002885577504466232,
      "loss": 4.7833,
      "mean_token_accuracy": 0.20545610934495925,
      "num_tokens": 11682498.0,
      "step": 5285
    },
    {
      "entropy": 5.4002796649932865,
      "epoch": 5.332829046898638,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00028816178461868425,
      "loss": 4.7978,
      "mean_token_accuracy": 0.20466247498989104,
      "num_tokens": 11694818.0,
      "step": 5290
    },
    {
      "entropy": 5.3366741180419925,
      "epoch": 5.337871911245587,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002877657778836151,
      "loss": 4.7894,
      "mean_token_accuracy": 0.1985916391015053,
      "num_tokens": 11705939.0,
      "step": 5295
    },
    {
      "entropy": 5.3842918395996096,
      "epoch": 5.342914775592536,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002873697314722138,
      "loss": 4.7956,
      "mean_token_accuracy": 0.20088584423065187,
      "num_tokens": 11716885.0,
      "step": 5300
    },
    {
      "entropy": 5.393453693389892,
      "epoch": 5.3479576399394855,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002869736466154018,
      "loss": 4.805,
      "mean_token_accuracy": 0.202781942486763,
      "num_tokens": 11728423.0,
      "step": 5305
    },
    {
      "entropy": 5.368320417404175,
      "epoch": 5.353000504286435,
      "grad_norm": 1.25,
      "learning_rate": 0.00028657752454421974,
      "loss": 4.7846,
      "mean_token_accuracy": 0.20440822392702102,
      "num_tokens": 11739293.0,
      "step": 5310
    },
    {
      "entropy": 5.429319047927857,
      "epoch": 5.358043368633384,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002861813664898245,
      "loss": 4.7757,
      "mean_token_accuracy": 0.20720991343259812,
      "num_tokens": 11749915.0,
      "step": 5315
    },
    {
      "entropy": 5.3167894840240475,
      "epoch": 5.363086232980333,
      "grad_norm": 1.34375,
      "learning_rate": 0.00028578517368348425,
      "loss": 4.8035,
      "mean_token_accuracy": 0.20666784793138504,
      "num_tokens": 11760055.0,
      "step": 5320
    },
    {
      "entropy": 5.328794956207275,
      "epoch": 5.368129097327282,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002853889473565753,
      "loss": 4.7749,
      "mean_token_accuracy": 0.2049415737390518,
      "num_tokens": 11771354.0,
      "step": 5325
    },
    {
      "entropy": 5.428283548355102,
      "epoch": 5.373171961674231,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002849926887405784,
      "loss": 4.8284,
      "mean_token_accuracy": 0.20337546467781067,
      "num_tokens": 11782628.0,
      "step": 5330
    },
    {
      "entropy": 5.342129516601562,
      "epoch": 5.37821482602118,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002845963990670744,
      "loss": 4.7402,
      "mean_token_accuracy": 0.204193115234375,
      "num_tokens": 11794733.0,
      "step": 5335
    },
    {
      "entropy": 5.46168270111084,
      "epoch": 5.383257690368129,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00028420007956774077,
      "loss": 4.8577,
      "mean_token_accuracy": 0.20001318603754042,
      "num_tokens": 11806532.0,
      "step": 5340
    },
    {
      "entropy": 5.363496541976929,
      "epoch": 5.388300554715078,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00028380373147434746,
      "loss": 4.8019,
      "mean_token_accuracy": 0.20032467395067216,
      "num_tokens": 11817305.0,
      "step": 5345
    },
    {
      "entropy": 5.3471982955932615,
      "epoch": 5.3933434190620275,
      "grad_norm": 1.25,
      "learning_rate": 0.00028340735601875376,
      "loss": 4.7019,
      "mean_token_accuracy": 0.2043310970067978,
      "num_tokens": 11827400.0,
      "step": 5350
    },
    {
      "entropy": 5.377983331680298,
      "epoch": 5.398386283408977,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002830109544329036,
      "loss": 4.7798,
      "mean_token_accuracy": 0.2055007964372635,
      "num_tokens": 11837675.0,
      "step": 5355
    },
    {
      "entropy": 5.247516250610351,
      "epoch": 5.403429147755926,
      "grad_norm": 1.4375,
      "learning_rate": 0.00028261452794882223,
      "loss": 4.6543,
      "mean_token_accuracy": 0.21412559747695922,
      "num_tokens": 11846567.0,
      "step": 5360
    },
    {
      "entropy": 5.433667755126953,
      "epoch": 5.408472012102875,
      "grad_norm": 1.296875,
      "learning_rate": 0.00028221807779861243,
      "loss": 4.835,
      "mean_token_accuracy": 0.2043476954102516,
      "num_tokens": 11856945.0,
      "step": 5365
    },
    {
      "entropy": 5.390807867050171,
      "epoch": 5.413514876449823,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002818216052144502,
      "loss": 4.7368,
      "mean_token_accuracy": 0.21132877767086028,
      "num_tokens": 11867302.0,
      "step": 5370
    },
    {
      "entropy": 5.367164421081543,
      "epoch": 5.418557740796772,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002814251114285816,
      "loss": 4.8306,
      "mean_token_accuracy": 0.2005585566163063,
      "num_tokens": 11879351.0,
      "step": 5375
    },
    {
      "entropy": 5.325662899017334,
      "epoch": 5.423600605143721,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002810285976733186,
      "loss": 4.7565,
      "mean_token_accuracy": 0.2065641164779663,
      "num_tokens": 11890337.0,
      "step": 5380
    },
    {
      "entropy": 5.402327060699463,
      "epoch": 5.4286434694906704,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002806320651810349,
      "loss": 4.751,
      "mean_token_accuracy": 0.2049296349287033,
      "num_tokens": 11900931.0,
      "step": 5385
    },
    {
      "entropy": 5.415978050231933,
      "epoch": 5.43368633383762,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002802355151841627,
      "loss": 4.807,
      "mean_token_accuracy": 0.20366040170192717,
      "num_tokens": 11911901.0,
      "step": 5390
    },
    {
      "entropy": 5.308761978149414,
      "epoch": 5.438729198184569,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002798389489151886,
      "loss": 4.7834,
      "mean_token_accuracy": 0.20754357129335405,
      "num_tokens": 11923349.0,
      "step": 5395
    },
    {
      "entropy": 5.314846229553223,
      "epoch": 5.443772062531518,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00027944236760664947,
      "loss": 4.7195,
      "mean_token_accuracy": 0.20861128568649293,
      "num_tokens": 11933496.0,
      "step": 5400
    },
    {
      "entropy": 5.388448333740234,
      "epoch": 5.448814926878467,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00027904577249112945,
      "loss": 4.8134,
      "mean_token_accuracy": 0.2080002471804619,
      "num_tokens": 11944626.0,
      "step": 5405
    },
    {
      "entropy": 5.423238897323609,
      "epoch": 5.453857791225416,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002786491648012552,
      "loss": 4.7632,
      "mean_token_accuracy": 0.20634396225214005,
      "num_tokens": 11956588.0,
      "step": 5410
    },
    {
      "entropy": 5.376740455627441,
      "epoch": 5.458900655572365,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0002782525457696926,
      "loss": 4.7909,
      "mean_token_accuracy": 0.20196923911571502,
      "num_tokens": 11967375.0,
      "step": 5415
    },
    {
      "entropy": 5.338318109512329,
      "epoch": 5.463943519919314,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002778559166291427,
      "loss": 4.7333,
      "mean_token_accuracy": 0.20497773289680482,
      "num_tokens": 11980651.0,
      "step": 5420
    },
    {
      "entropy": 5.345851087570191,
      "epoch": 5.468986384266263,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002774592786123379,
      "loss": 4.7371,
      "mean_token_accuracy": 0.2060674637556076,
      "num_tokens": 11991727.0,
      "step": 5425
    },
    {
      "entropy": 5.345818519592285,
      "epoch": 5.4740292486132125,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002770626329520386,
      "loss": 4.7685,
      "mean_token_accuracy": 0.20654116272926332,
      "num_tokens": 12001820.0,
      "step": 5430
    },
    {
      "entropy": 5.363340997695923,
      "epoch": 5.479072112960162,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002766659808810285,
      "loss": 4.749,
      "mean_token_accuracy": 0.20397913008928298,
      "num_tokens": 12012196.0,
      "step": 5435
    },
    {
      "entropy": 5.363338232040405,
      "epoch": 5.484114977307111,
      "grad_norm": 1.4375,
      "learning_rate": 0.00027626932363211134,
      "loss": 4.788,
      "mean_token_accuracy": 0.19482010900974273,
      "num_tokens": 12022701.0,
      "step": 5440
    },
    {
      "entropy": 5.346739339828491,
      "epoch": 5.48915784165406,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002758726624381071,
      "loss": 4.7594,
      "mean_token_accuracy": 0.21072799414396287,
      "num_tokens": 12032773.0,
      "step": 5445
    },
    {
      "entropy": 5.377016592025757,
      "epoch": 5.494200706001009,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002754759985318478,
      "loss": 4.8204,
      "mean_token_accuracy": 0.20592703074216842,
      "num_tokens": 12044140.0,
      "step": 5450
    },
    {
      "entropy": 5.410713768005371,
      "epoch": 5.499243570347957,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002750793331461741,
      "loss": 4.8205,
      "mean_token_accuracy": 0.19602502584457399,
      "num_tokens": 12055680.0,
      "step": 5455
    },
    {
      "entropy": 5.357589912414551,
      "epoch": 5.504286434694906,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002746826675139313,
      "loss": 4.78,
      "mean_token_accuracy": 0.20776030719280242,
      "num_tokens": 12068107.0,
      "step": 5460
    },
    {
      "entropy": 5.302773427963257,
      "epoch": 5.509329299041855,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00027428600286796503,
      "loss": 4.7353,
      "mean_token_accuracy": 0.20724163949489594,
      "num_tokens": 12079242.0,
      "step": 5465
    },
    {
      "entropy": 5.367038583755493,
      "epoch": 5.5143721633888045,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00027388934044111836,
      "loss": 4.8037,
      "mean_token_accuracy": 0.19698511213064193,
      "num_tokens": 12090720.0,
      "step": 5470
    },
    {
      "entropy": 5.347069215774536,
      "epoch": 5.519415027735754,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002734926814662271,
      "loss": 4.7155,
      "mean_token_accuracy": 0.2126438394188881,
      "num_tokens": 12101168.0,
      "step": 5475
    },
    {
      "entropy": 5.358303546905518,
      "epoch": 5.524457892082703,
      "grad_norm": 1.296875,
      "learning_rate": 0.00027309602717611673,
      "loss": 4.8148,
      "mean_token_accuracy": 0.19905234426259993,
      "num_tokens": 12112233.0,
      "step": 5480
    },
    {
      "entropy": 5.377463483810425,
      "epoch": 5.529500756429652,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00027269937880359786,
      "loss": 4.7977,
      "mean_token_accuracy": 0.20378277748823165,
      "num_tokens": 12122243.0,
      "step": 5485
    },
    {
      "entropy": 5.396564722061157,
      "epoch": 5.534543620776601,
      "grad_norm": 1.40625,
      "learning_rate": 0.00027230273758146263,
      "loss": 4.8166,
      "mean_token_accuracy": 0.1997758775949478,
      "num_tokens": 12132164.0,
      "step": 5490
    },
    {
      "entropy": 5.379419279098511,
      "epoch": 5.53958648512355,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002719061047424813,
      "loss": 4.696,
      "mean_token_accuracy": 0.21133473366498948,
      "num_tokens": 12141490.0,
      "step": 5495
    },
    {
      "entropy": 5.3451117992401125,
      "epoch": 5.544629349470499,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002715094815193978,
      "loss": 4.7629,
      "mean_token_accuracy": 0.19313400834798813,
      "num_tokens": 12151772.0,
      "step": 5500
    },
    {
      "epoch": 5.544629349470499,
      "eval_entropy": 5.179791746552057,
      "eval_loss": 5.666914463043213,
      "eval_mean_token_accuracy": 0.15457712009587057,
      "eval_num_tokens": 12151772.0,
      "eval_runtime": 2.2768,
      "eval_samples_per_second": 1378.26,
      "eval_steps_per_second": 172.612,
      "step": 5500
    },
    {
      "entropy": 5.308521366119384,
      "epoch": 5.549672213817448,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002711128691449262,
      "loss": 4.8011,
      "mean_token_accuracy": 0.20122693181037904,
      "num_tokens": 12163398.0,
      "step": 5505
    },
    {
      "entropy": 5.358075380325317,
      "epoch": 5.5547150781643975,
      "grad_norm": 1.28125,
      "learning_rate": 0.000270716268851747,
      "loss": 4.8369,
      "mean_token_accuracy": 0.19863139241933822,
      "num_tokens": 12173809.0,
      "step": 5510
    },
    {
      "entropy": 5.397147607803345,
      "epoch": 5.559757942511347,
      "grad_norm": 1.359375,
      "learning_rate": 0.00027031968187250305,
      "loss": 4.8418,
      "mean_token_accuracy": 0.19100669920444488,
      "num_tokens": 12185366.0,
      "step": 5515
    },
    {
      "entropy": 5.434380006790161,
      "epoch": 5.564800806858296,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002699231094397957,
      "loss": 4.7466,
      "mean_token_accuracy": 0.20518478751182556,
      "num_tokens": 12197613.0,
      "step": 5520
    },
    {
      "entropy": 5.3695777416229244,
      "epoch": 5.569843671205245,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002695265527861815,
      "loss": 4.7828,
      "mean_token_accuracy": 0.20141466408967973,
      "num_tokens": 12209152.0,
      "step": 5525
    },
    {
      "entropy": 5.330936002731323,
      "epoch": 5.574886535552194,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002691300131441674,
      "loss": 4.6804,
      "mean_token_accuracy": 0.20952833145856858,
      "num_tokens": 12219879.0,
      "step": 5530
    },
    {
      "entropy": 5.3092670917510985,
      "epoch": 5.579929399899143,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002687334917462079,
      "loss": 4.8176,
      "mean_token_accuracy": 0.19511986374855042,
      "num_tokens": 12231254.0,
      "step": 5535
    },
    {
      "entropy": 5.301174306869507,
      "epoch": 5.584972264246092,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002683369898247005,
      "loss": 4.6881,
      "mean_token_accuracy": 0.2124844580888748,
      "num_tokens": 12241545.0,
      "step": 5540
    },
    {
      "entropy": 5.304879474639892,
      "epoch": 5.590015128593041,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002679405086119825,
      "loss": 4.6931,
      "mean_token_accuracy": 0.21313273906707764,
      "num_tokens": 12252642.0,
      "step": 5545
    },
    {
      "entropy": 5.374314069747925,
      "epoch": 5.59505799293999,
      "grad_norm": 1.34375,
      "learning_rate": 0.00026754404934032657,
      "loss": 4.8264,
      "mean_token_accuracy": 0.2044908583164215,
      "num_tokens": 12264296.0,
      "step": 5550
    },
    {
      "entropy": 5.348439359664917,
      "epoch": 5.600100857286939,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002671476132419374,
      "loss": 4.7611,
      "mean_token_accuracy": 0.2004963144659996,
      "num_tokens": 12274127.0,
      "step": 5555
    },
    {
      "entropy": 5.371423721313477,
      "epoch": 5.605143721633888,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002667512015489472,
      "loss": 4.7992,
      "mean_token_accuracy": 0.20296185910701753,
      "num_tokens": 12287333.0,
      "step": 5560
    },
    {
      "entropy": 5.387567710876465,
      "epoch": 5.610186585980837,
      "grad_norm": 1.5078125,
      "learning_rate": 0.000266354815493413,
      "loss": 4.7263,
      "mean_token_accuracy": 0.21070191711187364,
      "num_tokens": 12297172.0,
      "step": 5565
    },
    {
      "entropy": 5.367141914367676,
      "epoch": 5.615229450327786,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00026595845630731166,
      "loss": 4.7883,
      "mean_token_accuracy": 0.20043358951807022,
      "num_tokens": 12308505.0,
      "step": 5570
    },
    {
      "entropy": 5.308610343933106,
      "epoch": 5.620272314674735,
      "grad_norm": 1.53125,
      "learning_rate": 0.00026556212522253666,
      "loss": 4.7349,
      "mean_token_accuracy": 0.20207765251398085,
      "num_tokens": 12319108.0,
      "step": 5575
    },
    {
      "entropy": 5.374286699295044,
      "epoch": 5.625315179021684,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002651658234708941,
      "loss": 4.7753,
      "mean_token_accuracy": 0.20696246474981309,
      "num_tokens": 12331935.0,
      "step": 5580
    },
    {
      "entropy": 5.346762466430664,
      "epoch": 5.630358043368633,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00026476955228409924,
      "loss": 4.7531,
      "mean_token_accuracy": 0.20637274235486985,
      "num_tokens": 12343380.0,
      "step": 5585
    },
    {
      "entropy": 5.357296562194824,
      "epoch": 5.635400907715582,
      "grad_norm": 1.34375,
      "learning_rate": 0.00026437331289377175,
      "loss": 4.7795,
      "mean_token_accuracy": 0.20298879593610764,
      "num_tokens": 12355216.0,
      "step": 5590
    },
    {
      "entropy": 5.436923885345459,
      "epoch": 5.640443772062532,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002639771065314329,
      "loss": 4.8707,
      "mean_token_accuracy": 0.20405201315879823,
      "num_tokens": 12367339.0,
      "step": 5595
    },
    {
      "entropy": 5.339682674407959,
      "epoch": 5.645486636409481,
      "grad_norm": 1.28125,
      "learning_rate": 0.00026358093442850134,
      "loss": 4.7805,
      "mean_token_accuracy": 0.20779405236244203,
      "num_tokens": 12378427.0,
      "step": 5600
    },
    {
      "entropy": 5.35480432510376,
      "epoch": 5.65052950075643,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002631847978162889,
      "loss": 4.7468,
      "mean_token_accuracy": 0.20558398365974426,
      "num_tokens": 12389264.0,
      "step": 5605
    },
    {
      "entropy": 5.330683422088623,
      "epoch": 5.655572365103379,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00026278869792599745,
      "loss": 4.8026,
      "mean_token_accuracy": 0.2091563433408737,
      "num_tokens": 12400080.0,
      "step": 5610
    },
    {
      "entropy": 5.370388555526733,
      "epoch": 5.660615229450328,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002623926359887145,
      "loss": 4.7818,
      "mean_token_accuracy": 0.20089555382728577,
      "num_tokens": 12410996.0,
      "step": 5615
    },
    {
      "entropy": 5.366828775405883,
      "epoch": 5.665658093797277,
      "grad_norm": 1.59375,
      "learning_rate": 0.00026199661323540973,
      "loss": 4.7926,
      "mean_token_accuracy": 0.20719375014305114,
      "num_tokens": 12421960.0,
      "step": 5620
    },
    {
      "entropy": 5.32580189704895,
      "epoch": 5.670700958144226,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00026160063089693096,
      "loss": 4.8086,
      "mean_token_accuracy": 0.20332509577274321,
      "num_tokens": 12433254.0,
      "step": 5625
    },
    {
      "entropy": 5.318446922302246,
      "epoch": 5.675743822491175,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002612046902040006,
      "loss": 4.6737,
      "mean_token_accuracy": 0.2056303083896637,
      "num_tokens": 12443278.0,
      "step": 5630
    },
    {
      "entropy": 5.384081745147705,
      "epoch": 5.680786686838124,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002608087923872111,
      "loss": 4.8052,
      "mean_token_accuracy": 0.20075951665639877,
      "num_tokens": 12456228.0,
      "step": 5635
    },
    {
      "entropy": 5.36103458404541,
      "epoch": 5.685829551185073,
      "grad_norm": 1.390625,
      "learning_rate": 0.00026041293867702235,
      "loss": 4.7578,
      "mean_token_accuracy": 0.20635222047567367,
      "num_tokens": 12467041.0,
      "step": 5640
    },
    {
      "entropy": 5.329710578918457,
      "epoch": 5.690872415532022,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00026001713030375664,
      "loss": 4.7485,
      "mean_token_accuracy": 0.20287285894155502,
      "num_tokens": 12477774.0,
      "step": 5645
    },
    {
      "entropy": 5.3519425868988035,
      "epoch": 5.695915279878971,
      "grad_norm": 1.21875,
      "learning_rate": 0.00025962136849759564,
      "loss": 4.7945,
      "mean_token_accuracy": 0.19733876287937163,
      "num_tokens": 12489621.0,
      "step": 5650
    },
    {
      "entropy": 5.38506851196289,
      "epoch": 5.70095814422592,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00025922565448857605,
      "loss": 4.8222,
      "mean_token_accuracy": 0.19799471646547318,
      "num_tokens": 12501885.0,
      "step": 5655
    },
    {
      "entropy": 5.439307594299317,
      "epoch": 5.706001008572869,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00025882998950658623,
      "loss": 4.8674,
      "mean_token_accuracy": 0.19519540071487426,
      "num_tokens": 12514066.0,
      "step": 5660
    },
    {
      "entropy": 5.297085952758789,
      "epoch": 5.711043872919818,
      "grad_norm": 1.25,
      "learning_rate": 0.00025843437478136196,
      "loss": 4.7512,
      "mean_token_accuracy": 0.20653219372034073,
      "num_tokens": 12524603.0,
      "step": 5665
    },
    {
      "entropy": 5.36145601272583,
      "epoch": 5.716086737266767,
      "grad_norm": 1.375,
      "learning_rate": 0.00025803881154248305,
      "loss": 4.7355,
      "mean_token_accuracy": 0.20963634848594664,
      "num_tokens": 12536339.0,
      "step": 5670
    },
    {
      "entropy": 5.418450498580933,
      "epoch": 5.7211296016137165,
      "grad_norm": 1.25,
      "learning_rate": 0.0002576433010193691,
      "loss": 4.7677,
      "mean_token_accuracy": 0.20334111899137497,
      "num_tokens": 12547346.0,
      "step": 5675
    },
    {
      "entropy": 5.2899822235107425,
      "epoch": 5.726172465960666,
      "grad_norm": 1.21875,
      "learning_rate": 0.000257247844441276,
      "loss": 4.7541,
      "mean_token_accuracy": 0.20494245737791061,
      "num_tokens": 12558064.0,
      "step": 5680
    },
    {
      "entropy": 5.267758131027222,
      "epoch": 5.731215330307615,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00025685244303729187,
      "loss": 4.7408,
      "mean_token_accuracy": 0.20933765769004822,
      "num_tokens": 12568908.0,
      "step": 5685
    },
    {
      "entropy": 5.340923833847046,
      "epoch": 5.736258194654564,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002564570980363334,
      "loss": 4.7815,
      "mean_token_accuracy": 0.20253028273582457,
      "num_tokens": 12579731.0,
      "step": 5690
    },
    {
      "entropy": 5.3510462760925295,
      "epoch": 5.741301059001513,
      "grad_norm": 1.390625,
      "learning_rate": 0.00025606181066714195,
      "loss": 4.7597,
      "mean_token_accuracy": 0.20619093328714372,
      "num_tokens": 12589438.0,
      "step": 5695
    },
    {
      "entropy": 5.3693594455719,
      "epoch": 5.746343923348462,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002556665821582798,
      "loss": 4.8165,
      "mean_token_accuracy": 0.19874174892902374,
      "num_tokens": 12600204.0,
      "step": 5700
    },
    {
      "entropy": 5.327193546295166,
      "epoch": 5.751386787695411,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002552714137381264,
      "loss": 4.7435,
      "mean_token_accuracy": 0.20999565124511718,
      "num_tokens": 12609441.0,
      "step": 5705
    },
    {
      "entropy": 5.344610548019409,
      "epoch": 5.75642965204236,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002548763066348741,
      "loss": 4.6972,
      "mean_token_accuracy": 0.20835570991039276,
      "num_tokens": 12620286.0,
      "step": 5710
    },
    {
      "entropy": 5.363843059539795,
      "epoch": 5.7614725163893095,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002544812620765251,
      "loss": 4.7303,
      "mean_token_accuracy": 0.2078951209783554,
      "num_tokens": 12631377.0,
      "step": 5715
    },
    {
      "entropy": 5.374969387054444,
      "epoch": 5.766515380736259,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002540862812908868,
      "loss": 4.8229,
      "mean_token_accuracy": 0.19874543100595474,
      "num_tokens": 12643550.0,
      "step": 5720
    },
    {
      "entropy": 5.316448211669922,
      "epoch": 5.771558245083208,
      "grad_norm": 1.25,
      "learning_rate": 0.0002536913655055687,
      "loss": 4.8067,
      "mean_token_accuracy": 0.20123717188835144,
      "num_tokens": 12654269.0,
      "step": 5725
    },
    {
      "entropy": 5.33986463546753,
      "epoch": 5.776601109430157,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002532965159479781,
      "loss": 4.7869,
      "mean_token_accuracy": 0.20129682570695878,
      "num_tokens": 12665146.0,
      "step": 5730
    },
    {
      "entropy": 5.362718820571899,
      "epoch": 5.781643973777105,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002529017338453167,
      "loss": 4.7957,
      "mean_token_accuracy": 0.20741688162088395,
      "num_tokens": 12675711.0,
      "step": 5735
    },
    {
      "entropy": 5.336122465133667,
      "epoch": 5.786686838124054,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002525070204245761,
      "loss": 4.6686,
      "mean_token_accuracy": 0.20992153733968735,
      "num_tokens": 12686065.0,
      "step": 5740
    },
    {
      "entropy": 5.336485433578491,
      "epoch": 5.791729702471003,
      "grad_norm": 1.171875,
      "learning_rate": 0.00025211237691253507,
      "loss": 4.731,
      "mean_token_accuracy": 0.2105412542819977,
      "num_tokens": 12696581.0,
      "step": 5745
    },
    {
      "entropy": 5.381826639175415,
      "epoch": 5.796772566817952,
      "grad_norm": 1.359375,
      "learning_rate": 0.00025171780453575447,
      "loss": 4.805,
      "mean_token_accuracy": 0.20175062417984008,
      "num_tokens": 12707696.0,
      "step": 5750
    },
    {
      "entropy": 5.339414787292481,
      "epoch": 5.8018154311649015,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002513233045205746,
      "loss": 4.7426,
      "mean_token_accuracy": 0.2086472913622856,
      "num_tokens": 12717992.0,
      "step": 5755
    },
    {
      "entropy": 5.34476261138916,
      "epoch": 5.806858295511851,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002509288780931102,
      "loss": 4.8038,
      "mean_token_accuracy": 0.2031818747520447,
      "num_tokens": 12729309.0,
      "step": 5760
    },
    {
      "entropy": 5.3823200225830075,
      "epoch": 5.8119011598588,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002505345264792482,
      "loss": 4.7625,
      "mean_token_accuracy": 0.21101685166358947,
      "num_tokens": 12738493.0,
      "step": 5765
    },
    {
      "entropy": 5.322897148132324,
      "epoch": 5.816944024205749,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00025014025090464205,
      "loss": 4.7676,
      "mean_token_accuracy": 0.20678919404745102,
      "num_tokens": 12748797.0,
      "step": 5770
    },
    {
      "entropy": 5.379063272476197,
      "epoch": 5.821986888552698,
      "grad_norm": 1.203125,
      "learning_rate": 0.00024974605259470975,
      "loss": 4.7489,
      "mean_token_accuracy": 0.20400178283452988,
      "num_tokens": 12760040.0,
      "step": 5775
    },
    {
      "entropy": 5.369988870620728,
      "epoch": 5.827029752899647,
      "grad_norm": 1.296875,
      "learning_rate": 0.00024935193277462835,
      "loss": 4.7715,
      "mean_token_accuracy": 0.2024065986275673,
      "num_tokens": 12770285.0,
      "step": 5780
    },
    {
      "entropy": 5.376403331756592,
      "epoch": 5.832072617246596,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002489578926693317,
      "loss": 4.7799,
      "mean_token_accuracy": 0.20566319823265075,
      "num_tokens": 12783012.0,
      "step": 5785
    },
    {
      "entropy": 5.403157567977905,
      "epoch": 5.837115481593545,
      "grad_norm": 1.25,
      "learning_rate": 0.0002485639335035053,
      "loss": 4.8372,
      "mean_token_accuracy": 0.20084059834480286,
      "num_tokens": 12794850.0,
      "step": 5790
    },
    {
      "entropy": 5.348498249053955,
      "epoch": 5.842158345940494,
      "grad_norm": 1.296875,
      "learning_rate": 0.00024817005650158345,
      "loss": 4.7696,
      "mean_token_accuracy": 0.20610379725694655,
      "num_tokens": 12805087.0,
      "step": 5795
    },
    {
      "entropy": 5.4035858631134035,
      "epoch": 5.8472012102874436,
      "grad_norm": 1.109375,
      "learning_rate": 0.00024777626288774473,
      "loss": 4.8669,
      "mean_token_accuracy": 0.19749905616045,
      "num_tokens": 12816144.0,
      "step": 5800
    },
    {
      "entropy": 5.3838475227355955,
      "epoch": 5.852244074634393,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002473825538859091,
      "loss": 4.7296,
      "mean_token_accuracy": 0.20245456844568252,
      "num_tokens": 12826942.0,
      "step": 5805
    },
    {
      "entropy": 5.3803564548492435,
      "epoch": 5.857286938981342,
      "grad_norm": 1.375,
      "learning_rate": 0.0002469889307197331,
      "loss": 4.7573,
      "mean_token_accuracy": 0.20649609416723252,
      "num_tokens": 12836777.0,
      "step": 5810
    },
    {
      "entropy": 5.371955490112304,
      "epoch": 5.86232980332829,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002465953946126065,
      "loss": 4.7949,
      "mean_token_accuracy": 0.2006831109523773,
      "num_tokens": 12848286.0,
      "step": 5815
    },
    {
      "entropy": 5.3131402969360355,
      "epoch": 5.867372667675239,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00024620194678764875,
      "loss": 4.7131,
      "mean_token_accuracy": 0.2055204078555107,
      "num_tokens": 12859383.0,
      "step": 5820
    },
    {
      "entropy": 5.299517059326172,
      "epoch": 5.872415532022188,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002458085884677046,
      "loss": 4.773,
      "mean_token_accuracy": 0.20928642451763152,
      "num_tokens": 12869897.0,
      "step": 5825
    },
    {
      "entropy": 5.363385105133057,
      "epoch": 5.877458396369137,
      "grad_norm": 1.609375,
      "learning_rate": 0.00024541532087534097,
      "loss": 4.8211,
      "mean_token_accuracy": 0.20196860581636428,
      "num_tokens": 12880052.0,
      "step": 5830
    },
    {
      "entropy": 5.429199647903443,
      "epoch": 5.8825012607160865,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002450221452328424,
      "loss": 4.8633,
      "mean_token_accuracy": 0.19865271747112273,
      "num_tokens": 12892322.0,
      "step": 5835
    },
    {
      "entropy": 5.401175260543823,
      "epoch": 5.887544125063036,
      "grad_norm": 1.296875,
      "learning_rate": 0.00024462906276220814,
      "loss": 4.7642,
      "mean_token_accuracy": 0.19862645864486694,
      "num_tokens": 12903399.0,
      "step": 5840
    },
    {
      "entropy": 5.322895479202271,
      "epoch": 5.892586989409985,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002442360746851473,
      "loss": 4.7343,
      "mean_token_accuracy": 0.20683542042970657,
      "num_tokens": 12913108.0,
      "step": 5845
    },
    {
      "entropy": 5.358834362030029,
      "epoch": 5.897629853756934,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00024384318222307605,
      "loss": 4.8091,
      "mean_token_accuracy": 0.20511840432882308,
      "num_tokens": 12924810.0,
      "step": 5850
    },
    {
      "entropy": 5.39119291305542,
      "epoch": 5.902672718103883,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002434503865971131,
      "loss": 4.8292,
      "mean_token_accuracy": 0.19755433201789857,
      "num_tokens": 12934896.0,
      "step": 5855
    },
    {
      "entropy": 5.3179301738739015,
      "epoch": 5.907715582450832,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002430576890280765,
      "loss": 4.7573,
      "mean_token_accuracy": 0.20801438689231871,
      "num_tokens": 12945873.0,
      "step": 5860
    },
    {
      "entropy": 5.367904186248779,
      "epoch": 5.912758446797781,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00024266509073647913,
      "loss": 4.8674,
      "mean_token_accuracy": 0.2009687677025795,
      "num_tokens": 12958701.0,
      "step": 5865
    },
    {
      "entropy": 5.3404923439025875,
      "epoch": 5.91780131114473,
      "grad_norm": 1.328125,
      "learning_rate": 0.00024227259294252578,
      "loss": 4.8332,
      "mean_token_accuracy": 0.199054816365242,
      "num_tokens": 12970729.0,
      "step": 5870
    },
    {
      "entropy": 5.379343032836914,
      "epoch": 5.922844175491679,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002418801968661084,
      "loss": 4.7776,
      "mean_token_accuracy": 0.20337541550397872,
      "num_tokens": 12980917.0,
      "step": 5875
    },
    {
      "entropy": 5.352252388000489,
      "epoch": 5.9278870398386285,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00024148790372680324,
      "loss": 4.8207,
      "mean_token_accuracy": 0.20628398805856704,
      "num_tokens": 12993123.0,
      "step": 5880
    },
    {
      "entropy": 5.257782077789306,
      "epoch": 5.932929904185578,
      "grad_norm": 1.328125,
      "learning_rate": 0.00024109571474386631,
      "loss": 4.6664,
      "mean_token_accuracy": 0.2172641158103943,
      "num_tokens": 13004031.0,
      "step": 5885
    },
    {
      "entropy": 5.434179401397705,
      "epoch": 5.937972768532527,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00024070363113623018,
      "loss": 4.8639,
      "mean_token_accuracy": 0.1999412104487419,
      "num_tokens": 13015407.0,
      "step": 5890
    },
    {
      "entropy": 5.4561511993408205,
      "epoch": 5.943015632879476,
      "grad_norm": 1.1875,
      "learning_rate": 0.00024031165412249938,
      "loss": 4.798,
      "mean_token_accuracy": 0.21083608269691467,
      "num_tokens": 13027293.0,
      "step": 5895
    },
    {
      "entropy": 5.398774671554565,
      "epoch": 5.948058497226425,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000239919784920948,
      "loss": 4.7794,
      "mean_token_accuracy": 0.20626269280910492,
      "num_tokens": 13037033.0,
      "step": 5900
    },
    {
      "entropy": 5.363014650344849,
      "epoch": 5.953101361573374,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00023952802474951418,
      "loss": 4.7988,
      "mean_token_accuracy": 0.1985170215368271,
      "num_tokens": 13049896.0,
      "step": 5905
    },
    {
      "entropy": 5.3304040908813475,
      "epoch": 5.958144225920323,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00023913637482579774,
      "loss": 4.7745,
      "mean_token_accuracy": 0.20336031764745713,
      "num_tokens": 13059745.0,
      "step": 5910
    },
    {
      "entropy": 5.3348002433776855,
      "epoch": 5.9631870902672715,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00023874483636705563,
      "loss": 4.7845,
      "mean_token_accuracy": 0.20156327039003372,
      "num_tokens": 13070993.0,
      "step": 5915
    },
    {
      "entropy": 5.430982398986816,
      "epoch": 5.968229954614221,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00023835341059019838,
      "loss": 4.9385,
      "mean_token_accuracy": 0.1884035587310791,
      "num_tokens": 13081743.0,
      "step": 5920
    },
    {
      "entropy": 5.414664697647095,
      "epoch": 5.97327281896117,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00023796209871178642,
      "loss": 4.8055,
      "mean_token_accuracy": 0.20064786523580552,
      "num_tokens": 13092894.0,
      "step": 5925
    },
    {
      "entropy": 5.323222064971924,
      "epoch": 5.978315683308119,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00023757090194802605,
      "loss": 4.7347,
      "mean_token_accuracy": 0.21286237835884095,
      "num_tokens": 13103886.0,
      "step": 5930
    },
    {
      "entropy": 5.366406583786011,
      "epoch": 5.983358547655068,
      "grad_norm": 1.296875,
      "learning_rate": 0.00023717982151476592,
      "loss": 4.8298,
      "mean_token_accuracy": 0.19448499381542206,
      "num_tokens": 13114858.0,
      "step": 5935
    },
    {
      "entropy": 5.417066764831543,
      "epoch": 5.988401412002017,
      "grad_norm": 1.3125,
      "learning_rate": 0.00023678885862749282,
      "loss": 4.8419,
      "mean_token_accuracy": 0.1975385770201683,
      "num_tokens": 13126359.0,
      "step": 5940
    },
    {
      "entropy": 5.315640783309936,
      "epoch": 5.993444276348966,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00023639801450132863,
      "loss": 4.699,
      "mean_token_accuracy": 0.20454249382019044,
      "num_tokens": 13136623.0,
      "step": 5945
    },
    {
      "entropy": 5.381450843811035,
      "epoch": 5.998487140695915,
      "grad_norm": 1.28125,
      "learning_rate": 0.00023600729035102585,
      "loss": 4.8112,
      "mean_token_accuracy": 0.2062505915760994,
      "num_tokens": 13147518.0,
      "step": 5950
    },
    {
      "entropy": 5.3663147820366754,
      "epoch": 6.0030257186081695,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002356166873909642,
      "loss": 4.6455,
      "mean_token_accuracy": 0.21081026726298863,
      "num_tokens": 13155773.0,
      "step": 5955
    },
    {
      "entropy": 5.374180841445923,
      "epoch": 6.008068582955119,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00023522620683514645,
      "loss": 4.6913,
      "mean_token_accuracy": 0.2164499804377556,
      "num_tokens": 13167682.0,
      "step": 5960
    },
    {
      "entropy": 5.318479347229004,
      "epoch": 6.013111447302068,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00023483584989719553,
      "loss": 4.5781,
      "mean_token_accuracy": 0.22809401154518127,
      "num_tokens": 13179069.0,
      "step": 5965
    },
    {
      "entropy": 5.348093795776367,
      "epoch": 6.018154311649017,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00023444561779034957,
      "loss": 4.6072,
      "mean_token_accuracy": 0.21669170409440994,
      "num_tokens": 13190548.0,
      "step": 5970
    },
    {
      "entropy": 5.296596050262451,
      "epoch": 6.023197175995966,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00023405551172745904,
      "loss": 4.6121,
      "mean_token_accuracy": 0.221450312435627,
      "num_tokens": 13202006.0,
      "step": 5975
    },
    {
      "entropy": 5.401267719268799,
      "epoch": 6.028240040342915,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00023366553292098247,
      "loss": 4.6254,
      "mean_token_accuracy": 0.22494206428527833,
      "num_tokens": 13212236.0,
      "step": 5980
    },
    {
      "entropy": 5.342278242111206,
      "epoch": 6.033282904689864,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002332756825829832,
      "loss": 4.5882,
      "mean_token_accuracy": 0.22302798330783843,
      "num_tokens": 13224704.0,
      "step": 5985
    },
    {
      "entropy": 5.342916822433471,
      "epoch": 6.038325769036813,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002328859619251249,
      "loss": 4.5748,
      "mean_token_accuracy": 0.22424897253513337,
      "num_tokens": 13236020.0,
      "step": 5990
    },
    {
      "entropy": 5.318588781356811,
      "epoch": 6.043368633383762,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00023249637215866847,
      "loss": 4.6069,
      "mean_token_accuracy": 0.21727624386548997,
      "num_tokens": 13247466.0,
      "step": 5995
    },
    {
      "entropy": 5.319881296157837,
      "epoch": 6.048411497730711,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002321069144944678,
      "loss": 4.6218,
      "mean_token_accuracy": 0.22042322307825088,
      "num_tokens": 13258927.0,
      "step": 6000
    },
    {
      "epoch": 6.048411497730711,
      "eval_entropy": 5.207631120244965,
      "eval_loss": 5.660191059112549,
      "eval_mean_token_accuracy": 0.15537597625765182,
      "eval_num_tokens": 13258927.0,
      "eval_runtime": 2.0532,
      "eval_samples_per_second": 1528.371,
      "eval_steps_per_second": 191.412,
      "step": 6000
    }
  ],
  "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": 3286226940825600.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}