{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 6.1100356597045336,
  "eval_steps": 500,
  "global_step": 6000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 10.691933536529541,
      "epoch": 0.005094243504839531,
      "grad_norm": 12.9375,
      "learning_rate": 2e-06,
      "loss": 10.8617,
      "mean_token_accuracy": 7.283321465365589e-05,
      "num_tokens": 12441.0,
      "step": 5
    },
    {
      "entropy": 10.69195785522461,
      "epoch": 0.010188487009679063,
      "grad_norm": 13.0625,
      "learning_rate": 4.5e-06,
      "loss": 10.7886,
      "mean_token_accuracy": 8.764241938479245e-05,
      "num_tokens": 23925.0,
      "step": 10
    },
    {
      "entropy": 10.691264724731445,
      "epoch": 0.015282730514518594,
      "grad_norm": 9.875,
      "learning_rate": 7e-06,
      "loss": 10.5167,
      "mean_token_accuracy": 0.023061814554966985,
      "num_tokens": 33958.0,
      "step": 15
    },
    {
      "entropy": 10.679726791381835,
      "epoch": 0.020376974019358125,
      "grad_norm": 5.84375,
      "learning_rate": 9.5e-06,
      "loss": 10.1946,
      "mean_token_accuracy": 0.043799113109707834,
      "num_tokens": 46188.0,
      "step": 20
    },
    {
      "entropy": 10.613013648986817,
      "epoch": 0.02547121752419766,
      "grad_norm": 3.9375,
      "learning_rate": 1.2e-05,
      "loss": 9.8889,
      "mean_token_accuracy": 0.04678206667304039,
      "num_tokens": 58080.0,
      "step": 25
    },
    {
      "entropy": 10.596247959136964,
      "epoch": 0.030565461029037188,
      "grad_norm": 3.171875,
      "learning_rate": 1.4500000000000002e-05,
      "loss": 9.7705,
      "mean_token_accuracy": 0.03978040032088757,
      "num_tokens": 72446.0,
      "step": 30
    },
    {
      "entropy": 10.60612449645996,
      "epoch": 0.03565970453387672,
      "grad_norm": 2.890625,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 9.6958,
      "mean_token_accuracy": 0.043326153233647344,
      "num_tokens": 85522.0,
      "step": 35
    },
    {
      "entropy": 10.586102485656738,
      "epoch": 0.04075394803871625,
      "grad_norm": 2.859375,
      "learning_rate": 1.95e-05,
      "loss": 9.6414,
      "mean_token_accuracy": 0.039182791113853456,
      "num_tokens": 96970.0,
      "step": 40
    },
    {
      "entropy": 10.593886470794677,
      "epoch": 0.04584819154355578,
      "grad_norm": 2.484375,
      "learning_rate": 2.2e-05,
      "loss": 9.6151,
      "mean_token_accuracy": 0.04260034691542387,
      "num_tokens": 108011.0,
      "step": 45
    },
    {
      "entropy": 10.594364643096924,
      "epoch": 0.05094243504839532,
      "grad_norm": 2.515625,
      "learning_rate": 2.4500000000000003e-05,
      "loss": 9.528,
      "mean_token_accuracy": 0.04013910107314587,
      "num_tokens": 119673.0,
      "step": 50
    },
    {
      "entropy": 10.586376857757568,
      "epoch": 0.056036678553234846,
      "grad_norm": 2.15625,
      "learning_rate": 2.7e-05,
      "loss": 9.5493,
      "mean_token_accuracy": 0.0417574979364872,
      "num_tokens": 133041.0,
      "step": 55
    },
    {
      "entropy": 10.560707283020019,
      "epoch": 0.061130922058074376,
      "grad_norm": 2.203125,
      "learning_rate": 2.95e-05,
      "loss": 9.4204,
      "mean_token_accuracy": 0.042354696244001386,
      "num_tokens": 143467.0,
      "step": 60
    },
    {
      "entropy": 10.495057964324952,
      "epoch": 0.06622516556291391,
      "grad_norm": 2.34375,
      "learning_rate": 3.2e-05,
      "loss": 9.3327,
      "mean_token_accuracy": 0.05373127907514572,
      "num_tokens": 154745.0,
      "step": 65
    },
    {
      "entropy": 10.405490398406982,
      "epoch": 0.07131940906775344,
      "grad_norm": 2.3125,
      "learning_rate": 3.4500000000000005e-05,
      "loss": 9.2218,
      "mean_token_accuracy": 0.053240276873111725,
      "num_tokens": 165526.0,
      "step": 70
    },
    {
      "entropy": 10.435882568359375,
      "epoch": 0.07641365257259297,
      "grad_norm": 2.453125,
      "learning_rate": 3.7e-05,
      "loss": 9.1812,
      "mean_token_accuracy": 0.054624059423804286,
      "num_tokens": 177735.0,
      "step": 75
    },
    {
      "entropy": 10.438795280456542,
      "epoch": 0.0815078960774325,
      "grad_norm": 2.375,
      "learning_rate": 3.95e-05,
      "loss": 9.0299,
      "mean_token_accuracy": 0.06236044131219387,
      "num_tokens": 189877.0,
      "step": 80
    },
    {
      "entropy": 10.311788082122803,
      "epoch": 0.08660213958227203,
      "grad_norm": 2.3125,
      "learning_rate": 4.2000000000000004e-05,
      "loss": 9.0334,
      "mean_token_accuracy": 0.058551866188645364,
      "num_tokens": 201740.0,
      "step": 85
    },
    {
      "entropy": 10.396924591064453,
      "epoch": 0.09169638308711156,
      "grad_norm": 2.21875,
      "learning_rate": 4.45e-05,
      "loss": 8.9084,
      "mean_token_accuracy": 0.05712463036179542,
      "num_tokens": 213385.0,
      "step": 90
    },
    {
      "entropy": 10.31311674118042,
      "epoch": 0.09679062659195109,
      "grad_norm": 1.9453125,
      "learning_rate": 4.7000000000000004e-05,
      "loss": 8.8263,
      "mean_token_accuracy": 0.06144661605358124,
      "num_tokens": 225501.0,
      "step": 95
    },
    {
      "entropy": 10.3136625289917,
      "epoch": 0.10188487009679063,
      "grad_norm": 1.9296875,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 8.7332,
      "mean_token_accuracy": 0.06830099895596504,
      "num_tokens": 239512.0,
      "step": 100
    },
    {
      "entropy": 10.148342895507813,
      "epoch": 0.10697911360163016,
      "grad_norm": 2.140625,
      "learning_rate": 5.2e-05,
      "loss": 8.6085,
      "mean_token_accuracy": 0.06609614193439484,
      "num_tokens": 252261.0,
      "step": 105
    },
    {
      "entropy": 10.107837390899657,
      "epoch": 0.11207335710646969,
      "grad_norm": 2.015625,
      "learning_rate": 5.45e-05,
      "loss": 8.4972,
      "mean_token_accuracy": 0.06363288164138795,
      "num_tokens": 263754.0,
      "step": 110
    },
    {
      "entropy": 9.950089359283448,
      "epoch": 0.11716760061130922,
      "grad_norm": 2.296875,
      "learning_rate": 5.7e-05,
      "loss": 8.377,
      "mean_token_accuracy": 0.07086736857891082,
      "num_tokens": 274953.0,
      "step": 115
    },
    {
      "entropy": 9.918815803527831,
      "epoch": 0.12226184411614875,
      "grad_norm": 2.421875,
      "learning_rate": 5.9499999999999996e-05,
      "loss": 8.2772,
      "mean_token_accuracy": 0.07252867110073566,
      "num_tokens": 284844.0,
      "step": 120
    },
    {
      "entropy": 9.788673591613769,
      "epoch": 0.1273560876209883,
      "grad_norm": 1.7890625,
      "learning_rate": 6.2e-05,
      "loss": 8.1819,
      "mean_token_accuracy": 0.07298735603690147,
      "num_tokens": 296833.0,
      "step": 125
    },
    {
      "entropy": 9.649177837371827,
      "epoch": 0.13245033112582782,
      "grad_norm": 1.8828125,
      "learning_rate": 6.450000000000001e-05,
      "loss": 8.0676,
      "mean_token_accuracy": 0.07574639841914177,
      "num_tokens": 308528.0,
      "step": 130
    },
    {
      "entropy": 9.483296966552734,
      "epoch": 0.13754457463066735,
      "grad_norm": 1.5703125,
      "learning_rate": 6.7e-05,
      "loss": 7.9082,
      "mean_token_accuracy": 0.07914942651987075,
      "num_tokens": 320599.0,
      "step": 135
    },
    {
      "entropy": 9.217451667785644,
      "epoch": 0.14263881813550688,
      "grad_norm": 1.609375,
      "learning_rate": 6.950000000000001e-05,
      "loss": 7.8002,
      "mean_token_accuracy": 0.07902043983340264,
      "num_tokens": 332264.0,
      "step": 140
    },
    {
      "entropy": 8.995352077484132,
      "epoch": 0.1477330616403464,
      "grad_norm": 1.1953125,
      "learning_rate": 7.2e-05,
      "loss": 7.7372,
      "mean_token_accuracy": 0.08456770777702331,
      "num_tokens": 344233.0,
      "step": 145
    },
    {
      "entropy": 8.8959490776062,
      "epoch": 0.15282730514518594,
      "grad_norm": 1.1875,
      "learning_rate": 7.45e-05,
      "loss": 7.7246,
      "mean_token_accuracy": 0.08076966851949692,
      "num_tokens": 356200.0,
      "step": 150
    },
    {
      "entropy": 8.64002866744995,
      "epoch": 0.15792154865002547,
      "grad_norm": 1.3828125,
      "learning_rate": 7.7e-05,
      "loss": 7.7279,
      "mean_token_accuracy": 0.07943359091877937,
      "num_tokens": 369791.0,
      "step": 155
    },
    {
      "entropy": 8.485534954071046,
      "epoch": 0.163015792154865,
      "grad_norm": 1.3125,
      "learning_rate": 7.950000000000001e-05,
      "loss": 7.506,
      "mean_token_accuracy": 0.08976328000426292,
      "num_tokens": 381127.0,
      "step": 160
    },
    {
      "entropy": 8.262951564788818,
      "epoch": 0.16811003565970453,
      "grad_norm": 1.4453125,
      "learning_rate": 8.2e-05,
      "loss": 7.6364,
      "mean_token_accuracy": 0.0788124717772007,
      "num_tokens": 393941.0,
      "step": 165
    },
    {
      "entropy": 8.217638492584229,
      "epoch": 0.17320427916454406,
      "grad_norm": 1.3359375,
      "learning_rate": 8.450000000000001e-05,
      "loss": 7.4653,
      "mean_token_accuracy": 0.08866559341549873,
      "num_tokens": 404883.0,
      "step": 170
    },
    {
      "entropy": 8.12268590927124,
      "epoch": 0.1782985226693836,
      "grad_norm": 1.171875,
      "learning_rate": 8.7e-05,
      "loss": 7.5067,
      "mean_token_accuracy": 0.0853340208530426,
      "num_tokens": 417961.0,
      "step": 175
    },
    {
      "entropy": 8.083796834945678,
      "epoch": 0.18339276617422312,
      "grad_norm": 1.25,
      "learning_rate": 8.95e-05,
      "loss": 7.4617,
      "mean_token_accuracy": 0.08598975539207458,
      "num_tokens": 430076.0,
      "step": 180
    },
    {
      "entropy": 7.989891767501831,
      "epoch": 0.18848700967906265,
      "grad_norm": 1.1640625,
      "learning_rate": 9.2e-05,
      "loss": 7.4382,
      "mean_token_accuracy": 0.08668889105319977,
      "num_tokens": 442091.0,
      "step": 185
    },
    {
      "entropy": 7.941371583938599,
      "epoch": 0.19358125318390218,
      "grad_norm": 1.203125,
      "learning_rate": 9.45e-05,
      "loss": 7.4342,
      "mean_token_accuracy": 0.08515545055270195,
      "num_tokens": 454272.0,
      "step": 190
    },
    {
      "entropy": 7.948634958267212,
      "epoch": 0.1986754966887417,
      "grad_norm": 1.4296875,
      "learning_rate": 9.7e-05,
      "loss": 7.368,
      "mean_token_accuracy": 0.07814025804400444,
      "num_tokens": 465237.0,
      "step": 195
    },
    {
      "entropy": 7.879694080352783,
      "epoch": 0.20376974019358127,
      "grad_norm": 1.2578125,
      "learning_rate": 9.95e-05,
      "loss": 7.3887,
      "mean_token_accuracy": 0.08662542253732682,
      "num_tokens": 476842.0,
      "step": 200
    },
    {
      "entropy": 7.8720118522644045,
      "epoch": 0.2088639836984208,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000102,
      "loss": 7.4507,
      "mean_token_accuracy": 0.0870475135743618,
      "num_tokens": 489198.0,
      "step": 205
    },
    {
      "entropy": 7.757378482818604,
      "epoch": 0.21395822720326033,
      "grad_norm": 1.0625,
      "learning_rate": 0.00010449999999999999,
      "loss": 7.3925,
      "mean_token_accuracy": 0.089558956772089,
      "num_tokens": 500320.0,
      "step": 210
    },
    {
      "entropy": 7.914836311340332,
      "epoch": 0.21905247070809986,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000107,
      "loss": 7.2584,
      "mean_token_accuracy": 0.0958659790456295,
      "num_tokens": 512117.0,
      "step": 215
    },
    {
      "entropy": 7.7668720245361325,
      "epoch": 0.22414671421293939,
      "grad_norm": 1.25,
      "learning_rate": 0.0001095,
      "loss": 7.39,
      "mean_token_accuracy": 0.08571948632597923,
      "num_tokens": 524106.0,
      "step": 220
    },
    {
      "entropy": 7.755286931991577,
      "epoch": 0.22924095771777891,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000112,
      "loss": 7.3119,
      "mean_token_accuracy": 0.08375829830765724,
      "num_tokens": 535815.0,
      "step": 225
    },
    {
      "entropy": 7.795973873138427,
      "epoch": 0.23433520122261844,
      "grad_norm": 1.03125,
      "learning_rate": 0.0001145,
      "loss": 7.3265,
      "mean_token_accuracy": 0.08689820542931556,
      "num_tokens": 548217.0,
      "step": 230
    },
    {
      "entropy": 7.734084177017212,
      "epoch": 0.23942944472745797,
      "grad_norm": 1.046875,
      "learning_rate": 0.00011700000000000001,
      "loss": 7.3159,
      "mean_token_accuracy": 0.09363964423537255,
      "num_tokens": 559218.0,
      "step": 235
    },
    {
      "entropy": 7.692209482192993,
      "epoch": 0.2445236882322975,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00011949999999999999,
      "loss": 7.2849,
      "mean_token_accuracy": 0.08573080748319625,
      "num_tokens": 571021.0,
      "step": 240
    },
    {
      "entropy": 7.653146457672119,
      "epoch": 0.24961793173713703,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000122,
      "loss": 7.3651,
      "mean_token_accuracy": 0.08879496678709983,
      "num_tokens": 582023.0,
      "step": 245
    },
    {
      "entropy": 7.691405630111694,
      "epoch": 0.2547121752419766,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0001245,
      "loss": 7.288,
      "mean_token_accuracy": 0.09284965097904205,
      "num_tokens": 594938.0,
      "step": 250
    },
    {
      "entropy": 7.677452039718628,
      "epoch": 0.2598064187468161,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000127,
      "loss": 7.2142,
      "mean_token_accuracy": 0.09495285749435425,
      "num_tokens": 606978.0,
      "step": 255
    },
    {
      "entropy": 7.555742835998535,
      "epoch": 0.26490066225165565,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001295,
      "loss": 7.224,
      "mean_token_accuracy": 0.09364139512181283,
      "num_tokens": 617653.0,
      "step": 260
    },
    {
      "entropy": 7.647906541824341,
      "epoch": 0.26999490575649515,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000132,
      "loss": 7.209,
      "mean_token_accuracy": 0.08683466166257858,
      "num_tokens": 628807.0,
      "step": 265
    },
    {
      "entropy": 7.582596492767334,
      "epoch": 0.2750891492613347,
      "grad_norm": 1.25,
      "learning_rate": 0.00013450000000000002,
      "loss": 7.2779,
      "mean_token_accuracy": 0.088419159501791,
      "num_tokens": 639512.0,
      "step": 270
    },
    {
      "entropy": 7.654620885848999,
      "epoch": 0.2801833927661742,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00013700000000000002,
      "loss": 7.2792,
      "mean_token_accuracy": 0.08684752136468887,
      "num_tokens": 652491.0,
      "step": 275
    },
    {
      "entropy": 7.6056561946868895,
      "epoch": 0.28527763627101377,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0001395,
      "loss": 7.1743,
      "mean_token_accuracy": 0.09496726170182228,
      "num_tokens": 664152.0,
      "step": 280
    },
    {
      "entropy": 7.49957766532898,
      "epoch": 0.29037187977585327,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00014199999999999998,
      "loss": 7.2412,
      "mean_token_accuracy": 0.08711678311228752,
      "num_tokens": 676774.0,
      "step": 285
    },
    {
      "entropy": 7.589282417297364,
      "epoch": 0.2954661232806928,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0001445,
      "loss": 7.256,
      "mean_token_accuracy": 0.09429504796862602,
      "num_tokens": 689459.0,
      "step": 290
    },
    {
      "entropy": 7.601702690124512,
      "epoch": 0.30056036678553233,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000147,
      "loss": 7.1178,
      "mean_token_accuracy": 0.1071195811033249,
      "num_tokens": 700338.0,
      "step": 295
    },
    {
      "entropy": 7.5238135814666744,
      "epoch": 0.3056546102903719,
      "grad_norm": 1.234375,
      "learning_rate": 0.0001495,
      "loss": 7.1756,
      "mean_token_accuracy": 0.10010429248213767,
      "num_tokens": 711764.0,
      "step": 300
    },
    {
      "entropy": 7.4707728862762455,
      "epoch": 0.3107488537952114,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000152,
      "loss": 7.1665,
      "mean_token_accuracy": 0.09938320368528367,
      "num_tokens": 725232.0,
      "step": 305
    },
    {
      "entropy": 7.504101991653442,
      "epoch": 0.31584309730005095,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00015450000000000001,
      "loss": 7.126,
      "mean_token_accuracy": 0.09905346557497978,
      "num_tokens": 737036.0,
      "step": 310
    },
    {
      "entropy": 7.535156774520874,
      "epoch": 0.32093734080489045,
      "grad_norm": 1.125,
      "learning_rate": 0.000157,
      "loss": 7.1663,
      "mean_token_accuracy": 0.09537394866347312,
      "num_tokens": 747844.0,
      "step": 315
    },
    {
      "entropy": 7.532956981658936,
      "epoch": 0.32603158430973,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001595,
      "loss": 7.196,
      "mean_token_accuracy": 0.09172512367367744,
      "num_tokens": 760730.0,
      "step": 320
    },
    {
      "entropy": 7.569301605224609,
      "epoch": 0.33112582781456956,
      "grad_norm": 1.21875,
      "learning_rate": 0.000162,
      "loss": 7.1939,
      "mean_token_accuracy": 0.08795616179704666,
      "num_tokens": 773020.0,
      "step": 325
    },
    {
      "entropy": 7.3828143119812015,
      "epoch": 0.33622007131940906,
      "grad_norm": 1.0,
      "learning_rate": 0.00016450000000000001,
      "loss": 7.1709,
      "mean_token_accuracy": 0.09435797706246377,
      "num_tokens": 786337.0,
      "step": 330
    },
    {
      "entropy": 7.46736159324646,
      "epoch": 0.3413143148242486,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00016700000000000002,
      "loss": 7.1946,
      "mean_token_accuracy": 0.09482537731528282,
      "num_tokens": 798490.0,
      "step": 335
    },
    {
      "entropy": 7.484088563919068,
      "epoch": 0.3464085583290881,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00016950000000000003,
      "loss": 7.0577,
      "mean_token_accuracy": 0.09521348923444747,
      "num_tokens": 810448.0,
      "step": 340
    },
    {
      "entropy": 7.479926538467407,
      "epoch": 0.3515028018339277,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00017199999999999998,
      "loss": 7.1475,
      "mean_token_accuracy": 0.09660323038697242,
      "num_tokens": 821443.0,
      "step": 345
    },
    {
      "entropy": 7.534337854385376,
      "epoch": 0.3565970453387672,
      "grad_norm": 1.0625,
      "learning_rate": 0.00017449999999999999,
      "loss": 7.1518,
      "mean_token_accuracy": 0.09190995395183563,
      "num_tokens": 833935.0,
      "step": 350
    },
    {
      "entropy": 7.371308040618897,
      "epoch": 0.36169128884360674,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000177,
      "loss": 7.1277,
      "mean_token_accuracy": 0.09297349452972412,
      "num_tokens": 845343.0,
      "step": 355
    },
    {
      "entropy": 7.360507678985596,
      "epoch": 0.36678553234844624,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0001795,
      "loss": 7.0661,
      "mean_token_accuracy": 0.10084544345736504,
      "num_tokens": 856683.0,
      "step": 360
    },
    {
      "entropy": 7.551660776138306,
      "epoch": 0.3718797758532858,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000182,
      "loss": 7.1496,
      "mean_token_accuracy": 0.09621937721967697,
      "num_tokens": 869190.0,
      "step": 365
    },
    {
      "entropy": 7.396897411346435,
      "epoch": 0.3769740193581253,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0001845,
      "loss": 7.1107,
      "mean_token_accuracy": 0.09819533750414848,
      "num_tokens": 880811.0,
      "step": 370
    },
    {
      "entropy": 7.379519033432007,
      "epoch": 0.38206826286296486,
      "grad_norm": 1.15625,
      "learning_rate": 0.000187,
      "loss": 7.062,
      "mean_token_accuracy": 0.09250850528478623,
      "num_tokens": 892559.0,
      "step": 375
    },
    {
      "entropy": 7.351970911026001,
      "epoch": 0.38716250636780436,
      "grad_norm": 1.1875,
      "learning_rate": 0.0001895,
      "loss": 6.9836,
      "mean_token_accuracy": 0.09942182153463364,
      "num_tokens": 903602.0,
      "step": 380
    },
    {
      "entropy": 7.36213903427124,
      "epoch": 0.3922567498726439,
      "grad_norm": 1.140625,
      "learning_rate": 0.000192,
      "loss": 7.0794,
      "mean_token_accuracy": 0.09812569916248322,
      "num_tokens": 915021.0,
      "step": 385
    },
    {
      "entropy": 7.400866937637329,
      "epoch": 0.3973509933774834,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0001945,
      "loss": 7.1524,
      "mean_token_accuracy": 0.09627576470375061,
      "num_tokens": 927084.0,
      "step": 390
    },
    {
      "entropy": 7.322605895996094,
      "epoch": 0.402445236882323,
      "grad_norm": 1.203125,
      "learning_rate": 0.00019700000000000002,
      "loss": 7.0569,
      "mean_token_accuracy": 0.09742958173155784,
      "num_tokens": 939900.0,
      "step": 395
    },
    {
      "entropy": 7.3560632228851315,
      "epoch": 0.40753948038716253,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00019950000000000002,
      "loss": 7.1292,
      "mean_token_accuracy": 0.08907266408205032,
      "num_tokens": 953956.0,
      "step": 400
    },
    {
      "entropy": 7.4070006847381595,
      "epoch": 0.41263372389200204,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000202,
      "loss": 7.0604,
      "mean_token_accuracy": 0.1046214260160923,
      "num_tokens": 966526.0,
      "step": 405
    },
    {
      "entropy": 7.261282110214234,
      "epoch": 0.4177279673968416,
      "grad_norm": 1.25,
      "learning_rate": 0.00020449999999999998,
      "loss": 6.9502,
      "mean_token_accuracy": 0.09846476316452027,
      "num_tokens": 977889.0,
      "step": 410
    },
    {
      "entropy": 7.307093572616577,
      "epoch": 0.4228222109016811,
      "grad_norm": 1.015625,
      "learning_rate": 0.000207,
      "loss": 7.0391,
      "mean_token_accuracy": 0.09500454217195511,
      "num_tokens": 989737.0,
      "step": 415
    },
    {
      "entropy": 7.332889413833618,
      "epoch": 0.42791645440652065,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002095,
      "loss": 7.0727,
      "mean_token_accuracy": 0.09839839786291123,
      "num_tokens": 1000820.0,
      "step": 420
    },
    {
      "entropy": 7.271477127075196,
      "epoch": 0.43301069791136015,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000212,
      "loss": 6.9583,
      "mean_token_accuracy": 0.10580145046114922,
      "num_tokens": 1012156.0,
      "step": 425
    },
    {
      "entropy": 7.290036249160766,
      "epoch": 0.4381049414161997,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002145,
      "loss": 6.8937,
      "mean_token_accuracy": 0.10478519052267074,
      "num_tokens": 1022825.0,
      "step": 430
    },
    {
      "entropy": 7.220993757247925,
      "epoch": 0.4431991849210392,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00021700000000000002,
      "loss": 6.9762,
      "mean_token_accuracy": 0.10158220753073692,
      "num_tokens": 1034725.0,
      "step": 435
    },
    {
      "entropy": 7.262880039215088,
      "epoch": 0.44829342842587877,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002195,
      "loss": 6.9638,
      "mean_token_accuracy": 0.09944465085864067,
      "num_tokens": 1045927.0,
      "step": 440
    },
    {
      "entropy": 7.248775577545166,
      "epoch": 0.4533876719307183,
      "grad_norm": 1.21875,
      "learning_rate": 0.000222,
      "loss": 6.9828,
      "mean_token_accuracy": 0.09821130633354187,
      "num_tokens": 1057943.0,
      "step": 445
    },
    {
      "entropy": 7.3402995586395265,
      "epoch": 0.45848191543555783,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002245,
      "loss": 7.0581,
      "mean_token_accuracy": 0.09455323070287705,
      "num_tokens": 1068949.0,
      "step": 450
    },
    {
      "entropy": 7.2297731876373295,
      "epoch": 0.46357615894039733,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00022700000000000002,
      "loss": 6.932,
      "mean_token_accuracy": 0.1041158229112625,
      "num_tokens": 1081298.0,
      "step": 455
    },
    {
      "entropy": 7.228163337707519,
      "epoch": 0.4686704024452369,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00022950000000000002,
      "loss": 6.9816,
      "mean_token_accuracy": 0.10278591811656952,
      "num_tokens": 1092522.0,
      "step": 460
    },
    {
      "entropy": 7.093091297149658,
      "epoch": 0.4737646459500764,
      "grad_norm": 1.15625,
      "learning_rate": 0.00023200000000000003,
      "loss": 6.8957,
      "mean_token_accuracy": 0.1022353433072567,
      "num_tokens": 1104133.0,
      "step": 465
    },
    {
      "entropy": 7.173185110092163,
      "epoch": 0.47885888945491595,
      "grad_norm": 1.125,
      "learning_rate": 0.00023449999999999998,
      "loss": 6.8605,
      "mean_token_accuracy": 0.1085218146443367,
      "num_tokens": 1116057.0,
      "step": 470
    },
    {
      "entropy": 7.134055089950562,
      "epoch": 0.48395313295975545,
      "grad_norm": 1.0,
      "learning_rate": 0.000237,
      "loss": 6.9546,
      "mean_token_accuracy": 0.09869935512542724,
      "num_tokens": 1128148.0,
      "step": 475
    },
    {
      "entropy": 7.2346595287322994,
      "epoch": 0.489047376464595,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002395,
      "loss": 6.9666,
      "mean_token_accuracy": 0.09960033446550369,
      "num_tokens": 1140104.0,
      "step": 480
    },
    {
      "entropy": 7.173930263519287,
      "epoch": 0.49414161996943456,
      "grad_norm": 1.078125,
      "learning_rate": 0.000242,
      "loss": 6.8951,
      "mean_token_accuracy": 0.10067171081900597,
      "num_tokens": 1152711.0,
      "step": 485
    },
    {
      "entropy": 7.140958261489868,
      "epoch": 0.49923586347427407,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002445,
      "loss": 6.8955,
      "mean_token_accuracy": 0.10290815010666847,
      "num_tokens": 1164223.0,
      "step": 490
    },
    {
      "entropy": 7.230376243591309,
      "epoch": 0.5043301069791136,
      "grad_norm": 1.21875,
      "learning_rate": 0.000247,
      "loss": 6.8962,
      "mean_token_accuracy": 0.10572774037718773,
      "num_tokens": 1174990.0,
      "step": 495
    },
    {
      "entropy": 7.084908771514892,
      "epoch": 0.5094243504839532,
      "grad_norm": 1.125,
      "learning_rate": 0.0002495,
      "loss": 6.8948,
      "mean_token_accuracy": 0.10063612461090088,
      "num_tokens": 1187561.0,
      "step": 500
    },
    {
      "epoch": 0.5094243504839532,
      "eval_entropy": 7.262294474943184,
      "eval_loss": 6.917758464813232,
      "eval_mean_token_accuracy": 0.10530804161120344,
      "eval_num_tokens": 1187561.0,
      "eval_runtime": 2.7331,
      "eval_samples_per_second": 1185.468,
      "eval_steps_per_second": 148.183,
      "step": 500
    },
    {
      "entropy": 7.168519735336304,
      "epoch": 0.5145185939887926,
      "grad_norm": 1.171875,
      "learning_rate": 0.000252,
      "loss": 6.8901,
      "mean_token_accuracy": 0.10450750216841698,
      "num_tokens": 1199108.0,
      "step": 505
    },
    {
      "entropy": 7.162225580215454,
      "epoch": 0.5196128374936322,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002545,
      "loss": 6.9639,
      "mean_token_accuracy": 0.10146603956818581,
      "num_tokens": 1211944.0,
      "step": 510
    },
    {
      "entropy": 7.139475297927857,
      "epoch": 0.5247070809984717,
      "grad_norm": 1.125,
      "learning_rate": 0.000257,
      "loss": 6.8997,
      "mean_token_accuracy": 0.09549630656838418,
      "num_tokens": 1223044.0,
      "step": 515
    },
    {
      "entropy": 7.140495634078979,
      "epoch": 0.5298013245033113,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002595,
      "loss": 6.923,
      "mean_token_accuracy": 0.10165961533784866,
      "num_tokens": 1235161.0,
      "step": 520
    },
    {
      "entropy": 7.138057518005371,
      "epoch": 0.5348955680081507,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000262,
      "loss": 6.9507,
      "mean_token_accuracy": 0.10160766541957855,
      "num_tokens": 1247732.0,
      "step": 525
    },
    {
      "entropy": 7.0351362228393555,
      "epoch": 0.5399898115129903,
      "grad_norm": 1.125,
      "learning_rate": 0.00026450000000000003,
      "loss": 6.7937,
      "mean_token_accuracy": 0.10965558364987374,
      "num_tokens": 1259349.0,
      "step": 530
    },
    {
      "entropy": 7.072154712677002,
      "epoch": 0.5450840550178299,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00026700000000000004,
      "loss": 6.76,
      "mean_token_accuracy": 0.11131609529256821,
      "num_tokens": 1269992.0,
      "step": 535
    },
    {
      "entropy": 6.97647762298584,
      "epoch": 0.5501782985226694,
      "grad_norm": 1.125,
      "learning_rate": 0.00026950000000000005,
      "loss": 6.8285,
      "mean_token_accuracy": 0.11009727045893669,
      "num_tokens": 1281502.0,
      "step": 540
    },
    {
      "entropy": 7.11279182434082,
      "epoch": 0.5552725420275089,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00027200000000000005,
      "loss": 6.7735,
      "mean_token_accuracy": 0.10252350270748138,
      "num_tokens": 1293880.0,
      "step": 545
    },
    {
      "entropy": 7.021842575073242,
      "epoch": 0.5603667855323484,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002745,
      "loss": 6.8787,
      "mean_token_accuracy": 0.10112191140651702,
      "num_tokens": 1306022.0,
      "step": 550
    },
    {
      "entropy": 7.022417259216309,
      "epoch": 0.565461029037188,
      "grad_norm": 1.140625,
      "learning_rate": 0.000277,
      "loss": 6.8458,
      "mean_token_accuracy": 0.10293983742594719,
      "num_tokens": 1317939.0,
      "step": 555
    },
    {
      "entropy": 6.911152935028076,
      "epoch": 0.5705552725420275,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002795,
      "loss": 6.6947,
      "mean_token_accuracy": 0.11798072904348374,
      "num_tokens": 1329917.0,
      "step": 560
    },
    {
      "entropy": 6.976852607727051,
      "epoch": 0.5756495160468671,
      "grad_norm": 1.109375,
      "learning_rate": 0.00028199999999999997,
      "loss": 6.8079,
      "mean_token_accuracy": 0.11052986085414887,
      "num_tokens": 1341733.0,
      "step": 565
    },
    {
      "entropy": 7.096958923339844,
      "epoch": 0.5807437595517065,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002845,
      "loss": 6.746,
      "mean_token_accuracy": 0.1133323922753334,
      "num_tokens": 1352454.0,
      "step": 570
    },
    {
      "entropy": 7.072584772109986,
      "epoch": 0.5858380030565461,
      "grad_norm": 1.21875,
      "learning_rate": 0.000287,
      "loss": 6.9179,
      "mean_token_accuracy": 0.10664055198431015,
      "num_tokens": 1363878.0,
      "step": 575
    },
    {
      "entropy": 7.035218572616577,
      "epoch": 0.5909322465613857,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002895,
      "loss": 6.8355,
      "mean_token_accuracy": 0.09741255789995193,
      "num_tokens": 1375159.0,
      "step": 580
    },
    {
      "entropy": 7.033307790756226,
      "epoch": 0.5960264900662252,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000292,
      "loss": 6.8531,
      "mean_token_accuracy": 0.10392893180251121,
      "num_tokens": 1386646.0,
      "step": 585
    },
    {
      "entropy": 7.03128490447998,
      "epoch": 0.6011207335710647,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002945,
      "loss": 6.7635,
      "mean_token_accuracy": 0.11585776656866073,
      "num_tokens": 1398020.0,
      "step": 590
    },
    {
      "entropy": 6.957646226882934,
      "epoch": 0.6062149770759042,
      "grad_norm": 1.078125,
      "learning_rate": 0.000297,
      "loss": 6.7512,
      "mean_token_accuracy": 0.11048697158694268,
      "num_tokens": 1408209.0,
      "step": 595
    },
    {
      "entropy": 7.011190795898438,
      "epoch": 0.6113092205807438,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002995,
      "loss": 6.8055,
      "mean_token_accuracy": 0.1033931739628315,
      "num_tokens": 1420391.0,
      "step": 600
    },
    {
      "entropy": 6.984811115264892,
      "epoch": 0.6164034640855833,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000302,
      "loss": 6.7731,
      "mean_token_accuracy": 0.10607712045311928,
      "num_tokens": 1432250.0,
      "step": 605
    },
    {
      "entropy": 6.982828664779663,
      "epoch": 0.6214977075904228,
      "grad_norm": 1.25,
      "learning_rate": 0.0003045,
      "loss": 6.7448,
      "mean_token_accuracy": 0.10987704768776893,
      "num_tokens": 1444823.0,
      "step": 610
    },
    {
      "entropy": 6.871470260620117,
      "epoch": 0.6265919510952623,
      "grad_norm": 1.15625,
      "learning_rate": 0.000307,
      "loss": 6.8034,
      "mean_token_accuracy": 0.10774814710021019,
      "num_tokens": 1455997.0,
      "step": 615
    },
    {
      "entropy": 7.018206071853638,
      "epoch": 0.6316861946001019,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003095,
      "loss": 6.7605,
      "mean_token_accuracy": 0.11087881028652191,
      "num_tokens": 1467707.0,
      "step": 620
    },
    {
      "entropy": 7.00219144821167,
      "epoch": 0.6367804381049414,
      "grad_norm": 1.125,
      "learning_rate": 0.000312,
      "loss": 6.789,
      "mean_token_accuracy": 0.10458588898181916,
      "num_tokens": 1477920.0,
      "step": 625
    },
    {
      "entropy": 7.005989694595337,
      "epoch": 0.6418746816097809,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003145,
      "loss": 6.8028,
      "mean_token_accuracy": 0.10765366032719612,
      "num_tokens": 1488176.0,
      "step": 630
    },
    {
      "entropy": 6.924004459381104,
      "epoch": 0.6469689251146205,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000317,
      "loss": 6.7612,
      "mean_token_accuracy": 0.10841824859380722,
      "num_tokens": 1499018.0,
      "step": 635
    },
    {
      "entropy": 7.021401691436767,
      "epoch": 0.65206316861946,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003195,
      "loss": 6.8063,
      "mean_token_accuracy": 0.10359535291790962,
      "num_tokens": 1510759.0,
      "step": 640
    },
    {
      "entropy": 6.915233039855957,
      "epoch": 0.6571574121242996,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000322,
      "loss": 6.7829,
      "mean_token_accuracy": 0.10922421887516975,
      "num_tokens": 1523028.0,
      "step": 645
    },
    {
      "entropy": 6.967010641098023,
      "epoch": 0.6622516556291391,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00032450000000000003,
      "loss": 6.7709,
      "mean_token_accuracy": 0.11187901347875595,
      "num_tokens": 1533505.0,
      "step": 650
    },
    {
      "entropy": 6.995042371749878,
      "epoch": 0.6673458991339786,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00032700000000000003,
      "loss": 6.8077,
      "mean_token_accuracy": 0.11219450533390045,
      "num_tokens": 1545366.0,
      "step": 655
    },
    {
      "entropy": 6.975549745559692,
      "epoch": 0.6724401426388181,
      "grad_norm": 1.15625,
      "learning_rate": 0.00032950000000000004,
      "loss": 6.6689,
      "mean_token_accuracy": 0.11294186711311341,
      "num_tokens": 1557976.0,
      "step": 660
    },
    {
      "entropy": 6.857566213607788,
      "epoch": 0.6775343861436577,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00033200000000000005,
      "loss": 6.7871,
      "mean_token_accuracy": 0.10875996574759483,
      "num_tokens": 1569822.0,
      "step": 665
    },
    {
      "entropy": 6.903141641616822,
      "epoch": 0.6826286296484972,
      "grad_norm": 1.140625,
      "learning_rate": 0.00033450000000000005,
      "loss": 6.6939,
      "mean_token_accuracy": 0.11107023805379868,
      "num_tokens": 1580485.0,
      "step": 670
    },
    {
      "entropy": 6.974818754196167,
      "epoch": 0.6877228731533367,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000337,
      "loss": 6.739,
      "mean_token_accuracy": 0.106230578571558,
      "num_tokens": 1591865.0,
      "step": 675
    },
    {
      "entropy": 6.857933616638183,
      "epoch": 0.6928171166581762,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003395,
      "loss": 6.6193,
      "mean_token_accuracy": 0.11482649818062782,
      "num_tokens": 1604580.0,
      "step": 680
    },
    {
      "entropy": 6.869587802886963,
      "epoch": 0.6979113601630158,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000342,
      "loss": 6.7781,
      "mean_token_accuracy": 0.10850742682814599,
      "num_tokens": 1615772.0,
      "step": 685
    },
    {
      "entropy": 6.902220678329468,
      "epoch": 0.7030056036678554,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00034449999999999997,
      "loss": 6.667,
      "mean_token_accuracy": 0.11990016177296639,
      "num_tokens": 1626965.0,
      "step": 690
    },
    {
      "entropy": 6.808654689788819,
      "epoch": 0.7080998471726948,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000347,
      "loss": 6.7007,
      "mean_token_accuracy": 0.11295107081532478,
      "num_tokens": 1638865.0,
      "step": 695
    },
    {
      "entropy": 6.851603651046753,
      "epoch": 0.7131940906775344,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003495,
      "loss": 6.673,
      "mean_token_accuracy": 0.1173295795917511,
      "num_tokens": 1649200.0,
      "step": 700
    },
    {
      "entropy": 6.836373996734619,
      "epoch": 0.7182883341823739,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000352,
      "loss": 6.6496,
      "mean_token_accuracy": 0.11917404681444169,
      "num_tokens": 1660726.0,
      "step": 705
    },
    {
      "entropy": 7.016081094741821,
      "epoch": 0.7233825776872135,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003545,
      "loss": 6.6981,
      "mean_token_accuracy": 0.11023402512073517,
      "num_tokens": 1672101.0,
      "step": 710
    },
    {
      "entropy": 6.692198133468628,
      "epoch": 0.7284768211920529,
      "grad_norm": 1.203125,
      "learning_rate": 0.000357,
      "loss": 6.647,
      "mean_token_accuracy": 0.11462761536240577,
      "num_tokens": 1683560.0,
      "step": 715
    },
    {
      "entropy": 6.85313572883606,
      "epoch": 0.7335710646968925,
      "grad_norm": 1.125,
      "learning_rate": 0.0003595,
      "loss": 6.6876,
      "mean_token_accuracy": 0.11544215381145477,
      "num_tokens": 1694084.0,
      "step": 720
    },
    {
      "entropy": 6.773692846298218,
      "epoch": 0.738665308201732,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000362,
      "loss": 6.6777,
      "mean_token_accuracy": 0.11710209399461746,
      "num_tokens": 1706686.0,
      "step": 725
    },
    {
      "entropy": 6.84816575050354,
      "epoch": 0.7437595517065716,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003645,
      "loss": 6.7517,
      "mean_token_accuracy": 0.11291900128126145,
      "num_tokens": 1719661.0,
      "step": 730
    },
    {
      "entropy": 6.915406131744385,
      "epoch": 0.7488537952114112,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000367,
      "loss": 6.6261,
      "mean_token_accuracy": 0.1130788080394268,
      "num_tokens": 1730521.0,
      "step": 735
    },
    {
      "entropy": 6.753418588638306,
      "epoch": 0.7539480387162506,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003695,
      "loss": 6.7005,
      "mean_token_accuracy": 0.11622855588793754,
      "num_tokens": 1742251.0,
      "step": 740
    },
    {
      "entropy": 6.651693344116211,
      "epoch": 0.7590422822210902,
      "grad_norm": 1.15625,
      "learning_rate": 0.000372,
      "loss": 6.5389,
      "mean_token_accuracy": 0.11808115169405937,
      "num_tokens": 1752871.0,
      "step": 745
    },
    {
      "entropy": 6.831832647323608,
      "epoch": 0.7641365257259297,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003745,
      "loss": 6.7215,
      "mean_token_accuracy": 0.11175637394189834,
      "num_tokens": 1765690.0,
      "step": 750
    },
    {
      "entropy": 6.77952823638916,
      "epoch": 0.7692307692307693,
      "grad_norm": 0.984375,
      "learning_rate": 0.000377,
      "loss": 6.6015,
      "mean_token_accuracy": 0.1158766008913517,
      "num_tokens": 1778398.0,
      "step": 755
    },
    {
      "entropy": 6.785519695281982,
      "epoch": 0.7743250127356087,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0003795,
      "loss": 6.6874,
      "mean_token_accuracy": 0.10822736173868179,
      "num_tokens": 1791496.0,
      "step": 760
    },
    {
      "entropy": 6.903592872619629,
      "epoch": 0.7794192562404483,
      "grad_norm": 1.0625,
      "learning_rate": 0.000382,
      "loss": 6.683,
      "mean_token_accuracy": 0.10840215981006622,
      "num_tokens": 1803152.0,
      "step": 765
    },
    {
      "entropy": 6.712223196029663,
      "epoch": 0.7845134997452878,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003845,
      "loss": 6.6069,
      "mean_token_accuracy": 0.11322797611355781,
      "num_tokens": 1814775.0,
      "step": 770
    },
    {
      "entropy": 6.67111029624939,
      "epoch": 0.7896077432501274,
      "grad_norm": 1.359375,
      "learning_rate": 0.00038700000000000003,
      "loss": 6.5962,
      "mean_token_accuracy": 0.11556519940495491,
      "num_tokens": 1826068.0,
      "step": 775
    },
    {
      "entropy": 6.748933267593384,
      "epoch": 0.7947019867549668,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038950000000000003,
      "loss": 6.5929,
      "mean_token_accuracy": 0.1146427683532238,
      "num_tokens": 1838115.0,
      "step": 780
    },
    {
      "entropy": 6.834074974060059,
      "epoch": 0.7997962302598064,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00039200000000000004,
      "loss": 6.692,
      "mean_token_accuracy": 0.11109071299433708,
      "num_tokens": 1850670.0,
      "step": 785
    },
    {
      "entropy": 6.708985233306885,
      "epoch": 0.804890473764646,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00039450000000000005,
      "loss": 6.6679,
      "mean_token_accuracy": 0.11313713565468789,
      "num_tokens": 1863500.0,
      "step": 790
    },
    {
      "entropy": 6.828090238571167,
      "epoch": 0.8099847172694855,
      "grad_norm": 0.984375,
      "learning_rate": 0.00039700000000000005,
      "loss": 6.6777,
      "mean_token_accuracy": 0.11277235373854637,
      "num_tokens": 1875840.0,
      "step": 795
    },
    {
      "entropy": 6.7314472675323485,
      "epoch": 0.8150789607743251,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003995,
      "loss": 6.6138,
      "mean_token_accuracy": 0.11551589146256447,
      "num_tokens": 1887265.0,
      "step": 800
    },
    {
      "entropy": 6.735307455062866,
      "epoch": 0.8201732042791645,
      "grad_norm": 1.140625,
      "learning_rate": 0.000402,
      "loss": 6.4903,
      "mean_token_accuracy": 0.1243679828941822,
      "num_tokens": 1898550.0,
      "step": 805
    },
    {
      "entropy": 6.597352409362793,
      "epoch": 0.8252674477840041,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004045,
      "loss": 6.4267,
      "mean_token_accuracy": 0.12633467689156533,
      "num_tokens": 1908990.0,
      "step": 810
    },
    {
      "entropy": 6.649030351638794,
      "epoch": 0.8303616912888436,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040699999999999997,
      "loss": 6.5239,
      "mean_token_accuracy": 0.1194791056215763,
      "num_tokens": 1920109.0,
      "step": 815
    },
    {
      "entropy": 6.675817346572876,
      "epoch": 0.8354559347936832,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004095,
      "loss": 6.5882,
      "mean_token_accuracy": 0.1158361554145813,
      "num_tokens": 1930936.0,
      "step": 820
    },
    {
      "entropy": 6.7580108642578125,
      "epoch": 0.8405501782985226,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000412,
      "loss": 6.58,
      "mean_token_accuracy": 0.11999981254339218,
      "num_tokens": 1940968.0,
      "step": 825
    },
    {
      "entropy": 6.54950590133667,
      "epoch": 0.8456444218033622,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004145,
      "loss": 6.5245,
      "mean_token_accuracy": 0.12037713453173637,
      "num_tokens": 1952955.0,
      "step": 830
    },
    {
      "entropy": 6.697274208068848,
      "epoch": 0.8507386653082017,
      "grad_norm": 1.03125,
      "learning_rate": 0.000417,
      "loss": 6.6222,
      "mean_token_accuracy": 0.11852857768535614,
      "num_tokens": 1966023.0,
      "step": 835
    },
    {
      "entropy": 6.72812933921814,
      "epoch": 0.8558329088130413,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004195,
      "loss": 6.5554,
      "mean_token_accuracy": 0.12165540158748626,
      "num_tokens": 1978036.0,
      "step": 840
    },
    {
      "entropy": 6.567156553268433,
      "epoch": 0.8609271523178808,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000422,
      "loss": 6.529,
      "mean_token_accuracy": 0.11831365376710892,
      "num_tokens": 1988870.0,
      "step": 845
    },
    {
      "entropy": 6.692898988723755,
      "epoch": 0.8660213958227203,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004245,
      "loss": 6.6079,
      "mean_token_accuracy": 0.12303238585591317,
      "num_tokens": 2000010.0,
      "step": 850
    },
    {
      "entropy": 6.666892623901367,
      "epoch": 0.8711156393275599,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000427,
      "loss": 6.5306,
      "mean_token_accuracy": 0.11588020399212837,
      "num_tokens": 2011127.0,
      "step": 855
    },
    {
      "entropy": 6.691326379776001,
      "epoch": 0.8762098828323994,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004295,
      "loss": 6.6304,
      "mean_token_accuracy": 0.1136388897895813,
      "num_tokens": 2023483.0,
      "step": 860
    },
    {
      "entropy": 6.642430973052979,
      "epoch": 0.8813041263372389,
      "grad_norm": 1.078125,
      "learning_rate": 0.000432,
      "loss": 6.6096,
      "mean_token_accuracy": 0.1140765056014061,
      "num_tokens": 2034699.0,
      "step": 865
    },
    {
      "entropy": 6.6784350872039795,
      "epoch": 0.8863983698420784,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004345,
      "loss": 6.557,
      "mean_token_accuracy": 0.11363079845905304,
      "num_tokens": 2046849.0,
      "step": 870
    },
    {
      "entropy": 6.733535099029541,
      "epoch": 0.891492613346918,
      "grad_norm": 1.046875,
      "learning_rate": 0.000437,
      "loss": 6.6706,
      "mean_token_accuracy": 0.11693223789334298,
      "num_tokens": 2060535.0,
      "step": 875
    },
    {
      "entropy": 6.701980686187744,
      "epoch": 0.8965868568517575,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004395,
      "loss": 6.5288,
      "mean_token_accuracy": 0.12081804201006889,
      "num_tokens": 2072576.0,
      "step": 880
    },
    {
      "entropy": 6.632142019271851,
      "epoch": 0.9016811003565971,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000442,
      "loss": 6.5451,
      "mean_token_accuracy": 0.12307053580880165,
      "num_tokens": 2083884.0,
      "step": 885
    },
    {
      "entropy": 6.692302131652832,
      "epoch": 0.9067753438614365,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004445,
      "loss": 6.575,
      "mean_token_accuracy": 0.11638538911938667,
      "num_tokens": 2095429.0,
      "step": 890
    },
    {
      "entropy": 6.6273736476898195,
      "epoch": 0.9118695873662761,
      "grad_norm": 1.046875,
      "learning_rate": 0.000447,
      "loss": 6.5777,
      "mean_token_accuracy": 0.11308116614818572,
      "num_tokens": 2108351.0,
      "step": 895
    },
    {
      "entropy": 6.797583866119385,
      "epoch": 0.9169638308711157,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044950000000000003,
      "loss": 6.6103,
      "mean_token_accuracy": 0.11703968867659569,
      "num_tokens": 2120227.0,
      "step": 900
    },
    {
      "entropy": 6.566512250900269,
      "epoch": 0.9220580743759552,
      "grad_norm": 0.96875,
      "learning_rate": 0.00045200000000000004,
      "loss": 6.5417,
      "mean_token_accuracy": 0.12239225804805756,
      "num_tokens": 2131624.0,
      "step": 905
    },
    {
      "entropy": 6.566377592086792,
      "epoch": 0.9271523178807947,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045450000000000004,
      "loss": 6.5082,
      "mean_token_accuracy": 0.11846095621585846,
      "num_tokens": 2143523.0,
      "step": 910
    },
    {
      "entropy": 6.590761947631836,
      "epoch": 0.9322465613856342,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045700000000000005,
      "loss": 6.524,
      "mean_token_accuracy": 0.12397759556770324,
      "num_tokens": 2154952.0,
      "step": 915
    },
    {
      "entropy": 6.632454824447632,
      "epoch": 0.9373408048904738,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045950000000000006,
      "loss": 6.5431,
      "mean_token_accuracy": 0.1184918574988842,
      "num_tokens": 2166407.0,
      "step": 920
    },
    {
      "entropy": 6.586945915222168,
      "epoch": 0.9424350483953133,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000462,
      "loss": 6.5346,
      "mean_token_accuracy": 0.11630669236183167,
      "num_tokens": 2179748.0,
      "step": 925
    },
    {
      "entropy": 6.59289584159851,
      "epoch": 0.9475292919001528,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004645,
      "loss": 6.617,
      "mean_token_accuracy": 0.12137511149048805,
      "num_tokens": 2191266.0,
      "step": 930
    },
    {
      "entropy": 6.706100225448608,
      "epoch": 0.9526235354049923,
      "grad_norm": 1.125,
      "learning_rate": 0.000467,
      "loss": 6.5404,
      "mean_token_accuracy": 0.12512246817350386,
      "num_tokens": 2203209.0,
      "step": 935
    },
    {
      "entropy": 6.561947679519653,
      "epoch": 0.9577177789098319,
      "grad_norm": 1.25,
      "learning_rate": 0.0004695,
      "loss": 6.5379,
      "mean_token_accuracy": 0.12010324597358704,
      "num_tokens": 2214693.0,
      "step": 940
    },
    {
      "entropy": 6.549541568756103,
      "epoch": 0.9628120224146715,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000472,
      "loss": 6.4383,
      "mean_token_accuracy": 0.12597796693444252,
      "num_tokens": 2226778.0,
      "step": 945
    },
    {
      "entropy": 6.511363363265991,
      "epoch": 0.9679062659195109,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004745,
      "loss": 6.5258,
      "mean_token_accuracy": 0.11091761514544488,
      "num_tokens": 2238528.0,
      "step": 950
    },
    {
      "entropy": 6.628715085983276,
      "epoch": 0.9730005094243505,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000477,
      "loss": 6.5948,
      "mean_token_accuracy": 0.11641908511519432,
      "num_tokens": 2251483.0,
      "step": 955
    },
    {
      "entropy": 6.526347684860229,
      "epoch": 0.97809475292919,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004795,
      "loss": 6.4457,
      "mean_token_accuracy": 0.12605334371328353,
      "num_tokens": 2262151.0,
      "step": 960
    },
    {
      "entropy": 6.524391222000122,
      "epoch": 0.9831889964340296,
      "grad_norm": 1.03125,
      "learning_rate": 0.000482,
      "loss": 6.4294,
      "mean_token_accuracy": 0.12696175500750542,
      "num_tokens": 2274533.0,
      "step": 965
    },
    {
      "entropy": 6.529171466827393,
      "epoch": 0.9882832399388691,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004845,
      "loss": 6.4701,
      "mean_token_accuracy": 0.12050113677978516,
      "num_tokens": 2285518.0,
      "step": 970
    },
    {
      "entropy": 6.5718752384185795,
      "epoch": 0.9933774834437086,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000487,
      "loss": 6.5593,
      "mean_token_accuracy": 0.1192559577524662,
      "num_tokens": 2297900.0,
      "step": 975
    },
    {
      "entropy": 6.551716423034668,
      "epoch": 0.9984717269485481,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004895,
      "loss": 6.5026,
      "mean_token_accuracy": 0.13176016956567765,
      "num_tokens": 2311035.0,
      "step": 980
    },
    {
      "entropy": 6.543602095709907,
      "epoch": 1.0030565461029037,
      "grad_norm": 1.03125,
      "learning_rate": 0.000492,
      "loss": 6.3259,
      "mean_token_accuracy": 0.13025057150257957,
      "num_tokens": 2321287.0,
      "step": 985
    },
    {
      "entropy": 6.562080955505371,
      "epoch": 1.0081507896077433,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004945,
      "loss": 6.3717,
      "mean_token_accuracy": 0.12037833482027054,
      "num_tokens": 2333276.0,
      "step": 990
    },
    {
      "entropy": 6.429567813873291,
      "epoch": 1.0132450331125828,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000497,
      "loss": 6.238,
      "mean_token_accuracy": 0.13079071342945098,
      "num_tokens": 2346554.0,
      "step": 995
    },
    {
      "entropy": 6.421297168731689,
      "epoch": 1.0183392766174224,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004995,
      "loss": 6.255,
      "mean_token_accuracy": 0.13205485045909882,
      "num_tokens": 2357564.0,
      "step": 1000
    },
    {
      "epoch": 1.0183392766174224,
      "eval_entropy": 6.351307462762903,
      "eval_loss": 6.465834617614746,
      "eval_mean_token_accuracy": 0.12646097975012696,
      "eval_num_tokens": 2357564.0,
      "eval_runtime": 2.7035,
      "eval_samples_per_second": 1198.459,
      "eval_steps_per_second": 149.807,
      "step": 1000
    },
    {
      "entropy": 6.417219495773315,
      "epoch": 1.0234335201222617,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004999997711134041,
      "loss": 6.2666,
      "mean_token_accuracy": 0.13421913608908653,
      "num_tokens": 2369749.0,
      "step": 1005
    },
    {
      "entropy": 6.509257507324219,
      "epoch": 1.0285277636271013,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999988412624068,
      "loss": 6.3632,
      "mean_token_accuracy": 0.12740975096821786,
      "num_tokens": 2382622.0,
      "step": 1010
    },
    {
      "entropy": 6.493946552276611,
      "epoch": 1.0336220071319409,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999971961445492,
      "loss": 6.3463,
      "mean_token_accuracy": 0.12851536944508551,
      "num_tokens": 2394749.0,
      "step": 1015
    },
    {
      "entropy": 6.49571852684021,
      "epoch": 1.0387162506367804,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999948357650613,
      "loss": 6.3532,
      "mean_token_accuracy": 0.12557510510087014,
      "num_tokens": 2407868.0,
      "step": 1020
    },
    {
      "entropy": 6.418413162231445,
      "epoch": 1.04381049414162,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999917601314468,
      "loss": 6.2644,
      "mean_token_accuracy": 0.1351904697716236,
      "num_tokens": 2419045.0,
      "step": 1025
    },
    {
      "entropy": 6.423746490478516,
      "epoch": 1.0489047376464595,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004999879692534828,
      "loss": 6.3565,
      "mean_token_accuracy": 0.127122400701046,
      "num_tokens": 2431697.0,
      "step": 1030
    },
    {
      "entropy": 6.501759767532349,
      "epoch": 1.053998981151299,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999834631432208,
      "loss": 6.3452,
      "mean_token_accuracy": 0.12978592067956923,
      "num_tokens": 2442579.0,
      "step": 1035
    },
    {
      "entropy": 6.348742437362671,
      "epoch": 1.0590932246561386,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999782418149852,
      "loss": 6.1903,
      "mean_token_accuracy": 0.13898445516824723,
      "num_tokens": 2453601.0,
      "step": 1040
    },
    {
      "entropy": 6.388893413543701,
      "epoch": 1.064187468160978,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999723052853747,
      "loss": 6.2305,
      "mean_token_accuracy": 0.13225354850292206,
      "num_tokens": 2465500.0,
      "step": 1045
    },
    {
      "entropy": 6.277715349197388,
      "epoch": 1.0692817116658175,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999656535732615,
      "loss": 6.256,
      "mean_token_accuracy": 0.1289329580962658,
      "num_tokens": 2477948.0,
      "step": 1050
    },
    {
      "entropy": 6.455788087844849,
      "epoch": 1.074375955170657,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999582866997913,
      "loss": 6.2824,
      "mean_token_accuracy": 0.1348245456814766,
      "num_tokens": 2489575.0,
      "step": 1055
    },
    {
      "entropy": 6.437306356430054,
      "epoch": 1.0794701986754967,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000499950204688383,
      "loss": 6.2831,
      "mean_token_accuracy": 0.1257219858467579,
      "num_tokens": 2501615.0,
      "step": 1060
    },
    {
      "entropy": 6.318784904479981,
      "epoch": 1.0845644421803362,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999414075647292,
      "loss": 6.2665,
      "mean_token_accuracy": 0.13291230499744416,
      "num_tokens": 2513164.0,
      "step": 1065
    },
    {
      "entropy": 6.469882678985596,
      "epoch": 1.0896586856851758,
      "grad_norm": 1.203125,
      "learning_rate": 0.000499931895356796,
      "loss": 6.3728,
      "mean_token_accuracy": 0.13190104067325592,
      "num_tokens": 2524349.0,
      "step": 1070
    },
    {
      "entropy": 6.368467235565186,
      "epoch": 1.0947529291900153,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999216680948223,
      "loss": 6.2449,
      "mean_token_accuracy": 0.12897544875741004,
      "num_tokens": 2535834.0,
      "step": 1075
    },
    {
      "entropy": 6.293046236038208,
      "epoch": 1.0998471726948549,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999107258113204,
      "loss": 6.1801,
      "mean_token_accuracy": 0.13650451526045798,
      "num_tokens": 2547690.0,
      "step": 1080
    },
    {
      "entropy": 6.357415246963501,
      "epoch": 1.1049414161996944,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004998990685410759,
      "loss": 6.2374,
      "mean_token_accuracy": 0.13259905725717544,
      "num_tokens": 2559249.0,
      "step": 1085
    },
    {
      "entropy": 6.391598272323608,
      "epoch": 1.110035659704534,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998866963211465,
      "loss": 6.3529,
      "mean_token_accuracy": 0.1308108612895012,
      "num_tokens": 2573014.0,
      "step": 1090
    },
    {
      "entropy": 6.377676296234131,
      "epoch": 1.1151299032093733,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004998736091908636,
      "loss": 6.2418,
      "mean_token_accuracy": 0.13509300872683525,
      "num_tokens": 2584720.0,
      "step": 1095
    },
    {
      "entropy": 6.372010564804077,
      "epoch": 1.1202241467142129,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998598071918309,
      "loss": 6.2614,
      "mean_token_accuracy": 0.1345648229122162,
      "num_tokens": 2596711.0,
      "step": 1100
    },
    {
      "entropy": 6.382832479476929,
      "epoch": 1.1253183902190524,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998452903679246,
      "loss": 6.2082,
      "mean_token_accuracy": 0.13323341384530069,
      "num_tokens": 2606709.0,
      "step": 1105
    },
    {
      "entropy": 6.278462076187134,
      "epoch": 1.130412633723892,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004998300587652936,
      "loss": 6.2504,
      "mean_token_accuracy": 0.13396723121404647,
      "num_tokens": 2617233.0,
      "step": 1110
    },
    {
      "entropy": 6.448177194595337,
      "epoch": 1.1355068772287316,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004998141124323586,
      "loss": 6.2313,
      "mean_token_accuracy": 0.13162466883659363,
      "num_tokens": 2628313.0,
      "step": 1115
    },
    {
      "entropy": 6.1972558975219725,
      "epoch": 1.1406011207335711,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004997974514198126,
      "loss": 6.1596,
      "mean_token_accuracy": 0.1385033383965492,
      "num_tokens": 2639198.0,
      "step": 1120
    },
    {
      "entropy": 6.381111478805542,
      "epoch": 1.1456953642384107,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004997800757806209,
      "loss": 6.2595,
      "mean_token_accuracy": 0.13240717723965645,
      "num_tokens": 2650914.0,
      "step": 1125
    },
    {
      "entropy": 6.278826713562012,
      "epoch": 1.15078960774325,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004997619855700203,
      "loss": 6.2428,
      "mean_token_accuracy": 0.13494674637913703,
      "num_tokens": 2664335.0,
      "step": 1130
    },
    {
      "entropy": 6.331632471084594,
      "epoch": 1.1558838512480896,
      "grad_norm": 1.0625,
      "learning_rate": 0.000499743180845519,
      "loss": 6.3043,
      "mean_token_accuracy": 0.1335383266210556,
      "num_tokens": 2675839.0,
      "step": 1135
    },
    {
      "entropy": 6.197550296783447,
      "epoch": 1.1609780947529291,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000499723661666897,
      "loss": 6.0912,
      "mean_token_accuracy": 0.14703805819153787,
      "num_tokens": 2686793.0,
      "step": 1140
    },
    {
      "entropy": 6.288230514526367,
      "epoch": 1.1660723382577687,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004997034280962054,
      "loss": 6.3004,
      "mean_token_accuracy": 0.12843396440148352,
      "num_tokens": 2698504.0,
      "step": 1145
    },
    {
      "entropy": 6.3738954067230225,
      "epoch": 1.1711665817626082,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004996824801977663,
      "loss": 6.1969,
      "mean_token_accuracy": 0.13494786694645883,
      "num_tokens": 2710678.0,
      "step": 1150
    },
    {
      "entropy": 6.318151998519897,
      "epoch": 1.1762608252674478,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004996608180381727,
      "loss": 6.3212,
      "mean_token_accuracy": 0.13294611647725105,
      "num_tokens": 2722644.0,
      "step": 1155
    },
    {
      "entropy": 6.280697631835937,
      "epoch": 1.1813550687722874,
      "grad_norm": 1.078125,
      "learning_rate": 0.000499638441686288,
      "loss": 6.1941,
      "mean_token_accuracy": 0.13483970239758492,
      "num_tokens": 2735279.0,
      "step": 1160
    },
    {
      "entropy": 6.307227325439453,
      "epoch": 1.186449312277127,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004996153512132462,
      "loss": 6.2653,
      "mean_token_accuracy": 0.13486257940530777,
      "num_tokens": 2746905.0,
      "step": 1165
    },
    {
      "entropy": 6.3911340713500975,
      "epoch": 1.1915435557819665,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004995915466924517,
      "loss": 6.3079,
      "mean_token_accuracy": 0.13066884726285935,
      "num_tokens": 2760081.0,
      "step": 1170
    },
    {
      "entropy": 6.279552936553955,
      "epoch": 1.196637799286806,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004995670281995784,
      "loss": 6.2574,
      "mean_token_accuracy": 0.13016019761562347,
      "num_tokens": 2771416.0,
      "step": 1175
    },
    {
      "entropy": 6.394925212860107,
      "epoch": 1.2017320427916454,
      "grad_norm": 1.0,
      "learning_rate": 0.0004995417958125703,
      "loss": 6.3563,
      "mean_token_accuracy": 0.12249920591711998,
      "num_tokens": 2784216.0,
      "step": 1180
    },
    {
      "entropy": 6.398853206634522,
      "epoch": 1.206826286296485,
      "grad_norm": 1.0,
      "learning_rate": 0.0004995158496116402,
      "loss": 6.3365,
      "mean_token_accuracy": 0.13531623110175134,
      "num_tokens": 2795706.0,
      "step": 1185
    },
    {
      "entropy": 6.367800998687744,
      "epoch": 1.2119205298013245,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004994891896792712,
      "loss": 6.259,
      "mean_token_accuracy": 0.13469970226287842,
      "num_tokens": 2806317.0,
      "step": 1190
    },
    {
      "entropy": 6.373945808410644,
      "epoch": 1.217014773306164,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004994618161002142,
      "loss": 6.2667,
      "mean_token_accuracy": 0.13018080592155457,
      "num_tokens": 2817432.0,
      "step": 1195
    },
    {
      "entropy": 6.390775060653686,
      "epoch": 1.2221090168110036,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004994337289614893,
      "loss": 6.2062,
      "mean_token_accuracy": 0.13436981067061424,
      "num_tokens": 2828145.0,
      "step": 1200
    },
    {
      "entropy": 6.194883871078491,
      "epoch": 1.2272032603158431,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004994049283523854,
      "loss": 6.1799,
      "mean_token_accuracy": 0.14733498990535737,
      "num_tokens": 2838962.0,
      "step": 1205
    },
    {
      "entropy": 6.318769741058349,
      "epoch": 1.2322975038206827,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004993754143644585,
      "loss": 6.2335,
      "mean_token_accuracy": 0.13520962446928025,
      "num_tokens": 2850941.0,
      "step": 1210
    },
    {
      "entropy": 6.2552954196929935,
      "epoch": 1.237391747325522,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004993451870915333,
      "loss": 6.1947,
      "mean_token_accuracy": 0.13884784355759622,
      "num_tokens": 2863265.0,
      "step": 1215
    },
    {
      "entropy": 6.272042751312256,
      "epoch": 1.2424859908303616,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004993142466297016,
      "loss": 6.2024,
      "mean_token_accuracy": 0.13500213846564293,
      "num_tokens": 2876235.0,
      "step": 1220
    },
    {
      "entropy": 6.289383506774902,
      "epoch": 1.2475802343352012,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004992825930773228,
      "loss": 6.2629,
      "mean_token_accuracy": 0.12987583950161935,
      "num_tokens": 2889518.0,
      "step": 1225
    },
    {
      "entropy": 6.2832972526550295,
      "epoch": 1.2526744778400407,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004992502265350224,
      "loss": 6.2856,
      "mean_token_accuracy": 0.13477277979254723,
      "num_tokens": 2902417.0,
      "step": 1230
    },
    {
      "entropy": 6.272687149047852,
      "epoch": 1.2577687213448803,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004992171471056933,
      "loss": 6.1611,
      "mean_token_accuracy": 0.14579299539327623,
      "num_tokens": 2912904.0,
      "step": 1235
    },
    {
      "entropy": 6.331518888473511,
      "epoch": 1.2628629648497198,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004991833548944945,
      "loss": 6.2294,
      "mean_token_accuracy": 0.1335999645292759,
      "num_tokens": 2925583.0,
      "step": 1240
    },
    {
      "entropy": 6.22144513130188,
      "epoch": 1.2679572083545594,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004991488500088505,
      "loss": 6.1588,
      "mean_token_accuracy": 0.14021274149417878,
      "num_tokens": 2937916.0,
      "step": 1245
    },
    {
      "entropy": 6.189096450805664,
      "epoch": 1.273051451859399,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004991136325584516,
      "loss": 6.1824,
      "mean_token_accuracy": 0.13783264458179473,
      "num_tokens": 2949583.0,
      "step": 1250
    },
    {
      "entropy": 6.280501842498779,
      "epoch": 1.2781456953642385,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004990777026552536,
      "loss": 6.1986,
      "mean_token_accuracy": 0.1347845807671547,
      "num_tokens": 2962044.0,
      "step": 1255
    },
    {
      "entropy": 6.409678268432617,
      "epoch": 1.283239938869078,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004990410604134769,
      "loss": 6.3647,
      "mean_token_accuracy": 0.12630252316594123,
      "num_tokens": 2973570.0,
      "step": 1260
    },
    {
      "entropy": 6.199240875244141,
      "epoch": 1.2883341823739174,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004990037059496064,
      "loss": 6.2218,
      "mean_token_accuracy": 0.13618834540247918,
      "num_tokens": 2983683.0,
      "step": 1265
    },
    {
      "entropy": 6.397557020187378,
      "epoch": 1.293428425878757,
      "grad_norm": 1.125,
      "learning_rate": 0.0004989656393823913,
      "loss": 6.2489,
      "mean_token_accuracy": 0.13301505297422409,
      "num_tokens": 2995100.0,
      "step": 1270
    },
    {
      "entropy": 6.195611143112183,
      "epoch": 1.2985226693835965,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004989268608328443,
      "loss": 6.2716,
      "mean_token_accuracy": 0.1302342250943184,
      "num_tokens": 3006887.0,
      "step": 1275
    },
    {
      "entropy": 6.336117124557495,
      "epoch": 1.303616912888436,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004988873704242418,
      "loss": 6.2255,
      "mean_token_accuracy": 0.13720177486538887,
      "num_tokens": 3020444.0,
      "step": 1280
    },
    {
      "entropy": 6.2508533000946045,
      "epoch": 1.3087111563932756,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004988471682821229,
      "loss": 6.1674,
      "mean_token_accuracy": 0.13537554293870926,
      "num_tokens": 3032487.0,
      "step": 1285
    },
    {
      "entropy": 6.28198561668396,
      "epoch": 1.3138053998981152,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004988062545342894,
      "loss": 6.2708,
      "mean_token_accuracy": 0.13203708529472352,
      "num_tokens": 3044652.0,
      "step": 1290
    },
    {
      "entropy": 6.253536605834961,
      "epoch": 1.3188996434029547,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004987646293108056,
      "loss": 6.1579,
      "mean_token_accuracy": 0.13946222364902497,
      "num_tokens": 3056252.0,
      "step": 1295
    },
    {
      "entropy": 6.187241125106811,
      "epoch": 1.323993886907794,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004987222927439967,
      "loss": 6.175,
      "mean_token_accuracy": 0.14278096705675125,
      "num_tokens": 3066959.0,
      "step": 1300
    },
    {
      "entropy": 6.314873886108399,
      "epoch": 1.3290881304126336,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004986792449684504,
      "loss": 6.2857,
      "mean_token_accuracy": 0.13692365884780883,
      "num_tokens": 3079113.0,
      "step": 1305
    },
    {
      "entropy": 6.333209562301636,
      "epoch": 1.3341823739174732,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004986354861210144,
      "loss": 6.2074,
      "mean_token_accuracy": 0.13738809376955033,
      "num_tokens": 3090290.0,
      "step": 1310
    },
    {
      "entropy": 6.23097414970398,
      "epoch": 1.3392766174223127,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004985910163407973,
      "loss": 6.1646,
      "mean_token_accuracy": 0.14096444249153137,
      "num_tokens": 3101031.0,
      "step": 1315
    },
    {
      "entropy": 6.222257280349732,
      "epoch": 1.3443708609271523,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004985458357691675,
      "loss": 6.2189,
      "mean_token_accuracy": 0.13166088685393335,
      "num_tokens": 3113371.0,
      "step": 1320
    },
    {
      "entropy": 6.228028869628906,
      "epoch": 1.3494651044319919,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004984999445497533,
      "loss": 6.2237,
      "mean_token_accuracy": 0.13785334452986717,
      "num_tokens": 3125167.0,
      "step": 1325
    },
    {
      "entropy": 6.183880138397217,
      "epoch": 1.3545593479368314,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000498453342828442,
      "loss": 6.24,
      "mean_token_accuracy": 0.1362723782658577,
      "num_tokens": 3136945.0,
      "step": 1330
    },
    {
      "entropy": 6.322224140167236,
      "epoch": 1.359653591441671,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004984060307533794,
      "loss": 6.1555,
      "mean_token_accuracy": 0.14208487123250962,
      "num_tokens": 3147435.0,
      "step": 1335
    },
    {
      "entropy": 6.202198314666748,
      "epoch": 1.3647478349465105,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004983580084749698,
      "loss": 6.1869,
      "mean_token_accuracy": 0.1455744594335556,
      "num_tokens": 3157693.0,
      "step": 1340
    },
    {
      "entropy": 6.214845132827759,
      "epoch": 1.36984207845135,
      "grad_norm": 0.9296875,
      "learning_rate": 0.000498309276145875,
      "loss": 6.1204,
      "mean_token_accuracy": 0.14093155562877654,
      "num_tokens": 3169561.0,
      "step": 1345
    },
    {
      "entropy": 6.275221824645996,
      "epoch": 1.3749363219561896,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004982598339210142,
      "loss": 6.3096,
      "mean_token_accuracy": 0.12878431007266045,
      "num_tokens": 3183142.0,
      "step": 1350
    },
    {
      "entropy": 6.235415983200073,
      "epoch": 1.380030565461029,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004982096819575631,
      "loss": 6.2222,
      "mean_token_accuracy": 0.13532169610261918,
      "num_tokens": 3195703.0,
      "step": 1355
    },
    {
      "entropy": 6.144252157211303,
      "epoch": 1.3851248089658685,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004981588204149538,
      "loss": 6.1176,
      "mean_token_accuracy": 0.14307456761598586,
      "num_tokens": 3206603.0,
      "step": 1360
    },
    {
      "entropy": 6.285677289962768,
      "epoch": 1.390219052470708,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004981072494548743,
      "loss": 6.2249,
      "mean_token_accuracy": 0.13540513813495636,
      "num_tokens": 3218909.0,
      "step": 1365
    },
    {
      "entropy": 6.245230484008789,
      "epoch": 1.3953132959755477,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004980549692412676,
      "loss": 6.1677,
      "mean_token_accuracy": 0.1433960661292076,
      "num_tokens": 3230252.0,
      "step": 1370
    },
    {
      "entropy": 6.1964904308319095,
      "epoch": 1.4004075394803872,
      "grad_norm": 1.125,
      "learning_rate": 0.0004980019799403313,
      "loss": 6.2193,
      "mean_token_accuracy": 0.13967160061001777,
      "num_tokens": 3241227.0,
      "step": 1375
    },
    {
      "entropy": 6.211932706832886,
      "epoch": 1.4055017829852268,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004979482817205176,
      "loss": 6.0619,
      "mean_token_accuracy": 0.14620037674903869,
      "num_tokens": 3253569.0,
      "step": 1380
    },
    {
      "entropy": 6.2361126899719235,
      "epoch": 1.410596026490066,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004978938747525319,
      "loss": 6.1672,
      "mean_token_accuracy": 0.138123869150877,
      "num_tokens": 3265371.0,
      "step": 1385
    },
    {
      "entropy": 6.251908636093139,
      "epoch": 1.4156902699949057,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004978387592093329,
      "loss": 6.2013,
      "mean_token_accuracy": 0.13564864099025725,
      "num_tokens": 3276612.0,
      "step": 1390
    },
    {
      "entropy": 6.239566612243652,
      "epoch": 1.4207845134997452,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004977829352661318,
      "loss": 6.2196,
      "mean_token_accuracy": 0.1351855956017971,
      "num_tokens": 3288847.0,
      "step": 1395
    },
    {
      "entropy": 6.063872957229615,
      "epoch": 1.4258787570045848,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004977264031003919,
      "loss": 6.1702,
      "mean_token_accuracy": 0.14193566218018533,
      "num_tokens": 3300738.0,
      "step": 1400
    },
    {
      "entropy": 6.229835319519043,
      "epoch": 1.4309730005094243,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004976691628918278,
      "loss": 6.1568,
      "mean_token_accuracy": 0.14138704612851144,
      "num_tokens": 3312369.0,
      "step": 1405
    },
    {
      "entropy": 6.212840414047241,
      "epoch": 1.4360672440142639,
      "grad_norm": 0.87890625,
      "learning_rate": 0.000497611214822405,
      "loss": 6.1656,
      "mean_token_accuracy": 0.1398153379559517,
      "num_tokens": 3325277.0,
      "step": 1410
    },
    {
      "entropy": 6.102501726150512,
      "epoch": 1.4411614875191034,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004975525590763393,
      "loss": 6.0414,
      "mean_token_accuracy": 0.1459734968841076,
      "num_tokens": 3335881.0,
      "step": 1415
    },
    {
      "entropy": 6.301107358932495,
      "epoch": 1.446255731023943,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000497493195840096,
      "loss": 6.2092,
      "mean_token_accuracy": 0.13697200790047645,
      "num_tokens": 3348942.0,
      "step": 1420
    },
    {
      "entropy": 6.175320100784302,
      "epoch": 1.4513499745287826,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004974331253023898,
      "loss": 6.2037,
      "mean_token_accuracy": 0.14206972047686578,
      "num_tokens": 3361890.0,
      "step": 1425
    },
    {
      "entropy": 6.227373933792114,
      "epoch": 1.4564442180336221,
      "grad_norm": 1.0,
      "learning_rate": 0.0004973723476541836,
      "loss": 6.179,
      "mean_token_accuracy": 0.1412748746573925,
      "num_tokens": 3373297.0,
      "step": 1430
    },
    {
      "entropy": 6.052916145324707,
      "epoch": 1.4615384615384617,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004973108630886885,
      "loss": 6.0592,
      "mean_token_accuracy": 0.14644164815545083,
      "num_tokens": 3384623.0,
      "step": 1435
    },
    {
      "entropy": 6.1409101486206055,
      "epoch": 1.466632705043301,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004972486718013626,
      "loss": 6.1425,
      "mean_token_accuracy": 0.14146411642432213,
      "num_tokens": 3398379.0,
      "step": 1440
    },
    {
      "entropy": 6.276405096054077,
      "epoch": 1.4717269485481406,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004971857739899109,
      "loss": 6.1401,
      "mean_token_accuracy": 0.13381285667419435,
      "num_tokens": 3410928.0,
      "step": 1445
    },
    {
      "entropy": 6.092034387588501,
      "epoch": 1.4768211920529801,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004971221698542842,
      "loss": 6.0856,
      "mean_token_accuracy": 0.1479926437139511,
      "num_tokens": 3422173.0,
      "step": 1450
    },
    {
      "entropy": 6.1313581466674805,
      "epoch": 1.4819154355578197,
      "grad_norm": 1.0,
      "learning_rate": 0.0004970578595966789,
      "loss": 6.1405,
      "mean_token_accuracy": 0.14326228201389313,
      "num_tokens": 3434341.0,
      "step": 1455
    },
    {
      "entropy": 6.1404660701751705,
      "epoch": 1.4870096790626592,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004969928434215358,
      "loss": 6.1099,
      "mean_token_accuracy": 0.13853721991181372,
      "num_tokens": 3445605.0,
      "step": 1460
    },
    {
      "entropy": 6.229519939422607,
      "epoch": 1.4921039225674988,
      "grad_norm": 1.0,
      "learning_rate": 0.0004969271215355402,
      "loss": 6.1073,
      "mean_token_accuracy": 0.14102545380592346,
      "num_tokens": 3456957.0,
      "step": 1465
    },
    {
      "entropy": 6.172272682189941,
      "epoch": 1.4971981660723381,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004968606941476207,
      "loss": 6.1453,
      "mean_token_accuracy": 0.14370948895812036,
      "num_tokens": 3468340.0,
      "step": 1470
    },
    {
      "entropy": 6.181661653518677,
      "epoch": 1.5022924095771777,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004967935614689485,
      "loss": 6.1106,
      "mean_token_accuracy": 0.14410856515169143,
      "num_tokens": 3479137.0,
      "step": 1475
    },
    {
      "entropy": 6.121802949905396,
      "epoch": 1.5073866530820172,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004967257237129373,
      "loss": 6.0129,
      "mean_token_accuracy": 0.15014424845576285,
      "num_tokens": 3490525.0,
      "step": 1480
    },
    {
      "entropy": 6.079214286804199,
      "epoch": 1.5124808965868568,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004966571810952417,
      "loss": 6.0056,
      "mean_token_accuracy": 0.15413511991500856,
      "num_tokens": 3500731.0,
      "step": 1485
    },
    {
      "entropy": 6.110242938995361,
      "epoch": 1.5175751400916964,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004965879338337577,
      "loss": 6.1093,
      "mean_token_accuracy": 0.14118633270263672,
      "num_tokens": 3512944.0,
      "step": 1490
    },
    {
      "entropy": 6.12162446975708,
      "epoch": 1.522669383596536,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004965179821486205,
      "loss": 6.0753,
      "mean_token_accuracy": 0.13991874754428862,
      "num_tokens": 3524110.0,
      "step": 1495
    },
    {
      "entropy": 6.107229280471802,
      "epoch": 1.5277636271013755,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004964473262622054,
      "loss": 6.1368,
      "mean_token_accuracy": 0.14842215329408645,
      "num_tokens": 3536162.0,
      "step": 1500
    },
    {
      "epoch": 1.5277636271013755,
      "eval_entropy": 6.009506080768727,
      "eval_loss": 6.184627056121826,
      "eval_mean_token_accuracy": 0.14316514546488537,
      "eval_num_tokens": 3536162.0,
      "eval_runtime": 2.6604,
      "eval_samples_per_second": 1217.879,
      "eval_steps_per_second": 152.235,
      "step": 1500
    },
    {
      "entropy": 6.133478784561158,
      "epoch": 1.532857870606215,
      "grad_norm": 1.03125,
      "learning_rate": 0.000496375966399126,
      "loss": 6.0538,
      "mean_token_accuracy": 0.1511278361082077,
      "num_tokens": 3547756.0,
      "step": 1505
    },
    {
      "entropy": 5.997680425643921,
      "epoch": 1.5379521141110546,
      "grad_norm": 1.0,
      "learning_rate": 0.000496303902786234,
      "loss": 6.0479,
      "mean_token_accuracy": 0.15588610172271727,
      "num_tokens": 3559218.0,
      "step": 1510
    },
    {
      "entropy": 6.21874737739563,
      "epoch": 1.5430463576158941,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004962311356526181,
      "loss": 6.1004,
      "mean_token_accuracy": 0.14346015974879264,
      "num_tokens": 3571875.0,
      "step": 1515
    },
    {
      "entropy": 6.06115870475769,
      "epoch": 1.5481406011207337,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004961576652296036,
      "loss": 6.0109,
      "mean_token_accuracy": 0.14653350710868834,
      "num_tokens": 3583297.0,
      "step": 1520
    },
    {
      "entropy": 6.078774929046631,
      "epoch": 1.5532348446255733,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004960834917507516,
      "loss": 6.0777,
      "mean_token_accuracy": 0.14559792429208757,
      "num_tokens": 3595302.0,
      "step": 1525
    },
    {
      "entropy": 6.114213848114014,
      "epoch": 1.5583290881304126,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004960086154518578,
      "loss": 6.1155,
      "mean_token_accuracy": 0.1444586381316185,
      "num_tokens": 3606940.0,
      "step": 1530
    },
    {
      "entropy": 6.140858173370361,
      "epoch": 1.5634233316352522,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004959330365709528,
      "loss": 6.1742,
      "mean_token_accuracy": 0.13936522379517555,
      "num_tokens": 3619364.0,
      "step": 1535
    },
    {
      "entropy": 6.136679077148438,
      "epoch": 1.5685175751400917,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004958567553483003,
      "loss": 6.1883,
      "mean_token_accuracy": 0.1398473523557186,
      "num_tokens": 3631789.0,
      "step": 1540
    },
    {
      "entropy": 6.210715007781983,
      "epoch": 1.5736118186449313,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004957797720263967,
      "loss": 6.0968,
      "mean_token_accuracy": 0.1433316245675087,
      "num_tokens": 3643126.0,
      "step": 1545
    },
    {
      "entropy": 6.055673503875733,
      "epoch": 1.5787060621497706,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004957020868499706,
      "loss": 6.1212,
      "mean_token_accuracy": 0.14359405040740966,
      "num_tokens": 3654253.0,
      "step": 1550
    },
    {
      "entropy": 6.220635080337525,
      "epoch": 1.5838003056546102,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004956237000659815,
      "loss": 6.0909,
      "mean_token_accuracy": 0.14231853038072587,
      "num_tokens": 3666025.0,
      "step": 1555
    },
    {
      "entropy": 6.071724557876587,
      "epoch": 1.5888945491594497,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004955446119236195,
      "loss": 6.1059,
      "mean_token_accuracy": 0.14634145349264144,
      "num_tokens": 3676103.0,
      "step": 1560
    },
    {
      "entropy": 6.272754764556884,
      "epoch": 1.5939887926642893,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004954648226743043,
      "loss": 6.2349,
      "mean_token_accuracy": 0.13702599853277206,
      "num_tokens": 3688956.0,
      "step": 1565
    },
    {
      "entropy": 6.046952247619629,
      "epoch": 1.5990830361691288,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004953843325716841,
      "loss": 6.0908,
      "mean_token_accuracy": 0.1434723451733589,
      "num_tokens": 3700288.0,
      "step": 1570
    },
    {
      "entropy": 6.251407861709595,
      "epoch": 1.6041772796739684,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004953031418716356,
      "loss": 6.1313,
      "mean_token_accuracy": 0.1386256992816925,
      "num_tokens": 3711342.0,
      "step": 1575
    },
    {
      "entropy": 6.078168249130249,
      "epoch": 1.609271523178808,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004952212508322623,
      "loss": 6.0284,
      "mean_token_accuracy": 0.15199928134679794,
      "num_tokens": 3722195.0,
      "step": 1580
    },
    {
      "entropy": 6.161301946640014,
      "epoch": 1.6143657666836475,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004951386597138943,
      "loss": 6.0596,
      "mean_token_accuracy": 0.13910518661141397,
      "num_tokens": 3733345.0,
      "step": 1585
    },
    {
      "entropy": 6.0794189453125,
      "epoch": 1.619460010188487,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004950553687790871,
      "loss": 6.1234,
      "mean_token_accuracy": 0.14309294521808624,
      "num_tokens": 3746975.0,
      "step": 1590
    },
    {
      "entropy": 6.060288143157959,
      "epoch": 1.6245542536933266,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004949713782926208,
      "loss": 5.982,
      "mean_token_accuracy": 0.14610447585582734,
      "num_tokens": 3758027.0,
      "step": 1595
    },
    {
      "entropy": 6.145478296279907,
      "epoch": 1.6296484971981662,
      "grad_norm": 1.0,
      "learning_rate": 0.0004948866885214996,
      "loss": 6.024,
      "mean_token_accuracy": 0.1490251824259758,
      "num_tokens": 3769046.0,
      "step": 1600
    },
    {
      "entropy": 6.070356655120849,
      "epoch": 1.6347427407030057,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004948012997349506,
      "loss": 6.1225,
      "mean_token_accuracy": 0.13770004659891127,
      "num_tokens": 3781487.0,
      "step": 1605
    },
    {
      "entropy": 6.106816005706787,
      "epoch": 1.6398369842078453,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004947152122044232,
      "loss": 6.0784,
      "mean_token_accuracy": 0.1434733733534813,
      "num_tokens": 3793089.0,
      "step": 1610
    },
    {
      "entropy": 6.085491609573364,
      "epoch": 1.6449312277126846,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004946284262035875,
      "loss": 6.0534,
      "mean_token_accuracy": 0.1486929565668106,
      "num_tokens": 3804871.0,
      "step": 1615
    },
    {
      "entropy": 6.07642707824707,
      "epoch": 1.6500254712175242,
      "grad_norm": 1.03125,
      "learning_rate": 0.000494540942008335,
      "loss": 6.0325,
      "mean_token_accuracy": 0.1559026300907135,
      "num_tokens": 3815258.0,
      "step": 1620
    },
    {
      "entropy": 6.128446197509765,
      "epoch": 1.6551197147223637,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000494452759896776,
      "loss": 6.0568,
      "mean_token_accuracy": 0.14560223817825318,
      "num_tokens": 3826578.0,
      "step": 1625
    },
    {
      "entropy": 6.194732284545898,
      "epoch": 1.6602139582272033,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004943638801492396,
      "loss": 6.0686,
      "mean_token_accuracy": 0.14353293254971505,
      "num_tokens": 3837675.0,
      "step": 1630
    },
    {
      "entropy": 6.103059434890747,
      "epoch": 1.6653082017320426,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004942743030482728,
      "loss": 6.0959,
      "mean_token_accuracy": 0.14127544909715653,
      "num_tokens": 3850108.0,
      "step": 1635
    },
    {
      "entropy": 6.089843034744263,
      "epoch": 1.6704024452368822,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004941840288786397,
      "loss": 5.9985,
      "mean_token_accuracy": 0.14871155768632888,
      "num_tokens": 3861975.0,
      "step": 1640
    },
    {
      "entropy": 6.037532472610474,
      "epoch": 1.6754966887417218,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004940930579273198,
      "loss": 6.0814,
      "mean_token_accuracy": 0.14704819321632384,
      "num_tokens": 3873819.0,
      "step": 1645
    },
    {
      "entropy": 6.140864372253418,
      "epoch": 1.6805909322465613,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004940013904835079,
      "loss": 6.1276,
      "mean_token_accuracy": 0.14329199567437173,
      "num_tokens": 3885821.0,
      "step": 1650
    },
    {
      "entropy": 6.020464944839477,
      "epoch": 1.6856851757514009,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004939090268386132,
      "loss": 6.061,
      "mean_token_accuracy": 0.14725224077701568,
      "num_tokens": 3897078.0,
      "step": 1655
    },
    {
      "entropy": 6.177228784561157,
      "epoch": 1.6907794192562404,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004938159672862578,
      "loss": 6.1186,
      "mean_token_accuracy": 0.14301334992051123,
      "num_tokens": 3909073.0,
      "step": 1660
    },
    {
      "entropy": 6.1155922412872314,
      "epoch": 1.69587366276108,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004937222121222763,
      "loss": 6.0403,
      "mean_token_accuracy": 0.15009486079216003,
      "num_tokens": 3920209.0,
      "step": 1665
    },
    {
      "entropy": 6.1138163089752195,
      "epoch": 1.7009679062659195,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004936277616447144,
      "loss": 6.0812,
      "mean_token_accuracy": 0.14103334173560142,
      "num_tokens": 3933202.0,
      "step": 1670
    },
    {
      "entropy": 6.031678533554077,
      "epoch": 1.706062149770759,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004935326161538281,
      "loss": 5.9336,
      "mean_token_accuracy": 0.15366721600294114,
      "num_tokens": 3944574.0,
      "step": 1675
    },
    {
      "entropy": 6.11631498336792,
      "epoch": 1.7111563932755987,
      "grad_norm": 1.0,
      "learning_rate": 0.0004934367759520834,
      "loss": 6.1349,
      "mean_token_accuracy": 0.13778666779398918,
      "num_tokens": 3956614.0,
      "step": 1680
    },
    {
      "entropy": 6.039500665664673,
      "epoch": 1.7162506367804382,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004933402413441541,
      "loss": 5.9878,
      "mean_token_accuracy": 0.14830894619226456,
      "num_tokens": 3967854.0,
      "step": 1685
    },
    {
      "entropy": 6.038022136688232,
      "epoch": 1.7213448802852778,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004932430126369221,
      "loss": 6.0042,
      "mean_token_accuracy": 0.15413039326667785,
      "num_tokens": 3980267.0,
      "step": 1690
    },
    {
      "entropy": 6.1142689228057865,
      "epoch": 1.7264391237901173,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004931450901394753,
      "loss": 6.0956,
      "mean_token_accuracy": 0.14080657660961152,
      "num_tokens": 3991551.0,
      "step": 1695
    },
    {
      "entropy": 5.961064291000366,
      "epoch": 1.7315333672949567,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004930464741631073,
      "loss": 5.9363,
      "mean_token_accuracy": 0.15858148485422136,
      "num_tokens": 4001685.0,
      "step": 1700
    },
    {
      "entropy": 6.097273063659668,
      "epoch": 1.7366276107997962,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004929471650213167,
      "loss": 6.023,
      "mean_token_accuracy": 0.1522778809070587,
      "num_tokens": 4011960.0,
      "step": 1705
    },
    {
      "entropy": 6.09654483795166,
      "epoch": 1.7417218543046358,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004928471630298051,
      "loss": 6.0452,
      "mean_token_accuracy": 0.14763895571231841,
      "num_tokens": 4023608.0,
      "step": 1710
    },
    {
      "entropy": 6.096526288986206,
      "epoch": 1.7468160978094753,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004927464685064767,
      "loss": 6.0664,
      "mean_token_accuracy": 0.14474863559007645,
      "num_tokens": 4035365.0,
      "step": 1715
    },
    {
      "entropy": 6.047125577926636,
      "epoch": 1.7519103413143147,
      "grad_norm": 1.125,
      "learning_rate": 0.0004926450817714376,
      "loss": 5.976,
      "mean_token_accuracy": 0.1497434765100479,
      "num_tokens": 4046796.0,
      "step": 1720
    },
    {
      "entropy": 6.0006410598754885,
      "epoch": 1.7570045848191542,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004925430031469941,
      "loss": 6.0487,
      "mean_token_accuracy": 0.13952309116721154,
      "num_tokens": 4057749.0,
      "step": 1725
    },
    {
      "entropy": 6.124393129348755,
      "epoch": 1.7620988283239938,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004924402329576522,
      "loss": 6.1128,
      "mean_token_accuracy": 0.14142398536205292,
      "num_tokens": 4069491.0,
      "step": 1730
    },
    {
      "entropy": 6.0156410217285154,
      "epoch": 1.7671930718288333,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004923367715301164,
      "loss": 5.9604,
      "mean_token_accuracy": 0.149125587195158,
      "num_tokens": 4081552.0,
      "step": 1735
    },
    {
      "entropy": 6.112949228286743,
      "epoch": 1.772287315333673,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004922326191932882,
      "loss": 6.1289,
      "mean_token_accuracy": 0.14075233936309814,
      "num_tokens": 4094701.0,
      "step": 1740
    },
    {
      "entropy": 5.994672060012817,
      "epoch": 1.7773815588385125,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004921277762782661,
      "loss": 5.9122,
      "mean_token_accuracy": 0.15512796491384506,
      "num_tokens": 4105202.0,
      "step": 1745
    },
    {
      "entropy": 6.082731771469116,
      "epoch": 1.782475802343352,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004920222431183437,
      "loss": 6.1068,
      "mean_token_accuracy": 0.14282587841153144,
      "num_tokens": 4116447.0,
      "step": 1750
    },
    {
      "entropy": 6.129996824264526,
      "epoch": 1.7875700458481916,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004919160200490085,
      "loss": 6.1383,
      "mean_token_accuracy": 0.14118616431951522,
      "num_tokens": 4128103.0,
      "step": 1755
    },
    {
      "entropy": 6.121463251113892,
      "epoch": 1.7926642893530311,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004918091074079419,
      "loss": 6.1258,
      "mean_token_accuracy": 0.14120445325970649,
      "num_tokens": 4140259.0,
      "step": 1760
    },
    {
      "entropy": 6.058639049530029,
      "epoch": 1.7977585328578707,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004917015055350167,
      "loss": 6.0027,
      "mean_token_accuracy": 0.15162411332130432,
      "num_tokens": 4151513.0,
      "step": 1765
    },
    {
      "entropy": 6.050025558471679,
      "epoch": 1.8028527763627102,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004915932147722974,
      "loss": 6.0446,
      "mean_token_accuracy": 0.1424715355038643,
      "num_tokens": 4163539.0,
      "step": 1770
    },
    {
      "entropy": 6.072246265411377,
      "epoch": 1.8079470198675498,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004914842354640382,
      "loss": 6.0393,
      "mean_token_accuracy": 0.14208936542272568,
      "num_tokens": 4177173.0,
      "step": 1775
    },
    {
      "entropy": 6.0362903594970705,
      "epoch": 1.8130412633723894,
      "grad_norm": 0.8984375,
      "learning_rate": 0.000491374567956682,
      "loss": 6.0433,
      "mean_token_accuracy": 0.1477975770831108,
      "num_tokens": 4189756.0,
      "step": 1780
    },
    {
      "entropy": 6.149643087387085,
      "epoch": 1.8181355068772287,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004912642125988597,
      "loss": 6.1003,
      "mean_token_accuracy": 0.1537694126367569,
      "num_tokens": 4203594.0,
      "step": 1785
    },
    {
      "entropy": 6.061221504211426,
      "epoch": 1.8232297503820682,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004911531697413888,
      "loss": 6.0252,
      "mean_token_accuracy": 0.1486793577671051,
      "num_tokens": 4216189.0,
      "step": 1790
    },
    {
      "entropy": 6.0181516170501705,
      "epoch": 1.8283239938869078,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004910414397372724,
      "loss": 6.0248,
      "mean_token_accuracy": 0.1441796153783798,
      "num_tokens": 4228971.0,
      "step": 1795
    },
    {
      "entropy": 6.129302215576172,
      "epoch": 1.8334182373917474,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004909290229416979,
      "loss": 6.0547,
      "mean_token_accuracy": 0.1431230567395687,
      "num_tokens": 4240671.0,
      "step": 1800
    },
    {
      "entropy": 6.030335569381714,
      "epoch": 1.8385124808965867,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000490815919712036,
      "loss": 6.0351,
      "mean_token_accuracy": 0.1441195696592331,
      "num_tokens": 4253522.0,
      "step": 1805
    },
    {
      "entropy": 6.055653190612793,
      "epoch": 1.8436067244014263,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004907021304078398,
      "loss": 6.0676,
      "mean_token_accuracy": 0.1464722380042076,
      "num_tokens": 4266545.0,
      "step": 1810
    },
    {
      "entropy": 6.024935054779053,
      "epoch": 1.8487009679062658,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004905876553908432,
      "loss": 5.9734,
      "mean_token_accuracy": 0.15227250307798385,
      "num_tokens": 4278692.0,
      "step": 1815
    },
    {
      "entropy": 6.068737316131592,
      "epoch": 1.8537952114111054,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004904724950249599,
      "loss": 5.9898,
      "mean_token_accuracy": 0.15321767032146455,
      "num_tokens": 4291154.0,
      "step": 1820
    },
    {
      "entropy": 6.084896469116211,
      "epoch": 1.858889454915945,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004903566496762824,
      "loss": 6.0975,
      "mean_token_accuracy": 0.14444440975785255,
      "num_tokens": 4303963.0,
      "step": 1825
    },
    {
      "entropy": 5.955971765518188,
      "epoch": 1.8639836984207845,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000490240119713081,
      "loss": 5.894,
      "mean_token_accuracy": 0.15113551169633865,
      "num_tokens": 4316046.0,
      "step": 1830
    },
    {
      "entropy": 6.058926010131836,
      "epoch": 1.869077941925624,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004901229055058019,
      "loss": 5.9533,
      "mean_token_accuracy": 0.15235763043165207,
      "num_tokens": 4327631.0,
      "step": 1835
    },
    {
      "entropy": 6.0447791576385494,
      "epoch": 1.8741721854304636,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004900050074270669,
      "loss": 5.9223,
      "mean_token_accuracy": 0.15083706974983216,
      "num_tokens": 4338994.0,
      "step": 1840
    },
    {
      "entropy": 5.9091675758361815,
      "epoch": 1.8792664289353032,
      "grad_norm": 1.0,
      "learning_rate": 0.0004898864258516717,
      "loss": 5.9759,
      "mean_token_accuracy": 0.158118836581707,
      "num_tokens": 4351197.0,
      "step": 1845
    },
    {
      "entropy": 6.0877978801727295,
      "epoch": 1.8843606724401427,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004897671611565845,
      "loss": 6.0025,
      "mean_token_accuracy": 0.14755566269159318,
      "num_tokens": 4363497.0,
      "step": 1850
    },
    {
      "entropy": 6.058438396453857,
      "epoch": 1.8894549159449823,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004896472137209457,
      "loss": 6.0393,
      "mean_token_accuracy": 0.14204568788409233,
      "num_tokens": 4376518.0,
      "step": 1855
    },
    {
      "entropy": 5.988859415054321,
      "epoch": 1.8945491594498218,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004895265839260657,
      "loss": 5.911,
      "mean_token_accuracy": 0.1513100005686283,
      "num_tokens": 4387905.0,
      "step": 1860
    },
    {
      "entropy": 5.94207067489624,
      "epoch": 1.8996434029546614,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004894052721554242,
      "loss": 5.9952,
      "mean_token_accuracy": 0.15157082080841064,
      "num_tokens": 4400188.0,
      "step": 1865
    },
    {
      "entropy": 6.07070140838623,
      "epoch": 1.9047376464595007,
      "grad_norm": 1.125,
      "learning_rate": 0.0004892832787946691,
      "loss": 6.0288,
      "mean_token_accuracy": 0.144869889318943,
      "num_tokens": 4411578.0,
      "step": 1870
    },
    {
      "entropy": 6.027485084533692,
      "epoch": 1.9098318899643403,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004891606042316146,
      "loss": 5.9551,
      "mean_token_accuracy": 0.1512698881328106,
      "num_tokens": 4423405.0,
      "step": 1875
    },
    {
      "entropy": 5.933491897583008,
      "epoch": 1.9149261334691798,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004890372488562408,
      "loss": 5.9792,
      "mean_token_accuracy": 0.1508818507194519,
      "num_tokens": 4435049.0,
      "step": 1880
    },
    {
      "entropy": 6.040270662307739,
      "epoch": 1.9200203769740194,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004889132130606921,
      "loss": 5.9578,
      "mean_token_accuracy": 0.15416083931922914,
      "num_tokens": 4445613.0,
      "step": 1885
    },
    {
      "entropy": 5.975605201721192,
      "epoch": 1.9251146204788587,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004887884972392758,
      "loss": 5.9572,
      "mean_token_accuracy": 0.1484675571322441,
      "num_tokens": 4458356.0,
      "step": 1890
    },
    {
      "entropy": 6.027165937423706,
      "epoch": 1.9302088639836983,
      "grad_norm": 1.109375,
      "learning_rate": 0.000488663101788461,
      "loss": 5.9582,
      "mean_token_accuracy": 0.14620108753442765,
      "num_tokens": 4470279.0,
      "step": 1895
    },
    {
      "entropy": 6.105115556716919,
      "epoch": 1.9353031074885378,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004885370271068773,
      "loss": 6.1719,
      "mean_token_accuracy": 0.14128794223070146,
      "num_tokens": 4481611.0,
      "step": 1900
    },
    {
      "entropy": 6.095979356765747,
      "epoch": 1.9403973509933774,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004884102735953136,
      "loss": 5.9509,
      "mean_token_accuracy": 0.15084906220436095,
      "num_tokens": 4493384.0,
      "step": 1905
    },
    {
      "entropy": 5.874639368057251,
      "epoch": 1.945491594498217,
      "grad_norm": 1.03125,
      "learning_rate": 0.000488282841656717,
      "loss": 5.8558,
      "mean_token_accuracy": 0.15816857516765595,
      "num_tokens": 4504059.0,
      "step": 1910
    },
    {
      "entropy": 6.0805902004241945,
      "epoch": 1.9505858380030565,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00048815473169619097,
      "loss": 6.0322,
      "mean_token_accuracy": 0.14308103621006013,
      "num_tokens": 4516155.0,
      "step": 1915
    },
    {
      "entropy": 6.023840379714966,
      "epoch": 1.955680081507896,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004880259441209945,
      "loss": 5.9207,
      "mean_token_accuracy": 0.15768046155571938,
      "num_tokens": 4528030.0,
      "step": 1920
    },
    {
      "entropy": 6.012329912185669,
      "epoch": 1.9607743250127356,
      "grad_norm": 1.0,
      "learning_rate": 0.0004878964793405408,
      "loss": 6.0323,
      "mean_token_accuracy": 0.1385513499379158,
      "num_tokens": 4539217.0,
      "step": 1925
    },
    {
      "entropy": 6.005829763412476,
      "epoch": 1.9658685685175752,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048776633776639584,
      "loss": 5.8414,
      "mean_token_accuracy": 0.159397454559803,
      "num_tokens": 4550524.0,
      "step": 1930
    },
    {
      "entropy": 5.987122201919556,
      "epoch": 1.9709628120224147,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00048763551981227703,
      "loss": 5.9538,
      "mean_token_accuracy": 0.15249492228031158,
      "num_tokens": 4561693.0,
      "step": 1935
    },
    {
      "entropy": 5.944087219238281,
      "epoch": 1.9760570555272543,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048750402589405224,
      "loss": 5.9474,
      "mean_token_accuracy": 0.14776530414819716,
      "num_tokens": 4573971.0,
      "step": 1940
    },
    {
      "entropy": 6.08344521522522,
      "epoch": 1.9811512990320939,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048737185642973784,
      "loss": 5.9511,
      "mean_token_accuracy": 0.15248930752277373,
      "num_tokens": 4584870.0,
      "step": 1945
    },
    {
      "entropy": 5.944841814041138,
      "epoch": 1.9862455425369334,
      "grad_norm": 1.078125,
      "learning_rate": 0.000487239011839498,
      "loss": 6.0011,
      "mean_token_accuracy": 0.1480586051940918,
      "num_tokens": 4595973.0,
      "step": 1950
    },
    {
      "entropy": 6.001732110977173,
      "epoch": 1.9913397860417728,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004871054925456431,
      "loss": 5.849,
      "mean_token_accuracy": 0.15668184459209442,
      "num_tokens": 4607516.0,
      "step": 1955
    },
    {
      "entropy": 5.914603996276855,
      "epoch": 1.9964340295466123,
      "grad_norm": 0.90625,
      "learning_rate": 0.00048697129897262833,
      "loss": 5.9035,
      "mean_token_accuracy": 0.1592019245028496,
      "num_tokens": 4619382.0,
      "step": 1960
    },
    {
      "entropy": 6.0304126209682885,
      "epoch": 2.001018848700968,
      "grad_norm": 1.0,
      "learning_rate": 0.0004868364315470523,
      "loss": 5.9977,
      "mean_token_accuracy": 0.1434525102376938,
      "num_tokens": 4630049.0,
      "step": 1965
    },
    {
      "entropy": 5.960834646224976,
      "epoch": 2.0061130922058075,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00048670089069765594,
      "loss": 5.6245,
      "mean_token_accuracy": 0.16735252141952514,
      "num_tokens": 4642919.0,
      "step": 1970
    },
    {
      "entropy": 5.999763917922974,
      "epoch": 2.011207335710647,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004865646768553209,
      "loss": 5.7162,
      "mean_token_accuracy": 0.1654820777475834,
      "num_tokens": 4656620.0,
      "step": 1975
    },
    {
      "entropy": 5.935464525222779,
      "epoch": 2.0163015792154866,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004864277904530684,
      "loss": 5.6283,
      "mean_token_accuracy": 0.1659976363182068,
      "num_tokens": 4668764.0,
      "step": 1980
    },
    {
      "entropy": 5.838066720962525,
      "epoch": 2.021395822720326,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00048629023192605747,
      "loss": 5.5223,
      "mean_token_accuracy": 0.17317576110363006,
      "num_tokens": 4680014.0,
      "step": 1985
    },
    {
      "entropy": 5.937760353088379,
      "epoch": 2.0264900662251657,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048615200171158395,
      "loss": 5.5979,
      "mean_token_accuracy": 0.1677281975746155,
      "num_tokens": 4692677.0,
      "step": 1990
    },
    {
      "entropy": 5.756949186325073,
      "epoch": 2.0315843097300053,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00048601310024907904,
      "loss": 5.5487,
      "mean_token_accuracy": 0.17209948599338531,
      "num_tokens": 4704585.0,
      "step": 1995
    },
    {
      "entropy": 5.923199081420899,
      "epoch": 2.036678553234845,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048587352798010766,
      "loss": 5.6052,
      "mean_token_accuracy": 0.16866799145936967,
      "num_tokens": 4716418.0,
      "step": 2000
    },
    {
      "epoch": 2.036678553234845,
      "eval_entropy": 5.7684895903975875,
      "eval_loss": 5.99842643737793,
      "eval_mean_token_accuracy": 0.15436349115621897,
      "eval_num_tokens": 4716418.0,
      "eval_runtime": 2.6838,
      "eval_samples_per_second": 1207.227,
      "eval_steps_per_second": 150.903,
      "step": 2000
    },
    {
      "entropy": 5.864035272598267,
      "epoch": 2.0417727967396844,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00048573328534836735,
      "loss": 5.6371,
      "mean_token_accuracy": 0.16328690499067305,
      "num_tokens": 4727471.0,
      "step": 2005
    },
    {
      "entropy": 5.880073833465576,
      "epoch": 2.0468670402445235,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004855923727996866,
      "loss": 5.6391,
      "mean_token_accuracy": 0.16733508110046386,
      "num_tokens": 4740139.0,
      "step": 2010
    },
    {
      "entropy": 5.8417088985443115,
      "epoch": 2.051961283749363,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004854507907820237,
      "loss": 5.5478,
      "mean_token_accuracy": 0.1754288911819458,
      "num_tokens": 4750757.0,
      "step": 2015
    },
    {
      "entropy": 5.875588130950928,
      "epoch": 2.0570555272542026,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00048530853974546495,
      "loss": 5.6349,
      "mean_token_accuracy": 0.16513933688402177,
      "num_tokens": 4763291.0,
      "step": 2020
    },
    {
      "entropy": 5.898016309738159,
      "epoch": 2.062149770759042,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004851656201422237,
      "loss": 5.5786,
      "mean_token_accuracy": 0.16381702572107315,
      "num_tokens": 4773925.0,
      "step": 2025
    },
    {
      "entropy": 5.860568046569824,
      "epoch": 2.0672440142638817,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048502203242663846,
      "loss": 5.5818,
      "mean_token_accuracy": 0.17210538536310196,
      "num_tokens": 4785264.0,
      "step": 2030
    },
    {
      "entropy": 5.868568563461304,
      "epoch": 2.0723382577687213,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00048487777705517177,
      "loss": 5.721,
      "mean_token_accuracy": 0.1673053503036499,
      "num_tokens": 4797670.0,
      "step": 2035
    },
    {
      "entropy": 5.931161165237427,
      "epoch": 2.077432501273561,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00048473285448640853,
      "loss": 5.6954,
      "mean_token_accuracy": 0.1602186843752861,
      "num_tokens": 4809408.0,
      "step": 2040
    },
    {
      "entropy": 5.822083425521851,
      "epoch": 2.0825267447784004,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004845872651810549,
      "loss": 5.6777,
      "mean_token_accuracy": 0.15986730307340621,
      "num_tokens": 4821655.0,
      "step": 2045
    },
    {
      "entropy": 5.858307266235352,
      "epoch": 2.08762098828324,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00048444100960193617,
      "loss": 5.5558,
      "mean_token_accuracy": 0.17199701070785522,
      "num_tokens": 4833879.0,
      "step": 2050
    },
    {
      "entropy": 5.790414524078369,
      "epoch": 2.0927152317880795,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004842940882139961,
      "loss": 5.6486,
      "mean_token_accuracy": 0.16194527447223664,
      "num_tokens": 4846054.0,
      "step": 2055
    },
    {
      "entropy": 5.9548438549041744,
      "epoch": 2.097809475292919,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00048414650148429465,
      "loss": 5.5922,
      "mean_token_accuracy": 0.17124049961566926,
      "num_tokens": 4857508.0,
      "step": 2060
    },
    {
      "entropy": 5.797548818588257,
      "epoch": 2.1029037187977586,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048399824988200747,
      "loss": 5.65,
      "mean_token_accuracy": 0.16566074639558792,
      "num_tokens": 4868291.0,
      "step": 2065
    },
    {
      "entropy": 5.945477485656738,
      "epoch": 2.107997962302598,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00048384933387842313,
      "loss": 5.6588,
      "mean_token_accuracy": 0.1591070517897606,
      "num_tokens": 4881769.0,
      "step": 2070
    },
    {
      "entropy": 5.8316374778747555,
      "epoch": 2.1130922058074377,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004836997539469429,
      "loss": 5.6624,
      "mean_token_accuracy": 0.16767859160900117,
      "num_tokens": 4893518.0,
      "step": 2075
    },
    {
      "entropy": 5.816946315765381,
      "epoch": 2.1181864493122773,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048354951056307814,
      "loss": 5.5837,
      "mean_token_accuracy": 0.16843218803405763,
      "num_tokens": 4903881.0,
      "step": 2080
    },
    {
      "entropy": 5.846394300460815,
      "epoch": 2.123280692817117,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048339860420444994,
      "loss": 5.6629,
      "mean_token_accuracy": 0.16391829177737235,
      "num_tokens": 4915749.0,
      "step": 2085
    },
    {
      "entropy": 5.935555076599121,
      "epoch": 2.128374936321956,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048324703535078633,
      "loss": 5.7187,
      "mean_token_accuracy": 0.16316656917333602,
      "num_tokens": 4927856.0,
      "step": 2090
    },
    {
      "entropy": 5.842387676239014,
      "epoch": 2.1334691798267955,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048309480448392185,
      "loss": 5.6139,
      "mean_token_accuracy": 0.16263346821069719,
      "num_tokens": 4939096.0,
      "step": 2095
    },
    {
      "entropy": 5.804424142837524,
      "epoch": 2.138563423331635,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004829419120877954,
      "loss": 5.6006,
      "mean_token_accuracy": 0.17247907668352128,
      "num_tokens": 4950309.0,
      "step": 2100
    },
    {
      "entropy": 5.797757148742676,
      "epoch": 2.1436576668364746,
      "grad_norm": 1.0,
      "learning_rate": 0.0004827883586484488,
      "loss": 5.5648,
      "mean_token_accuracy": 0.1749133437871933,
      "num_tokens": 4962283.0,
      "step": 2105
    },
    {
      "entropy": 5.8026713848114015,
      "epoch": 2.148751910341314,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004826341446540254,
      "loss": 5.6037,
      "mean_token_accuracy": 0.16955575197935105,
      "num_tokens": 4973398.0,
      "step": 2110
    },
    {
      "entropy": 5.781595659255982,
      "epoch": 2.1538461538461537,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048247927059476866,
      "loss": 5.6195,
      "mean_token_accuracy": 0.16578615158796312,
      "num_tokens": 4984866.0,
      "step": 2115
    },
    {
      "entropy": 5.857525587081909,
      "epoch": 2.1589403973509933,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048232373696301985,
      "loss": 5.6608,
      "mean_token_accuracy": 0.16607604920864105,
      "num_tokens": 4996078.0,
      "step": 2120
    },
    {
      "entropy": 5.85476131439209,
      "epoch": 2.164034640855833,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048216754425321734,
      "loss": 5.7085,
      "mean_token_accuracy": 0.15740856230258943,
      "num_tokens": 5008575.0,
      "step": 2125
    },
    {
      "entropy": 5.965959453582764,
      "epoch": 2.1691288843606724,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004820106929618948,
      "loss": 5.7389,
      "mean_token_accuracy": 0.16028655618429183,
      "num_tokens": 5020762.0,
      "step": 2130
    },
    {
      "entropy": 5.710131502151489,
      "epoch": 2.174223127865512,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048185318358767926,
      "loss": 5.5819,
      "mean_token_accuracy": 0.1731780633330345,
      "num_tokens": 5032700.0,
      "step": 2135
    },
    {
      "entropy": 5.781336498260498,
      "epoch": 2.1793173713703515,
      "grad_norm": 1.0,
      "learning_rate": 0.00048169501663129007,
      "loss": 5.5361,
      "mean_token_accuracy": 0.176752208173275,
      "num_tokens": 5044189.0,
      "step": 2140
    },
    {
      "entropy": 5.842349052429199,
      "epoch": 2.184411614875191,
      "grad_norm": 1.0,
      "learning_rate": 0.00048153619259553665,
      "loss": 5.619,
      "mean_token_accuracy": 0.1670036718249321,
      "num_tokens": 5055674.0,
      "step": 2145
    },
    {
      "entropy": 5.748504686355591,
      "epoch": 2.1895058583800306,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00048137671198531765,
      "loss": 5.7283,
      "mean_token_accuracy": 0.16206004619598388,
      "num_tokens": 5068641.0,
      "step": 2150
    },
    {
      "entropy": 5.860818862915039,
      "epoch": 2.19460010188487,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048121657530761857,
      "loss": 5.6305,
      "mean_token_accuracy": 0.16791221201419831,
      "num_tokens": 5080012.0,
      "step": 2155
    },
    {
      "entropy": 5.751377248764038,
      "epoch": 2.1996943453897098,
      "grad_norm": 1.0,
      "learning_rate": 0.00048105578307151094,
      "loss": 5.6591,
      "mean_token_accuracy": 0.16100457757711412,
      "num_tokens": 5092379.0,
      "step": 2160
    },
    {
      "entropy": 5.901716804504394,
      "epoch": 2.2047885888945493,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048089433578814994,
      "loss": 5.6864,
      "mean_token_accuracy": 0.167290398478508,
      "num_tokens": 5104126.0,
      "step": 2165
    },
    {
      "entropy": 5.851080226898193,
      "epoch": 2.209882832399389,
      "grad_norm": 1.0,
      "learning_rate": 0.0004807322339707733,
      "loss": 5.6347,
      "mean_token_accuracy": 0.16827466040849687,
      "num_tokens": 5115888.0,
      "step": 2170
    },
    {
      "entropy": 5.851242160797119,
      "epoch": 2.214977075904228,
      "grad_norm": 1.171875,
      "learning_rate": 0.00048056947813469967,
      "loss": 5.6182,
      "mean_token_accuracy": 0.1718982547521591,
      "num_tokens": 5126675.0,
      "step": 2175
    },
    {
      "entropy": 5.8321527481079105,
      "epoch": 2.220071319409068,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004804060687973264,
      "loss": 5.6637,
      "mean_token_accuracy": 0.1679067075252533,
      "num_tokens": 5140403.0,
      "step": 2180
    },
    {
      "entropy": 5.832393741607666,
      "epoch": 2.225165562913907,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004802420064781286,
      "loss": 5.6442,
      "mean_token_accuracy": 0.1621223583817482,
      "num_tokens": 5153367.0,
      "step": 2185
    },
    {
      "entropy": 5.721229553222656,
      "epoch": 2.2302598064187467,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004800772916986572,
      "loss": 5.563,
      "mean_token_accuracy": 0.17193944603204728,
      "num_tokens": 5165254.0,
      "step": 2190
    },
    {
      "entropy": 5.812705612182617,
      "epoch": 2.235354049923586,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047991192498253715,
      "loss": 5.5421,
      "mean_token_accuracy": 0.1765172675251961,
      "num_tokens": 5176107.0,
      "step": 2195
    },
    {
      "entropy": 5.749564123153687,
      "epoch": 2.2404482934284258,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000479745906855466,
      "loss": 5.6645,
      "mean_token_accuracy": 0.16781434416770935,
      "num_tokens": 5187193.0,
      "step": 2200
    },
    {
      "entropy": 5.786741256713867,
      "epoch": 2.2455425369332653,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047957923784521206,
      "loss": 5.6125,
      "mean_token_accuracy": 0.16534093022346497,
      "num_tokens": 5198234.0,
      "step": 2205
    },
    {
      "entropy": 5.81756010055542,
      "epoch": 2.250636780438105,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004794119184816129,
      "loss": 5.6576,
      "mean_token_accuracy": 0.16438238620758056,
      "num_tokens": 5211195.0,
      "step": 2210
    },
    {
      "entropy": 5.848905324935913,
      "epoch": 2.2557310239429444,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004792439492965734,
      "loss": 5.6329,
      "mean_token_accuracy": 0.17122048288583755,
      "num_tokens": 5222032.0,
      "step": 2215
    },
    {
      "entropy": 5.802203178405762,
      "epoch": 2.260825267447784,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004790753308240642,
      "loss": 5.6236,
      "mean_token_accuracy": 0.16637725085020066,
      "num_tokens": 5233891.0,
      "step": 2220
    },
    {
      "entropy": 5.774746561050415,
      "epoch": 2.2659195109526236,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004789060636001202,
      "loss": 5.6555,
      "mean_token_accuracy": 0.1662521168589592,
      "num_tokens": 5245258.0,
      "step": 2225
    },
    {
      "entropy": 5.850039482116699,
      "epoch": 2.271013754457463,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004787361481628385,
      "loss": 5.6726,
      "mean_token_accuracy": 0.16486700773239135,
      "num_tokens": 5257738.0,
      "step": 2230
    },
    {
      "entropy": 5.734911489486694,
      "epoch": 2.2761079979623027,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004785655850523769,
      "loss": 5.5604,
      "mean_token_accuracy": 0.1680548220872879,
      "num_tokens": 5268577.0,
      "step": 2235
    },
    {
      "entropy": 5.799193048477173,
      "epoch": 2.2812022414671422,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047839437481095223,
      "loss": 5.6456,
      "mean_token_accuracy": 0.16766868084669112,
      "num_tokens": 5279920.0,
      "step": 2240
    },
    {
      "entropy": 5.759489250183106,
      "epoch": 2.286296484971982,
      "grad_norm": 1.0,
      "learning_rate": 0.00047822251798283836,
      "loss": 5.6756,
      "mean_token_accuracy": 0.16465742886066437,
      "num_tokens": 5291543.0,
      "step": 2245
    },
    {
      "entropy": 5.816216707229614,
      "epoch": 2.2913907284768213,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004780500151143647,
      "loss": 5.595,
      "mean_token_accuracy": 0.16568690836429595,
      "num_tokens": 5302076.0,
      "step": 2250
    },
    {
      "entropy": 5.809090328216553,
      "epoch": 2.296484971981661,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004778768667539147,
      "loss": 5.5667,
      "mean_token_accuracy": 0.16760059148073198,
      "num_tokens": 5312902.0,
      "step": 2255
    },
    {
      "entropy": 5.769825649261475,
      "epoch": 2.3015792154865,
      "grad_norm": 0.9375,
      "learning_rate": 0.00047770307345192326,
      "loss": 5.6814,
      "mean_token_accuracy": 0.17060534060001373,
      "num_tokens": 5325229.0,
      "step": 2260
    },
    {
      "entropy": 5.801876735687256,
      "epoch": 2.30667345899134,
      "grad_norm": 0.96875,
      "learning_rate": 0.00047752863576087603,
      "loss": 5.6027,
      "mean_token_accuracy": 0.17036211788654326,
      "num_tokens": 5336942.0,
      "step": 2265
    },
    {
      "entropy": 5.777912712097168,
      "epoch": 2.311767702496179,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000477353554235307,
      "loss": 5.6544,
      "mean_token_accuracy": 0.16362376362085343,
      "num_tokens": 5349148.0,
      "step": 2270
    },
    {
      "entropy": 5.825894737243653,
      "epoch": 2.3168619460010187,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047717782943179664,
      "loss": 5.6254,
      "mean_token_accuracy": 0.16521700322628022,
      "num_tokens": 5361452.0,
      "step": 2275
    },
    {
      "entropy": 5.766107845306396,
      "epoch": 2.3219561895058582,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004770014619089709,
      "loss": 5.573,
      "mean_token_accuracy": 0.1703086569905281,
      "num_tokens": 5372766.0,
      "step": 2280
    },
    {
      "entropy": 5.826212167739868,
      "epoch": 2.327050433010698,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004768244522274985,
      "loss": 5.7871,
      "mean_token_accuracy": 0.1534041740000248,
      "num_tokens": 5385744.0,
      "step": 2285
    },
    {
      "entropy": 5.8942187309265135,
      "epoch": 2.3321446765155374,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00047664680095008973,
      "loss": 5.6314,
      "mean_token_accuracy": 0.1623341202735901,
      "num_tokens": 5397701.0,
      "step": 2290
    },
    {
      "entropy": 5.819622945785523,
      "epoch": 2.337238920020377,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004764685086414945,
      "loss": 5.7039,
      "mean_token_accuracy": 0.155191071331501,
      "num_tokens": 5410025.0,
      "step": 2295
    },
    {
      "entropy": 5.783815097808838,
      "epoch": 2.3423331635252165,
      "grad_norm": 0.96875,
      "learning_rate": 0.00047628957586850053,
      "loss": 5.653,
      "mean_token_accuracy": 0.16540868431329728,
      "num_tokens": 5421298.0,
      "step": 2300
    },
    {
      "entropy": 5.726611375808716,
      "epoch": 2.347427407030056,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004761100031999315,
      "loss": 5.588,
      "mean_token_accuracy": 0.1762511447072029,
      "num_tokens": 5433010.0,
      "step": 2305
    },
    {
      "entropy": 5.692564344406128,
      "epoch": 2.3525216505348956,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004759297912066454,
      "loss": 5.6095,
      "mean_token_accuracy": 0.16813477128744125,
      "num_tokens": 5445780.0,
      "step": 2310
    },
    {
      "entropy": 5.82792558670044,
      "epoch": 2.357615894039735,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047574894046153266,
      "loss": 5.6448,
      "mean_token_accuracy": 0.17046891897916794,
      "num_tokens": 5457012.0,
      "step": 2315
    },
    {
      "entropy": 5.660221290588379,
      "epoch": 2.3627101375445747,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004755674515395141,
      "loss": 5.6803,
      "mean_token_accuracy": 0.16038270592689513,
      "num_tokens": 5469602.0,
      "step": 2320
    },
    {
      "entropy": 5.884269332885742,
      "epoch": 2.3678043810494143,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004753853250175395,
      "loss": 5.6395,
      "mean_token_accuracy": 0.16682495325803756,
      "num_tokens": 5480266.0,
      "step": 2325
    },
    {
      "entropy": 5.727888298034668,
      "epoch": 2.372898624554254,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047520256147458545,
      "loss": 5.605,
      "mean_token_accuracy": 0.1654973953962326,
      "num_tokens": 5491269.0,
      "step": 2330
    },
    {
      "entropy": 5.808749675750732,
      "epoch": 2.3779928680590934,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004750191614916536,
      "loss": 5.6353,
      "mean_token_accuracy": 0.16049928814172745,
      "num_tokens": 5502690.0,
      "step": 2335
    },
    {
      "entropy": 5.739553165435791,
      "epoch": 2.383087111563933,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004748351256517689,
      "loss": 5.5995,
      "mean_token_accuracy": 0.16872989684343337,
      "num_tokens": 5514278.0,
      "step": 2340
    },
    {
      "entropy": 5.77966799736023,
      "epoch": 2.388181355068772,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047465045453997756,
      "loss": 5.7252,
      "mean_token_accuracy": 0.1624407634139061,
      "num_tokens": 5526993.0,
      "step": 2345
    },
    {
      "entropy": 5.821387529373169,
      "epoch": 2.393275598573612,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004744651487433455,
      "loss": 5.6013,
      "mean_token_accuracy": 0.1661293163895607,
      "num_tokens": 5538504.0,
      "step": 2350
    },
    {
      "entropy": 5.689996433258057,
      "epoch": 2.398369842078451,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004742792088509561,
      "loss": 5.5442,
      "mean_token_accuracy": 0.17447483390569687,
      "num_tokens": 5550848.0,
      "step": 2355
    },
    {
      "entropy": 5.726719665527344,
      "epoch": 2.4034640855832907,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00047409263545390864,
      "loss": 5.5186,
      "mean_token_accuracy": 0.17505260407924653,
      "num_tokens": 5561678.0,
      "step": 2360
    },
    {
      "entropy": 5.708486413955688,
      "epoch": 2.4085583290881303,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004739054291453162,
      "loss": 5.5358,
      "mean_token_accuracy": 0.17320602983236313,
      "num_tokens": 5574320.0,
      "step": 2365
    },
    {
      "entropy": 5.749895763397217,
      "epoch": 2.41365257259297,
      "grad_norm": 1.0,
      "learning_rate": 0.00047371759052030395,
      "loss": 5.7173,
      "mean_token_accuracy": 0.1599662110209465,
      "num_tokens": 5587387.0,
      "step": 2370
    },
    {
      "entropy": 5.725869083404541,
      "epoch": 2.4187468160978094,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004735291201760072,
      "loss": 5.65,
      "mean_token_accuracy": 0.17233884781599046,
      "num_tokens": 5599760.0,
      "step": 2375
    },
    {
      "entropy": 5.844428396224975,
      "epoch": 2.423841059602649,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004733400187115694,
      "loss": 5.6392,
      "mean_token_accuracy": 0.1669354572892189,
      "num_tokens": 5611451.0,
      "step": 2380
    },
    {
      "entropy": 5.825902318954467,
      "epoch": 2.4289353031074885,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004731502867281404,
      "loss": 5.6862,
      "mean_token_accuracy": 0.16543307900428772,
      "num_tokens": 5622687.0,
      "step": 2385
    },
    {
      "entropy": 5.793111181259155,
      "epoch": 2.434029546612328,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047295992482887435,
      "loss": 5.7076,
      "mean_token_accuracy": 0.1622299462556839,
      "num_tokens": 5634126.0,
      "step": 2390
    },
    {
      "entropy": 5.809430932998657,
      "epoch": 2.4391237901171676,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047276893361892803,
      "loss": 5.6028,
      "mean_token_accuracy": 0.17397123277187349,
      "num_tokens": 5646322.0,
      "step": 2395
    },
    {
      "entropy": 5.758405590057373,
      "epoch": 2.444218033622007,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047257731370545865,
      "loss": 5.6194,
      "mean_token_accuracy": 0.17260859459638594,
      "num_tokens": 5657530.0,
      "step": 2400
    },
    {
      "entropy": 5.685599422454834,
      "epoch": 2.4493122771268467,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004723850656976221,
      "loss": 5.5587,
      "mean_token_accuracy": 0.16950662434101105,
      "num_tokens": 5669471.0,
      "step": 2405
    },
    {
      "entropy": 5.683564805984497,
      "epoch": 2.4544065206316863,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004721921902065711,
      "loss": 5.6158,
      "mean_token_accuracy": 0.16663512587547302,
      "num_tokens": 5681582.0,
      "step": 2410
    },
    {
      "entropy": 5.806542825698853,
      "epoch": 2.459500764136526,
      "grad_norm": 0.984375,
      "learning_rate": 0.000471998687845453,
      "loss": 5.6616,
      "mean_token_accuracy": 0.16524447202682496,
      "num_tokens": 5694195.0,
      "step": 2415
    },
    {
      "entropy": 5.713834476470947,
      "epoch": 2.4645950076413654,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00047180455922940783,
      "loss": 5.653,
      "mean_token_accuracy": 0.16521431356668473,
      "num_tokens": 5707677.0,
      "step": 2420
    },
    {
      "entropy": 5.837003231048584,
      "epoch": 2.469689251146205,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004716098049755668,
      "loss": 5.6944,
      "mean_token_accuracy": 0.16017566323280336,
      "num_tokens": 5719538.0,
      "step": 2425
    },
    {
      "entropy": 5.76138482093811,
      "epoch": 2.474783494651044,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004714144257030497,
      "loss": 5.6475,
      "mean_token_accuracy": 0.16634842604398728,
      "num_tokens": 5730323.0,
      "step": 2430
    },
    {
      "entropy": 5.6930094242095945,
      "epoch": 2.479877738155884,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004712184220329634,
      "loss": 5.5474,
      "mean_token_accuracy": 0.1701219290494919,
      "num_tokens": 5741322.0,
      "step": 2435
    },
    {
      "entropy": 5.7122878074646,
      "epoch": 2.484971981660723,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047102179458839965,
      "loss": 5.573,
      "mean_token_accuracy": 0.17167485058307647,
      "num_tokens": 5753145.0,
      "step": 2440
    },
    {
      "entropy": 5.745958232879639,
      "epoch": 2.4900662251655628,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004708245439944333,
      "loss": 5.6718,
      "mean_token_accuracy": 0.17136583626270294,
      "num_tokens": 5765379.0,
      "step": 2445
    },
    {
      "entropy": 5.837702226638794,
      "epoch": 2.4951604686704023,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047062667087812007,
      "loss": 5.6926,
      "mean_token_accuracy": 0.15801315978169442,
      "num_tokens": 5777440.0,
      "step": 2450
    },
    {
      "entropy": 5.855507898330688,
      "epoch": 2.500254712175242,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004704281758684946,
      "loss": 5.67,
      "mean_token_accuracy": 0.16719134002923966,
      "num_tokens": 5789111.0,
      "step": 2455
    },
    {
      "entropy": 5.765543460845947,
      "epoch": 2.5053489556800814,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004702290595965686,
      "loss": 5.6529,
      "mean_token_accuracy": 0.16245744973421097,
      "num_tokens": 5800910.0,
      "step": 2460
    },
    {
      "entropy": 5.775801706314087,
      "epoch": 2.510443199184921,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004700293226953288,
      "loss": 5.6126,
      "mean_token_accuracy": 0.1708730563521385,
      "num_tokens": 5813125.0,
      "step": 2465
    },
    {
      "entropy": 5.654363536834717,
      "epoch": 2.5155374426897605,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004698289657997348,
      "loss": 5.5352,
      "mean_token_accuracy": 0.17443631142377852,
      "num_tokens": 5823461.0,
      "step": 2470
    },
    {
      "entropy": 5.779846668243408,
      "epoch": 2.5206316861946,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046962798954671727,
      "loss": 5.6619,
      "mean_token_accuracy": 0.1620674878358841,
      "num_tokens": 5834879.0,
      "step": 2475
    },
    {
      "entropy": 5.674750089645386,
      "epoch": 2.5257259296994397,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046942639457517565,
      "loss": 5.5619,
      "mean_token_accuracy": 0.1699425533413887,
      "num_tokens": 5845807.0,
      "step": 2480
    },
    {
      "entropy": 5.782552289962768,
      "epoch": 2.530820173204279,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004692241815259765,
      "loss": 5.7069,
      "mean_token_accuracy": 0.16113019436597825,
      "num_tokens": 5859722.0,
      "step": 2485
    },
    {
      "entropy": 5.773889112472534,
      "epoch": 2.5359144167091188,
      "grad_norm": 0.953125,
      "learning_rate": 0.000469021351041951,
      "loss": 5.5834,
      "mean_token_accuracy": 0.1676306575536728,
      "num_tokens": 5871108.0,
      "step": 2490
    },
    {
      "entropy": 5.802457666397094,
      "epoch": 2.5410086602139583,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004688179037678933,
      "loss": 5.6609,
      "mean_token_accuracy": 0.16495586633682252,
      "num_tokens": 5882759.0,
      "step": 2495
    },
    {
      "entropy": 5.788983678817749,
      "epoch": 2.546102903718798,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004686138403505582,
      "loss": 5.6743,
      "mean_token_accuracy": 0.16444195061922073,
      "num_tokens": 5894104.0,
      "step": 2500
    },
    {
      "epoch": 2.546102903718798,
      "eval_entropy": 5.498270017129403,
      "eval_loss": 5.916449546813965,
      "eval_mean_token_accuracy": 0.16029709029344866,
      "eval_num_tokens": 5894104.0,
      "eval_runtime": 2.6827,
      "eval_samples_per_second": 1207.752,
      "eval_steps_per_second": 150.969,
      "step": 2500
    },
    {
      "entropy": 5.636739015579224,
      "epoch": 2.5511971472236374,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046840916143865913,
      "loss": 5.5956,
      "mean_token_accuracy": 0.1692618429660797,
      "num_tokens": 5904869.0,
      "step": 2505
    },
    {
      "entropy": 5.7637975215911865,
      "epoch": 2.556291390728477,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004682038676828666,
      "loss": 5.5924,
      "mean_token_accuracy": 0.16839471310377122,
      "num_tokens": 5918463.0,
      "step": 2510
    },
    {
      "entropy": 5.75229287147522,
      "epoch": 2.561385634233316,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00046799795973580515,
      "loss": 5.6292,
      "mean_token_accuracy": 0.1668006271123886,
      "num_tokens": 5929645.0,
      "step": 2515
    },
    {
      "entropy": 5.8215991973876955,
      "epoch": 2.566479877738156,
      "grad_norm": 0.890625,
      "learning_rate": 0.00046779143825205217,
      "loss": 5.621,
      "mean_token_accuracy": 0.17295834869146348,
      "num_tokens": 5941516.0,
      "step": 2520
    },
    {
      "entropy": 5.744787502288818,
      "epoch": 2.5715741212429952,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004675843038881354,
      "loss": 5.6248,
      "mean_token_accuracy": 0.17074517160654068,
      "num_tokens": 5952482.0,
      "step": 2525
    },
    {
      "entropy": 5.642045879364014,
      "epoch": 2.576668364747835,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046737655730253076,
      "loss": 5.4886,
      "mean_token_accuracy": 0.1740100786089897,
      "num_tokens": 5963513.0,
      "step": 2530
    },
    {
      "entropy": 5.7616454601287845,
      "epoch": 2.5817626082526743,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046716819915566067,
      "loss": 5.5284,
      "mean_token_accuracy": 0.17381125539541245,
      "num_tokens": 5974254.0,
      "step": 2535
    },
    {
      "entropy": 5.66745548248291,
      "epoch": 2.586856851757514,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046695923010989145,
      "loss": 5.5274,
      "mean_token_accuracy": 0.1786910444498062,
      "num_tokens": 5985917.0,
      "step": 2540
    },
    {
      "entropy": 5.7848122119903564,
      "epoch": 2.5919510952623535,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046674965082953157,
      "loss": 5.6478,
      "mean_token_accuracy": 0.16492172032594682,
      "num_tokens": 5997316.0,
      "step": 2545
    },
    {
      "entropy": 5.711985158920288,
      "epoch": 2.597045338767193,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046653946198082943,
      "loss": 5.6931,
      "mean_token_accuracy": 0.1697121500968933,
      "num_tokens": 6010369.0,
      "step": 2550
    },
    {
      "entropy": 5.720060634613037,
      "epoch": 2.6021395822720326,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00046632866423197125,
      "loss": 5.5791,
      "mean_token_accuracy": 0.1741783171892166,
      "num_tokens": 6023568.0,
      "step": 2555
    },
    {
      "entropy": 5.731412744522094,
      "epoch": 2.607233825776872,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004661172582530788,
      "loss": 5.6073,
      "mean_token_accuracy": 0.16563851088285447,
      "num_tokens": 6035488.0,
      "step": 2560
    },
    {
      "entropy": 5.731129217147827,
      "epoch": 2.6123280692817117,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004659052447162076,
      "loss": 5.5885,
      "mean_token_accuracy": 0.17423500269651412,
      "num_tokens": 6047581.0,
      "step": 2565
    },
    {
      "entropy": 5.717588138580322,
      "epoch": 2.6174223127865512,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004656926242953445,
      "loss": 5.6462,
      "mean_token_accuracy": 0.1672428384423256,
      "num_tokens": 6060482.0,
      "step": 2570
    },
    {
      "entropy": 5.69965181350708,
      "epoch": 2.622516556291391,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004654793976664055,
      "loss": 5.6525,
      "mean_token_accuracy": 0.16912020295858382,
      "num_tokens": 6073051.0,
      "step": 2575
    },
    {
      "entropy": 5.760418891906738,
      "epoch": 2.6276107997962304,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00046526556550723395,
      "loss": 5.6735,
      "mean_token_accuracy": 0.16043009608983994,
      "num_tokens": 6084870.0,
      "step": 2580
    },
    {
      "entropy": 5.770459604263306,
      "epoch": 2.63270504330107,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046505112849759794,
      "loss": 5.6482,
      "mean_token_accuracy": 0.1581513911485672,
      "num_tokens": 6096305.0,
      "step": 2585
    },
    {
      "entropy": 5.724056577682495,
      "epoch": 2.6377992868059095,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046483608731918855,
      "loss": 5.5926,
      "mean_token_accuracy": 0.16795676797628403,
      "num_tokens": 6108504.0,
      "step": 2590
    },
    {
      "entropy": 5.6456928730010985,
      "epoch": 2.642893530310749,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00046462044265561735,
      "loss": 5.6014,
      "mean_token_accuracy": 0.16678556203842163,
      "num_tokens": 6120001.0,
      "step": 2595
    },
    {
      "entropy": 5.793216943740845,
      "epoch": 2.647987773815588,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046440419519241455,
      "loss": 5.6556,
      "mean_token_accuracy": 0.16507133692502976,
      "num_tokens": 6132059.0,
      "step": 2600
    },
    {
      "entropy": 5.7890002727508545,
      "epoch": 2.653082017320428,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004641873456170265,
      "loss": 5.6431,
      "mean_token_accuracy": 0.16495248973369597,
      "num_tokens": 6143082.0,
      "step": 2605
    },
    {
      "entropy": 5.6799695014953615,
      "epoch": 2.6581762608252673,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004639698946188137,
      "loss": 5.6202,
      "mean_token_accuracy": 0.16697055846452713,
      "num_tokens": 6155307.0,
      "step": 2610
    },
    {
      "entropy": 5.816330814361573,
      "epoch": 2.663270504330107,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004637518428890486,
      "loss": 5.6481,
      "mean_token_accuracy": 0.16570667177438736,
      "num_tokens": 6166623.0,
      "step": 2615
    },
    {
      "entropy": 5.784613418579101,
      "epoch": 2.6683647478349464,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004635331911209133,
      "loss": 5.6115,
      "mean_token_accuracy": 0.16887350976467133,
      "num_tokens": 6177606.0,
      "step": 2620
    },
    {
      "entropy": 5.6948333263397215,
      "epoch": 2.673458991339786,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004633139400094974,
      "loss": 5.5747,
      "mean_token_accuracy": 0.1651701346039772,
      "num_tokens": 6189617.0,
      "step": 2625
    },
    {
      "entropy": 5.746786403656006,
      "epoch": 2.6785532348446255,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000463094090251796,
      "loss": 5.6235,
      "mean_token_accuracy": 0.16294585168361664,
      "num_tokens": 6201273.0,
      "step": 2630
    },
    {
      "entropy": 5.71042833328247,
      "epoch": 2.683647478349465,
      "grad_norm": 1.0,
      "learning_rate": 0.00046287364254670696,
      "loss": 5.4985,
      "mean_token_accuracy": 0.18142603188753129,
      "num_tokens": 6211281.0,
      "step": 2635
    },
    {
      "entropy": 5.693270111083985,
      "epoch": 2.6887417218543046,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004626525975950292,
      "loss": 5.6167,
      "mean_token_accuracy": 0.16980593502521515,
      "num_tokens": 6221520.0,
      "step": 2640
    },
    {
      "entropy": 5.746152925491333,
      "epoch": 2.693835965359144,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004624309560994603,
      "loss": 5.5793,
      "mean_token_accuracy": 0.16716546863317489,
      "num_tokens": 6234517.0,
      "step": 2645
    },
    {
      "entropy": 5.736660766601562,
      "epoch": 2.6989302088639837,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004622087187645942,
      "loss": 5.5796,
      "mean_token_accuracy": 0.17094964385032654,
      "num_tokens": 6245770.0,
      "step": 2650
    },
    {
      "entropy": 5.722123718261718,
      "epoch": 2.7040244523688233,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004619858862969188,
      "loss": 5.6211,
      "mean_token_accuracy": 0.17120955884456635,
      "num_tokens": 6257756.0,
      "step": 2655
    },
    {
      "entropy": 5.6861255168914795,
      "epoch": 2.709118695873663,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004617624594048143,
      "loss": 5.592,
      "mean_token_accuracy": 0.17201131731271743,
      "num_tokens": 6269292.0,
      "step": 2660
    },
    {
      "entropy": 5.559902048110962,
      "epoch": 2.7142129393785024,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004615384387985504,
      "loss": 5.5509,
      "mean_token_accuracy": 0.17285647690296174,
      "num_tokens": 6281892.0,
      "step": 2665
    },
    {
      "entropy": 5.789300012588501,
      "epoch": 2.719307182883342,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000461313825190284,
      "loss": 5.6272,
      "mean_token_accuracy": 0.16576700359582902,
      "num_tokens": 6294134.0,
      "step": 2670
    },
    {
      "entropy": 5.733413648605347,
      "epoch": 2.7244014263881815,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004610886192940574,
      "loss": 5.6071,
      "mean_token_accuracy": 0.16636348962783815,
      "num_tokens": 6306475.0,
      "step": 2675
    },
    {
      "entropy": 5.742194271087646,
      "epoch": 2.729495669893021,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00046086282182579566,
      "loss": 5.5889,
      "mean_token_accuracy": 0.16803767532110214,
      "num_tokens": 6319025.0,
      "step": 2680
    },
    {
      "entropy": 5.656324434280395,
      "epoch": 2.73458991339786,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004606364335033045,
      "loss": 5.624,
      "mean_token_accuracy": 0.1758906736969948,
      "num_tokens": 6330968.0,
      "step": 2685
    },
    {
      "entropy": 5.752261018753051,
      "epoch": 2.7396841569027,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046040945504626787,
      "loss": 5.6368,
      "mean_token_accuracy": 0.16836125254631043,
      "num_tokens": 6344029.0,
      "step": 2690
    },
    {
      "entropy": 5.74548602104187,
      "epoch": 2.7447784004075393,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046018188717624576,
      "loss": 5.6695,
      "mean_token_accuracy": 0.1657738447189331,
      "num_tokens": 6356506.0,
      "step": 2695
    },
    {
      "entropy": 5.682445049285889,
      "epoch": 2.7498726439123793,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000459953730616672,
      "loss": 5.5981,
      "mean_token_accuracy": 0.17355468422174453,
      "num_tokens": 6367910.0,
      "step": 2700
    },
    {
      "entropy": 5.712604331970215,
      "epoch": 2.7549668874172184,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045972498609285175,
      "loss": 5.5375,
      "mean_token_accuracy": 0.17217673659324645,
      "num_tokens": 6378009.0,
      "step": 2705
    },
    {
      "entropy": 5.661414241790771,
      "epoch": 2.760061130922058,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00045949565433195925,
      "loss": 5.5836,
      "mean_token_accuracy": 0.1669108122587204,
      "num_tokens": 6389836.0,
      "step": 2710
    },
    {
      "entropy": 5.764407730102539,
      "epoch": 2.7651553744268975,
      "grad_norm": 1.015625,
      "learning_rate": 0.00045926573606303576,
      "loss": 5.5958,
      "mean_token_accuracy": 0.1682405263185501,
      "num_tokens": 6401499.0,
      "step": 2715
    },
    {
      "entropy": 5.764873743057251,
      "epoch": 2.770249617931737,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045903523201698684,
      "loss": 5.6563,
      "mean_token_accuracy": 0.16478636264801025,
      "num_tokens": 6413738.0,
      "step": 2720
    },
    {
      "entropy": 5.753376054763794,
      "epoch": 2.7753438614365766,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004588041429265803,
      "loss": 5.673,
      "mean_token_accuracy": 0.1616167888045311,
      "num_tokens": 6425810.0,
      "step": 2725
    },
    {
      "entropy": 5.765153551101685,
      "epoch": 2.780438104941416,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004585724695264436,
      "loss": 5.6706,
      "mean_token_accuracy": 0.16697462201118468,
      "num_tokens": 6435732.0,
      "step": 2730
    },
    {
      "entropy": 5.769167280197143,
      "epoch": 2.7855323484462557,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045834021255306223,
      "loss": 5.6737,
      "mean_token_accuracy": 0.16523695588111878,
      "num_tokens": 6447233.0,
      "step": 2735
    },
    {
      "entropy": 5.650635671615601,
      "epoch": 2.7906265919510953,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004581073727447762,
      "loss": 5.564,
      "mean_token_accuracy": 0.17355237603187562,
      "num_tokens": 6459150.0,
      "step": 2740
    },
    {
      "entropy": 5.7149739265441895,
      "epoch": 2.795720835455935,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004578739508417788,
      "loss": 5.6784,
      "mean_token_accuracy": 0.16392523050308228,
      "num_tokens": 6471893.0,
      "step": 2745
    },
    {
      "entropy": 5.78902759552002,
      "epoch": 2.8008150789607744,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004576399475861137,
      "loss": 5.6207,
      "mean_token_accuracy": 0.16507020592689514,
      "num_tokens": 6484039.0,
      "step": 2750
    },
    {
      "entropy": 5.675820684432983,
      "epoch": 2.805909322465614,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004574053637216724,
      "loss": 5.6656,
      "mean_token_accuracy": 0.16735105216503143,
      "num_tokens": 6495052.0,
      "step": 2755
    },
    {
      "entropy": 5.760396480560303,
      "epoch": 2.8110035659704535,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004571701999941926,
      "loss": 5.5839,
      "mean_token_accuracy": 0.1688750609755516,
      "num_tokens": 6506047.0,
      "step": 2760
    },
    {
      "entropy": 5.628135013580322,
      "epoch": 2.816097809475293,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000456934457151255,
      "loss": 5.5124,
      "mean_token_accuracy": 0.17793091088533403,
      "num_tokens": 6517591.0,
      "step": 2765
    },
    {
      "entropy": 5.744438505172729,
      "epoch": 2.821192052980132,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045669813594228137,
      "loss": 5.5389,
      "mean_token_accuracy": 0.1804353728890419,
      "num_tokens": 6528459.0,
      "step": 2770
    },
    {
      "entropy": 5.614086103439331,
      "epoch": 2.826286296484972,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045646123711853235,
      "loss": 5.5842,
      "mean_token_accuracy": 0.1740228310227394,
      "num_tokens": 6539027.0,
      "step": 2775
    },
    {
      "entropy": 5.5858015537261965,
      "epoch": 2.8313805399898113,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045622376143310435,
      "loss": 5.495,
      "mean_token_accuracy": 0.18643426448106765,
      "num_tokens": 6551188.0,
      "step": 2780
    },
    {
      "entropy": 5.595398473739624,
      "epoch": 2.8364747834946513,
      "grad_norm": 0.984375,
      "learning_rate": 0.00045598570964092817,
      "loss": 5.558,
      "mean_token_accuracy": 0.17872982174158097,
      "num_tokens": 6562759.0,
      "step": 2785
    },
    {
      "entropy": 5.7415930271148685,
      "epoch": 2.8415690269994904,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045574708249876547,
      "loss": 5.5538,
      "mean_token_accuracy": 0.16583310812711716,
      "num_tokens": 6574350.0,
      "step": 2790
    },
    {
      "entropy": 5.680415773391724,
      "epoch": 2.84666327050433,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045550788076520736,
      "loss": 5.5331,
      "mean_token_accuracy": 0.17263849675655366,
      "num_tokens": 6585292.0,
      "step": 2795
    },
    {
      "entropy": 5.710312271118164,
      "epoch": 2.8517575140091695,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004552681052006713,
      "loss": 5.6097,
      "mean_token_accuracy": 0.16250473111867905,
      "num_tokens": 6598143.0,
      "step": 2800
    },
    {
      "entropy": 5.657690525054932,
      "epoch": 2.856851757514009,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004550277565673992,
      "loss": 5.491,
      "mean_token_accuracy": 0.17814151346683502,
      "num_tokens": 6609284.0,
      "step": 2805
    },
    {
      "entropy": 5.646687078475952,
      "epoch": 2.8619460010188487,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045478683562945457,
      "loss": 5.5079,
      "mean_token_accuracy": 0.18061336725950242,
      "num_tokens": 6621215.0,
      "step": 2810
    },
    {
      "entropy": 5.691900205612183,
      "epoch": 2.867040244523688,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004545453431527202,
      "loss": 5.6462,
      "mean_token_accuracy": 0.16735012084245682,
      "num_tokens": 6633522.0,
      "step": 2815
    },
    {
      "entropy": 5.645006275177002,
      "epoch": 2.8721344880285278,
      "grad_norm": 1.046875,
      "learning_rate": 0.000454303279904896,
      "loss": 5.5361,
      "mean_token_accuracy": 0.169601172208786,
      "num_tokens": 6645645.0,
      "step": 2820
    },
    {
      "entropy": 5.67996826171875,
      "epoch": 2.8772287315333673,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045406064665549624,
      "loss": 5.5093,
      "mean_token_accuracy": 0.17295902371406555,
      "num_tokens": 6656329.0,
      "step": 2825
    },
    {
      "entropy": 5.646455383300781,
      "epoch": 2.882322975038207,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004538174441758471,
      "loss": 5.5571,
      "mean_token_accuracy": 0.17819010764360427,
      "num_tokens": 6666816.0,
      "step": 2830
    },
    {
      "entropy": 5.733512306213379,
      "epoch": 2.8874172185430464,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045357367323908463,
      "loss": 5.6371,
      "mean_token_accuracy": 0.17107812613248824,
      "num_tokens": 6678948.0,
      "step": 2835
    },
    {
      "entropy": 5.696351766586304,
      "epoch": 2.892511462047886,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045332933462015156,
      "loss": 5.6164,
      "mean_token_accuracy": 0.16568383127450942,
      "num_tokens": 6690272.0,
      "step": 2840
    },
    {
      "entropy": 5.740844058990478,
      "epoch": 2.8976057055527256,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004530844290957957,
      "loss": 5.6669,
      "mean_token_accuracy": 0.17213962376117706,
      "num_tokens": 6701612.0,
      "step": 2845
    },
    {
      "entropy": 5.736670446395874,
      "epoch": 2.902699949057565,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045283895744456694,
      "loss": 5.5405,
      "mean_token_accuracy": 0.17544633895158768,
      "num_tokens": 6713883.0,
      "step": 2850
    },
    {
      "entropy": 5.713712310791015,
      "epoch": 2.9077941925624042,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045259292044681456,
      "loss": 5.5988,
      "mean_token_accuracy": 0.17053105384111406,
      "num_tokens": 6725378.0,
      "step": 2855
    },
    {
      "entropy": 5.659952163696289,
      "epoch": 2.9128884360672442,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004523463188846855,
      "loss": 5.5881,
      "mean_token_accuracy": 0.17058228105306625,
      "num_tokens": 6737202.0,
      "step": 2860
    },
    {
      "entropy": 5.720408821105957,
      "epoch": 2.9179826795720833,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004520991535421211,
      "loss": 5.6035,
      "mean_token_accuracy": 0.16732127517461776,
      "num_tokens": 6749874.0,
      "step": 2865
    },
    {
      "entropy": 5.668585348129272,
      "epoch": 2.9230769230769234,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004518514252048552,
      "loss": 5.5438,
      "mean_token_accuracy": 0.1763650208711624,
      "num_tokens": 6761168.0,
      "step": 2870
    },
    {
      "entropy": 5.734693336486816,
      "epoch": 2.9281711665817625,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004516031346604112,
      "loss": 5.5656,
      "mean_token_accuracy": 0.16883667260408403,
      "num_tokens": 6772622.0,
      "step": 2875
    },
    {
      "entropy": 5.605511283874511,
      "epoch": 2.933265410086602,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004513542826980999,
      "loss": 5.4823,
      "mean_token_accuracy": 0.16999128162860871,
      "num_tokens": 6784854.0,
      "step": 2880
    },
    {
      "entropy": 5.701577281951904,
      "epoch": 2.9383596535914416,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004511048701090167,
      "loss": 5.5436,
      "mean_token_accuracy": 0.17507170438766478,
      "num_tokens": 6796079.0,
      "step": 2885
    },
    {
      "entropy": 5.637689161300659,
      "epoch": 2.943453897096281,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004508548976860392,
      "loss": 5.5843,
      "mean_token_accuracy": 0.16997636407613753,
      "num_tokens": 6808270.0,
      "step": 2890
    },
    {
      "entropy": 5.6968488693237305,
      "epoch": 2.9485481406011207,
      "grad_norm": 1.015625,
      "learning_rate": 0.000450604366223825,
      "loss": 5.5834,
      "mean_token_accuracy": 0.1691394954919815,
      "num_tokens": 6820829.0,
      "step": 2895
    },
    {
      "entropy": 5.726992988586426,
      "epoch": 2.9536423841059603,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00045035327651880846,
      "loss": 5.5721,
      "mean_token_accuracy": 0.16805260479450226,
      "num_tokens": 6832891.0,
      "step": 2900
    },
    {
      "entropy": 5.741576147079468,
      "epoch": 2.9587366276108,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045010162936919896,
      "loss": 5.5669,
      "mean_token_accuracy": 0.17039521634578705,
      "num_tokens": 6846624.0,
      "step": 2905
    },
    {
      "entropy": 5.679950284957886,
      "epoch": 2.9638308711156394,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004498494255749776,
      "loss": 5.6126,
      "mean_token_accuracy": 0.16650597602128983,
      "num_tokens": 6859577.0,
      "step": 2910
    },
    {
      "entropy": 5.541244840621948,
      "epoch": 2.968925114620479,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004495966659378954,
      "loss": 5.5123,
      "mean_token_accuracy": 0.17585458159446715,
      "num_tokens": 6871276.0,
      "step": 2915
    },
    {
      "entropy": 5.697320127487183,
      "epoch": 2.9740193581253185,
      "grad_norm": 1.03125,
      "learning_rate": 0.00044934335126147016,
      "loss": 5.5624,
      "mean_token_accuracy": 0.17635687291622162,
      "num_tokens": 6882341.0,
      "step": 2920
    },
    {
      "entropy": 5.686966323852539,
      "epoch": 2.979113601630158,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004490894823509843,
      "loss": 5.5436,
      "mean_token_accuracy": 0.1755341351032257,
      "num_tokens": 6893507.0,
      "step": 2925
    },
    {
      "entropy": 5.704105472564697,
      "epoch": 2.9842078451349976,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004488350600134819,
      "loss": 5.5332,
      "mean_token_accuracy": 0.17578678876161574,
      "num_tokens": 6905313.0,
      "step": 2930
    },
    {
      "entropy": 5.652353668212891,
      "epoch": 2.989302088639837,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004485800850577667,
      "loss": 5.585,
      "mean_token_accuracy": 0.17711285054683684,
      "num_tokens": 6917280.0,
      "step": 2935
    },
    {
      "entropy": 5.67321662902832,
      "epoch": 2.9943963321446763,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004483245582943989,
      "loss": 5.4955,
      "mean_token_accuracy": 0.17702803909778594,
      "num_tokens": 6929790.0,
      "step": 2940
    },
    {
      "entropy": 5.609502744674683,
      "epoch": 2.9994905756495163,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044806848053569314,
      "loss": 5.5389,
      "mean_token_accuracy": 0.17498539835214616,
      "num_tokens": 6941361.0,
      "step": 2945
    },
    {
      "entropy": 5.632062647077772,
      "epoch": 3.0040753948038716,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044781185259571544,
      "loss": 5.2869,
      "mean_token_accuracy": 0.18558117085032994,
      "num_tokens": 6951362.0,
      "step": 2950
    },
    {
      "entropy": 5.613099241256714,
      "epoch": 3.009169638308711,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004475546752902811,
      "loss": 5.2281,
      "mean_token_accuracy": 0.1877656176686287,
      "num_tokens": 6962089.0,
      "step": 2955
    },
    {
      "entropy": 5.614435482025146,
      "epoch": 3.0142638818135508,
      "grad_norm": 0.984375,
      "learning_rate": 0.00044729694943695155,
      "loss": 5.176,
      "mean_token_accuracy": 0.18906930238008499,
      "num_tokens": 6974049.0,
      "step": 2960
    },
    {
      "entropy": 5.6351393699646,
      "epoch": 3.0193581253183903,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004470386758550321,
      "loss": 5.1977,
      "mean_token_accuracy": 0.19047407507896424,
      "num_tokens": 6985401.0,
      "step": 2965
    },
    {
      "entropy": 5.613850688934326,
      "epoch": 3.02445236882323,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004467798553655695,
      "loss": 5.2465,
      "mean_token_accuracy": 0.18515696972608567,
      "num_tokens": 6996771.0,
      "step": 2970
    },
    {
      "entropy": 5.581202411651612,
      "epoch": 3.0295466123280694,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000446520488791349,
      "loss": 5.2419,
      "mean_token_accuracy": 0.19033248275518416,
      "num_tokens": 7007945.0,
      "step": 2975
    },
    {
      "entropy": 5.658330583572388,
      "epoch": 3.034640855832909,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004462605769568915,
      "loss": 5.2055,
      "mean_token_accuracy": 0.19026146531105043,
      "num_tokens": 7020309.0,
      "step": 2980
    },
    {
      "entropy": 5.56205382347107,
      "epoch": 3.0397350993377485,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004460001206884519,
      "loss": 5.2496,
      "mean_token_accuracy": 0.1896470904350281,
      "num_tokens": 7032040.0,
      "step": 2985
    },
    {
      "entropy": 5.66866455078125,
      "epoch": 3.0448293428425877,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004457391208140155,
      "loss": 5.2875,
      "mean_token_accuracy": 0.1815323680639267,
      "num_tokens": 7043552.0,
      "step": 2990
    },
    {
      "entropy": 5.547569131851196,
      "epoch": 3.049923586347427,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004454775781632955,
      "loss": 5.2683,
      "mean_token_accuracy": 0.18844338357448578,
      "num_tokens": 7055301.0,
      "step": 2995
    },
    {
      "entropy": 5.601464557647705,
      "epoch": 3.0550178298522668,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044521549356773096,
      "loss": 5.1743,
      "mean_token_accuracy": 0.19403870403766632,
      "num_tokens": 7066015.0,
      "step": 3000
    },
    {
      "epoch": 3.0550178298522668,
      "eval_entropy": 5.440786188620108,
      "eval_loss": 5.829931259155273,
      "eval_mean_token_accuracy": 0.16714958874163804,
      "eval_num_tokens": 7066015.0,
      "eval_runtime": 2.8883,
      "eval_samples_per_second": 1121.785,
      "eval_steps_per_second": 140.223,
      "step": 3000
    },
    {
      "entropy": 5.614556694030762,
      "epoch": 3.0601120733571063,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004449528678604836,
      "loss": 5.2799,
      "mean_token_accuracy": 0.18113382607698442,
      "num_tokens": 7075982.0,
      "step": 3005
    },
    {
      "entropy": 5.619668292999267,
      "epoch": 3.065206316861946,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004446897018764353,
      "loss": 5.2448,
      "mean_token_accuracy": 0.18539075404405594,
      "num_tokens": 7088245.0,
      "step": 3010
    },
    {
      "entropy": 5.578803205490113,
      "epoch": 3.0703005603667854,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044442599645218566,
      "loss": 5.1528,
      "mean_token_accuracy": 0.19120035618543624,
      "num_tokens": 7098543.0,
      "step": 3015
    },
    {
      "entropy": 5.559082937240601,
      "epoch": 3.075394803871625,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044416175242604887,
      "loss": 5.2385,
      "mean_token_accuracy": 0.18284711390733718,
      "num_tokens": 7110872.0,
      "step": 3020
    },
    {
      "entropy": 5.628260612487793,
      "epoch": 3.0804890473764646,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004438969706380514,
      "loss": 5.2694,
      "mean_token_accuracy": 0.18073297739028932,
      "num_tokens": 7123123.0,
      "step": 3025
    },
    {
      "entropy": 5.6057638168334964,
      "epoch": 3.085583290881304,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044363165192992947,
      "loss": 5.3441,
      "mean_token_accuracy": 0.17324751168489455,
      "num_tokens": 7135051.0,
      "step": 3030
    },
    {
      "entropy": 5.556230068206787,
      "epoch": 3.0906775343861437,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004433657971451259,
      "loss": 5.243,
      "mean_token_accuracy": 0.19202030599117278,
      "num_tokens": 7147272.0,
      "step": 3035
    },
    {
      "entropy": 5.567021179199219,
      "epoch": 3.0957717778909832,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044309940712878774,
      "loss": 5.3377,
      "mean_token_accuracy": 0.18122091591358186,
      "num_tokens": 7159530.0,
      "step": 3040
    },
    {
      "entropy": 5.567464399337768,
      "epoch": 3.100866021395823,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004428324827277637,
      "loss": 5.1435,
      "mean_token_accuracy": 0.19043394029140473,
      "num_tokens": 7171004.0,
      "step": 3045
    },
    {
      "entropy": 5.457118988037109,
      "epoch": 3.1059602649006623,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000442565024790601,
      "loss": 5.1639,
      "mean_token_accuracy": 0.19438861608505248,
      "num_tokens": 7182177.0,
      "step": 3050
    },
    {
      "entropy": 5.619878053665161,
      "epoch": 3.111054508405502,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044229703416754333,
      "loss": 5.2345,
      "mean_token_accuracy": 0.18157973140478134,
      "num_tokens": 7193275.0,
      "step": 3055
    },
    {
      "entropy": 5.573149633407593,
      "epoch": 3.1161487519103415,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004420285117105275,
      "loss": 5.1904,
      "mean_token_accuracy": 0.1889777660369873,
      "num_tokens": 7203865.0,
      "step": 3060
    },
    {
      "entropy": 5.5709387302398685,
      "epoch": 3.121242995415181,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044175945827318097,
      "loss": 5.3606,
      "mean_token_accuracy": 0.18194379210472106,
      "num_tokens": 7216119.0,
      "step": 3065
    },
    {
      "entropy": 5.63159589767456,
      "epoch": 3.1263372389200206,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044148987471081933,
      "loss": 5.3227,
      "mean_token_accuracy": 0.18258614838123322,
      "num_tokens": 7227669.0,
      "step": 3070
    },
    {
      "entropy": 5.524097681045532,
      "epoch": 3.13143148242486,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044121976188044357,
      "loss": 5.2112,
      "mean_token_accuracy": 0.19257961809635163,
      "num_tokens": 7239609.0,
      "step": 3075
    },
    {
      "entropy": 5.563867282867432,
      "epoch": 3.1365257259296992,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004409491206407367,
      "loss": 5.1695,
      "mean_token_accuracy": 0.2025545358657837,
      "num_tokens": 7250716.0,
      "step": 3080
    },
    {
      "entropy": 5.514932298660279,
      "epoch": 3.141619969434539,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004406779518520622,
      "loss": 5.1607,
      "mean_token_accuracy": 0.19224747121334076,
      "num_tokens": 7261432.0,
      "step": 3085
    },
    {
      "entropy": 5.4803986072540285,
      "epoch": 3.1467142129393784,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044040625637646006,
      "loss": 5.2667,
      "mean_token_accuracy": 0.18231366127729415,
      "num_tokens": 7274594.0,
      "step": 3090
    },
    {
      "entropy": 5.639348459243775,
      "epoch": 3.151808456444218,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004401340350776449,
      "loss": 5.3031,
      "mean_token_accuracy": 0.18331949710845946,
      "num_tokens": 7285882.0,
      "step": 3095
    },
    {
      "entropy": 5.454601144790649,
      "epoch": 3.1569026999490575,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043986128882100286,
      "loss": 5.1068,
      "mean_token_accuracy": 0.20472786873579024,
      "num_tokens": 7296682.0,
      "step": 3100
    },
    {
      "entropy": 5.472896385192871,
      "epoch": 3.161996943453897,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00043958801847358874,
      "loss": 5.2693,
      "mean_token_accuracy": 0.18683293610811233,
      "num_tokens": 7309402.0,
      "step": 3105
    },
    {
      "entropy": 5.596922397613525,
      "epoch": 3.1670911869587366,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00043931422490412376,
      "loss": 5.2016,
      "mean_token_accuracy": 0.19378906786441802,
      "num_tokens": 7321169.0,
      "step": 3110
    },
    {
      "entropy": 5.518866443634034,
      "epoch": 3.172185430463576,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043903990898299213,
      "loss": 5.2162,
      "mean_token_accuracy": 0.18622226268053055,
      "num_tokens": 7331913.0,
      "step": 3115
    },
    {
      "entropy": 5.599637365341186,
      "epoch": 3.1772796739684157,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00043876507158223865,
      "loss": 5.2596,
      "mean_token_accuracy": 0.18324704915285112,
      "num_tokens": 7343477.0,
      "step": 3120
    },
    {
      "entropy": 5.6364775657653805,
      "epoch": 3.1823739174732553,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004384897135755663,
      "loss": 5.4032,
      "mean_token_accuracy": 0.1882349207997322,
      "num_tokens": 7357385.0,
      "step": 3125
    },
    {
      "entropy": 5.52266001701355,
      "epoch": 3.187468160978095,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00043821383583833237,
      "loss": 5.2232,
      "mean_token_accuracy": 0.1869337350130081,
      "num_tokens": 7369426.0,
      "step": 3130
    },
    {
      "entropy": 5.465065860748291,
      "epoch": 3.1925624044829344,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004379374392475468,
      "loss": 5.1309,
      "mean_token_accuracy": 0.20164910405874253,
      "num_tokens": 7382952.0,
      "step": 3135
    },
    {
      "entropy": 5.526530361175537,
      "epoch": 3.197656647987774,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004376605246818691,
      "loss": 5.3127,
      "mean_token_accuracy": 0.18805583715438842,
      "num_tokens": 7395133.0,
      "step": 3140
    },
    {
      "entropy": 5.5906510829925535,
      "epoch": 3.2027508914926135,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000437383093021605,
      "loss": 5.2216,
      "mean_token_accuracy": 0.19437444359064102,
      "num_tokens": 7406029.0,
      "step": 3145
    },
    {
      "entropy": 5.545298767089844,
      "epoch": 3.207845134997453,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004371051451487042,
      "loss": 5.2317,
      "mean_token_accuracy": 0.18914845883846282,
      "num_tokens": 7417901.0,
      "step": 3150
    },
    {
      "entropy": 5.498559761047363,
      "epoch": 3.2129393785022926,
      "grad_norm": 1.03125,
      "learning_rate": 0.00043682668194675766,
      "loss": 5.2367,
      "mean_token_accuracy": 0.1817171424627304,
      "num_tokens": 7430445.0,
      "step": 3155
    },
    {
      "entropy": 5.567168140411377,
      "epoch": 3.218033622007132,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004365477043009943,
      "loss": 5.3247,
      "mean_token_accuracy": 0.17993049919605256,
      "num_tokens": 7441483.0,
      "step": 3160
    },
    {
      "entropy": 5.61783652305603,
      "epoch": 3.2231278655119713,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00043626821309827846,
      "loss": 5.2812,
      "mean_token_accuracy": 0.1795026659965515,
      "num_tokens": 7453561.0,
      "step": 3165
    },
    {
      "entropy": 5.574274587631225,
      "epoch": 3.228222109016811,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043598820922710727,
      "loss": 5.3181,
      "mean_token_accuracy": 0.17989187985658645,
      "num_tokens": 7465236.0,
      "step": 3170
    },
    {
      "entropy": 5.522154140472412,
      "epoch": 3.2333163525216504,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043570769357760737,
      "loss": 5.2163,
      "mean_token_accuracy": 0.1942819282412529,
      "num_tokens": 7475985.0,
      "step": 3175
    },
    {
      "entropy": 5.4718786716461185,
      "epoch": 3.23841059602649,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043542666704153257,
      "loss": 5.239,
      "mean_token_accuracy": 0.19178879261016846,
      "num_tokens": 7487833.0,
      "step": 3180
    },
    {
      "entropy": 5.592524147033691,
      "epoch": 3.2435048395313295,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004351451305122605,
      "loss": 5.284,
      "mean_token_accuracy": 0.1928950160741806,
      "num_tokens": 7499506.0,
      "step": 3185
    },
    {
      "entropy": 5.5970710754394535,
      "epoch": 3.248599083036169,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043486308488479045,
      "loss": 5.3216,
      "mean_token_accuracy": 0.16977053880691528,
      "num_tokens": 7511727.0,
      "step": 3190
    },
    {
      "entropy": 5.560200214385986,
      "epoch": 3.2536933265410086,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004345805310557397,
      "loss": 5.3532,
      "mean_token_accuracy": 0.183087357878685,
      "num_tokens": 7524505.0,
      "step": 3195
    },
    {
      "entropy": 5.58688440322876,
      "epoch": 3.258787570045848,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004342974699233412,
      "loss": 5.3564,
      "mean_token_accuracy": 0.17731336504220963,
      "num_tokens": 7536685.0,
      "step": 3200
    },
    {
      "entropy": 5.648311614990234,
      "epoch": 3.2638818135506877,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043401390238744094,
      "loss": 5.2127,
      "mean_token_accuracy": 0.19691305458545685,
      "num_tokens": 7547296.0,
      "step": 3205
    },
    {
      "entropy": 5.495570468902588,
      "epoch": 3.2689760570555273,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004337298293494943,
      "loss": 5.2596,
      "mean_token_accuracy": 0.18756755888462068,
      "num_tokens": 7559511.0,
      "step": 3210
    },
    {
      "entropy": 5.544582653045654,
      "epoch": 3.274070300560367,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043344525171256395,
      "loss": 5.2977,
      "mean_token_accuracy": 0.1879246026277542,
      "num_tokens": 7570881.0,
      "step": 3215
    },
    {
      "entropy": 5.577976131439209,
      "epoch": 3.2791645440652064,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004331601703813165,
      "loss": 5.2196,
      "mean_token_accuracy": 0.18650293946266175,
      "num_tokens": 7582723.0,
      "step": 3220
    },
    {
      "entropy": 5.485066366195679,
      "epoch": 3.284258787570046,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004328745862620198,
      "loss": 5.3482,
      "mean_token_accuracy": 0.18082359284162522,
      "num_tokens": 7594375.0,
      "step": 3225
    },
    {
      "entropy": 5.622646427154541,
      "epoch": 3.2893530310748855,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004325885002625403,
      "loss": 5.233,
      "mean_token_accuracy": 0.18945698887109758,
      "num_tokens": 7606481.0,
      "step": 3230
    },
    {
      "entropy": 5.555077838897705,
      "epoch": 3.294447274579725,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004323019132923396,
      "loss": 5.3339,
      "mean_token_accuracy": 0.18236346542835236,
      "num_tokens": 7618495.0,
      "step": 3235
    },
    {
      "entropy": 5.564655447006226,
      "epoch": 3.299541518084564,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004320148262624719,
      "loss": 5.267,
      "mean_token_accuracy": 0.18518765568733214,
      "num_tokens": 7630149.0,
      "step": 3240
    },
    {
      "entropy": 5.598733758926391,
      "epoch": 3.304635761589404,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004317272400855814,
      "loss": 5.2643,
      "mean_token_accuracy": 0.18013916909694672,
      "num_tokens": 7641580.0,
      "step": 3245
    },
    {
      "entropy": 5.499423837661743,
      "epoch": 3.3097300050942433,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043143915567589875,
      "loss": 5.267,
      "mean_token_accuracy": 0.18198563307523727,
      "num_tokens": 7654231.0,
      "step": 3250
    },
    {
      "entropy": 5.512173175811768,
      "epoch": 3.314824248599083,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004311505739492385,
      "loss": 5.2139,
      "mean_token_accuracy": 0.18966390788555146,
      "num_tokens": 7665979.0,
      "step": 3255
    },
    {
      "entropy": 5.487508821487427,
      "epoch": 3.3199184921039224,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043086149582299637,
      "loss": 5.2246,
      "mean_token_accuracy": 0.19439356327056884,
      "num_tokens": 7678072.0,
      "step": 3260
    },
    {
      "entropy": 5.5803870677948,
      "epoch": 3.325012735608762,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00043057192221614604,
      "loss": 5.2933,
      "mean_token_accuracy": 0.18085529059171676,
      "num_tokens": 7690927.0,
      "step": 3265
    },
    {
      "entropy": 5.472428989410401,
      "epoch": 3.3301069791136015,
      "grad_norm": 1.015625,
      "learning_rate": 0.00043028185404923614,
      "loss": 5.2968,
      "mean_token_accuracy": 0.1850164070725441,
      "num_tokens": 7702502.0,
      "step": 3270
    },
    {
      "entropy": 5.489730167388916,
      "epoch": 3.335201222618441,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004299912922443877,
      "loss": 5.2778,
      "mean_token_accuracy": 0.19059146493673323,
      "num_tokens": 7714676.0,
      "step": 3275
    },
    {
      "entropy": 5.542190837860107,
      "epoch": 3.3402954661232807,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004297002377252908,
      "loss": 5.2665,
      "mean_token_accuracy": 0.17979499101638793,
      "num_tokens": 7725664.0,
      "step": 3280
    },
    {
      "entropy": 5.493380689620972,
      "epoch": 3.34538970962812,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042940869141720224,
      "loss": 5.1951,
      "mean_token_accuracy": 0.19204748570919036,
      "num_tokens": 7737494.0,
      "step": 3285
    },
    {
      "entropy": 5.5412290573120115,
      "epoch": 3.3504839531329598,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004291166542469418,
      "loss": 5.2187,
      "mean_token_accuracy": 0.19737294912338257,
      "num_tokens": 7749314.0,
      "step": 3290
    },
    {
      "entropy": 5.493816995620728,
      "epoch": 3.3555781966377993,
      "grad_norm": 1.234375,
      "learning_rate": 0.00042882412714288987,
      "loss": 5.2828,
      "mean_token_accuracy": 0.18764964640140533,
      "num_tokens": 7760808.0,
      "step": 3295
    },
    {
      "entropy": 5.483426523208618,
      "epoch": 3.360672440142639,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042853111103498433,
      "loss": 5.2588,
      "mean_token_accuracy": 0.18581708073616027,
      "num_tokens": 7772056.0,
      "step": 3300
    },
    {
      "entropy": 5.541358566284179,
      "epoch": 3.3657666836474784,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042823760685471756,
      "loss": 5.3097,
      "mean_token_accuracy": 0.18453435748815536,
      "num_tokens": 7784862.0,
      "step": 3305
    },
    {
      "entropy": 5.6184193134307865,
      "epoch": 3.370860927152318,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004279436155351336,
      "loss": 5.3535,
      "mean_token_accuracy": 0.1799679085612297,
      "num_tokens": 7797090.0,
      "step": 3310
    },
    {
      "entropy": 5.496599102020264,
      "epoch": 3.3759551706571576,
      "grad_norm": 1.125,
      "learning_rate": 0.0004276491380108251,
      "loss": 5.3177,
      "mean_token_accuracy": 0.1821472480893135,
      "num_tokens": 7810352.0,
      "step": 3315
    },
    {
      "entropy": 5.480373573303223,
      "epoch": 3.381049414161997,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004273541752179301,
      "loss": 5.1528,
      "mean_token_accuracy": 0.196275295317173,
      "num_tokens": 7820991.0,
      "step": 3320
    },
    {
      "entropy": 5.551218843460083,
      "epoch": 3.3861436576668362,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042705872809412955,
      "loss": 5.3307,
      "mean_token_accuracy": 0.1872739389538765,
      "num_tokens": 7832694.0,
      "step": 3325
    },
    {
      "entropy": 5.506617021560669,
      "epoch": 3.3912379011716762,
      "grad_norm": 1.03125,
      "learning_rate": 0.00042676279757864404,
      "loss": 5.2619,
      "mean_token_accuracy": 0.20017192661762237,
      "num_tokens": 7844132.0,
      "step": 3330
    },
    {
      "entropy": 5.494177532196045,
      "epoch": 3.3963321446765153,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042646638461223076,
      "loss": 5.2625,
      "mean_token_accuracy": 0.19145260751247406,
      "num_tokens": 7854232.0,
      "step": 3335
    },
    {
      "entropy": 5.599319982528686,
      "epoch": 3.401426388181355,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004261694901371807,
      "loss": 5.2803,
      "mean_token_accuracy": 0.18350138813257216,
      "num_tokens": 7865906.0,
      "step": 3340
    },
    {
      "entropy": 5.517926216125488,
      "epoch": 3.4065206316861945,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004258721150973155,
      "loss": 5.2694,
      "mean_token_accuracy": 0.18290073275566102,
      "num_tokens": 7877029.0,
      "step": 3345
    },
    {
      "entropy": 5.522620391845703,
      "epoch": 3.411614875191034,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004255742604379845,
      "loss": 5.3202,
      "mean_token_accuracy": 0.1834985375404358,
      "num_tokens": 7889245.0,
      "step": 3350
    },
    {
      "entropy": 5.5225183963775635,
      "epoch": 3.4167091186958736,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004252759271060618,
      "loss": 5.2016,
      "mean_token_accuracy": 0.1915111646056175,
      "num_tokens": 7900117.0,
      "step": 3355
    },
    {
      "entropy": 5.54949049949646,
      "epoch": 3.421803362200713,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042497711604994314,
      "loss": 5.3096,
      "mean_token_accuracy": 0.183358334004879,
      "num_tokens": 7910746.0,
      "step": 3360
    },
    {
      "entropy": 5.481787014007568,
      "epoch": 3.4268976057055527,
      "grad_norm": 1.21875,
      "learning_rate": 0.00042467782821954306,
      "loss": 5.2159,
      "mean_token_accuracy": 0.19784331768751146,
      "num_tokens": 7922351.0,
      "step": 3365
    },
    {
      "entropy": 5.528353071212768,
      "epoch": 3.4319918492103922,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00042437806456629154,
      "loss": 5.3226,
      "mean_token_accuracy": 0.18049002885818483,
      "num_tokens": 7934902.0,
      "step": 3370
    },
    {
      "entropy": 5.4786989212036135,
      "epoch": 3.437086092715232,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004240778260431311,
      "loss": 5.2117,
      "mean_token_accuracy": 0.19243534207344054,
      "num_tokens": 7946920.0,
      "step": 3375
    },
    {
      "entropy": 5.505074405670166,
      "epoch": 3.4421803362200714,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004237771136045144,
      "loss": 5.2949,
      "mean_token_accuracy": 0.19107773751020432,
      "num_tokens": 7959667.0,
      "step": 3380
    },
    {
      "entropy": 5.474018812179565,
      "epoch": 3.447274579724911,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004234759282064001,
      "loss": 5.2371,
      "mean_token_accuracy": 0.19770409315824508,
      "num_tokens": 7971762.0,
      "step": 3385
    },
    {
      "entropy": 5.4705976963043215,
      "epoch": 3.4523688232297505,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004231742708062508,
      "loss": 5.2663,
      "mean_token_accuracy": 0.18874183744192125,
      "num_tokens": 7983586.0,
      "step": 3390
    },
    {
      "entropy": 5.567195320129395,
      "epoch": 3.45746306673459,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004228721423630292,
      "loss": 5.2862,
      "mean_token_accuracy": 0.18435157239437103,
      "num_tokens": 7994554.0,
      "step": 3395
    },
    {
      "entropy": 5.564591932296753,
      "epoch": 3.4625573102394296,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004225695438371958,
      "loss": 5.355,
      "mean_token_accuracy": 0.18007050156593324,
      "num_tokens": 8008486.0,
      "step": 3400
    },
    {
      "entropy": 5.5197381496429445,
      "epoch": 3.467651553744269,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004222664761907052,
      "loss": 5.3091,
      "mean_token_accuracy": 0.18644343614578246,
      "num_tokens": 8019684.0,
      "step": 3405
    },
    {
      "entropy": 5.514086294174194,
      "epoch": 3.4727457972491087,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004219629403870035,
      "loss": 5.3283,
      "mean_token_accuracy": 0.18536198884248734,
      "num_tokens": 8032093.0,
      "step": 3410
    },
    {
      "entropy": 5.524766445159912,
      "epoch": 3.4778400407539483,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00042165893739102504,
      "loss": 5.3424,
      "mean_token_accuracy": 0.18187156021595002,
      "num_tokens": 8044439.0,
      "step": 3415
    },
    {
      "entropy": 5.523378324508667,
      "epoch": 3.4829342842587874,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042135446816918927,
      "loss": 5.2412,
      "mean_token_accuracy": 0.18457507938146592,
      "num_tokens": 8055840.0,
      "step": 3420
    },
    {
      "entropy": 5.486256313323975,
      "epoch": 3.488028527763627,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042104953368939785,
      "loss": 5.2084,
      "mean_token_accuracy": 0.19623675495386123,
      "num_tokens": 8066362.0,
      "step": 3425
    },
    {
      "entropy": 5.463621425628662,
      "epoch": 3.4931227712684665,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004207441349210315,
      "loss": 5.2574,
      "mean_token_accuracy": 0.19175745099782943,
      "num_tokens": 8077934.0,
      "step": 3430
    },
    {
      "entropy": 5.444260501861573,
      "epoch": 3.498217014773306,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042043827283494667,
      "loss": 5.1187,
      "mean_token_accuracy": 0.20353888124227523,
      "num_tokens": 8089195.0,
      "step": 3435
    },
    {
      "entropy": 5.524917030334473,
      "epoch": 3.5033112582781456,
      "grad_norm": 1.125,
      "learning_rate": 0.0004201319484034731,
      "loss": 5.24,
      "mean_token_accuracy": 0.18563750833272935,
      "num_tokens": 8099862.0,
      "step": 3440
    },
    {
      "entropy": 5.429136943817139,
      "epoch": 3.508405501782985,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041982516260040994,
      "loss": 5.2604,
      "mean_token_accuracy": 0.19045042097568513,
      "num_tokens": 8111574.0,
      "step": 3445
    },
    {
      "entropy": 5.5911602020263675,
      "epoch": 3.5134997452878247,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041951791640102326,
      "loss": 5.2902,
      "mean_token_accuracy": 0.1805243521928787,
      "num_tokens": 8124165.0,
      "step": 3450
    },
    {
      "entropy": 5.521031284332276,
      "epoch": 3.5185939887926643,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004192102107820426,
      "loss": 5.2656,
      "mean_token_accuracy": 0.1881466954946518,
      "num_tokens": 8135409.0,
      "step": 3455
    },
    {
      "entropy": 5.494682836532593,
      "epoch": 3.523688232297504,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00041890204672165803,
      "loss": 5.2609,
      "mean_token_accuracy": 0.1869012787938118,
      "num_tokens": 8147816.0,
      "step": 3460
    },
    {
      "entropy": 5.526585340499878,
      "epoch": 3.5287824758023434,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00041859342519951696,
      "loss": 5.3119,
      "mean_token_accuracy": 0.17959794849157334,
      "num_tokens": 8160016.0,
      "step": 3465
    },
    {
      "entropy": 5.518149757385254,
      "epoch": 3.533876719307183,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004182843471967212,
      "loss": 5.3176,
      "mean_token_accuracy": 0.1795201450586319,
      "num_tokens": 8171110.0,
      "step": 3470
    },
    {
      "entropy": 5.546420907974243,
      "epoch": 3.5389709628120225,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041797481369582355,
      "loss": 5.3158,
      "mean_token_accuracy": 0.18323036879301072,
      "num_tokens": 8183927.0,
      "step": 3475
    },
    {
      "entropy": 5.536783504486084,
      "epoch": 3.544065206316862,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041766482568082486,
      "loss": 5.2717,
      "mean_token_accuracy": 0.1956411376595497,
      "num_tokens": 8195603.0,
      "step": 3480
    },
    {
      "entropy": 5.468293905258179,
      "epoch": 3.5491594498217016,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004173543841371709,
      "loss": 5.2733,
      "mean_token_accuracy": 0.19261570721864701,
      "num_tokens": 8207163.0,
      "step": 3485
    },
    {
      "entropy": 5.444919204711914,
      "epoch": 3.554253693326541,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004170434900517492,
      "loss": 5.1479,
      "mean_token_accuracy": 0.19642960131168366,
      "num_tokens": 8218751.0,
      "step": 3490
    },
    {
      "entropy": 5.498192214965821,
      "epoch": 3.5593479368313803,
      "grad_norm": 1.0,
      "learning_rate": 0.00041673214441288595,
      "loss": 5.3053,
      "mean_token_accuracy": 0.1814761459827423,
      "num_tokens": 8231404.0,
      "step": 3495
    },
    {
      "entropy": 5.551133537292481,
      "epoch": 3.5644421803362203,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004164203482103426,
      "loss": 5.3213,
      "mean_token_accuracy": 0.17414924055337905,
      "num_tokens": 8244087.0,
      "step": 3500
    },
    {
      "epoch": 3.5644421803362203,
      "eval_entropy": 5.314532568425308,
      "eval_loss": 5.78098726272583,
      "eval_mean_token_accuracy": 0.1709617973661717,
      "eval_num_tokens": 8244087.0,
      "eval_runtime": 2.7507,
      "eval_samples_per_second": 1177.895,
      "eval_steps_per_second": 147.237,
      "step": 3500
    },
    {
      "entropy": 5.505230569839478,
      "epoch": 3.5695364238410594,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004161081024353132,
      "loss": 5.2716,
      "mean_token_accuracy": 0.18772488534450532,
      "num_tokens": 8255107.0,
      "step": 3505
    },
    {
      "entropy": 5.519088315963745,
      "epoch": 3.5746306673458994,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041579540808042096,
      "loss": 5.3052,
      "mean_token_accuracy": 0.17553389221429824,
      "num_tokens": 8267268.0,
      "step": 3510
    },
    {
      "entropy": 5.502148675918579,
      "epoch": 3.5797249108507385,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000415482266139715,
      "loss": 5.3248,
      "mean_token_accuracy": 0.18211540579795837,
      "num_tokens": 8279022.0,
      "step": 3515
    },
    {
      "entropy": 5.507491874694824,
      "epoch": 3.584819154355578,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004151686776086673,
      "loss": 5.291,
      "mean_token_accuracy": 0.19002780616283416,
      "num_tokens": 8291460.0,
      "step": 3520
    },
    {
      "entropy": 5.56368579864502,
      "epoch": 3.5899133978604176,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00041485464348416963,
      "loss": 5.292,
      "mean_token_accuracy": 0.19022610038518906,
      "num_tokens": 8302912.0,
      "step": 3525
    },
    {
      "entropy": 5.506591844558716,
      "epoch": 3.595007641365257,
      "grad_norm": 1.046875,
      "learning_rate": 0.00041454016476453017,
      "loss": 5.2864,
      "mean_token_accuracy": 0.18261739909648894,
      "num_tokens": 8316021.0,
      "step": 3530
    },
    {
      "entropy": 5.481910228729248,
      "epoch": 3.6001018848700967,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041422524244947066,
      "loss": 5.2623,
      "mean_token_accuracy": 0.18677586317062378,
      "num_tokens": 8326686.0,
      "step": 3535
    },
    {
      "entropy": 5.479059791564941,
      "epoch": 3.6051961283749363,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004139098775401228,
      "loss": 5.3193,
      "mean_token_accuracy": 0.18399812579154967,
      "num_tokens": 8338233.0,
      "step": 3540
    },
    {
      "entropy": 5.540248775482178,
      "epoch": 3.610290371879776,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041359407103902556,
      "loss": 5.2666,
      "mean_token_accuracy": 0.18573987931013108,
      "num_tokens": 8350040.0,
      "step": 3545
    },
    {
      "entropy": 5.493269157409668,
      "epoch": 3.6153846153846154,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004132778239501214,
      "loss": 5.286,
      "mean_token_accuracy": 0.18652140200138093,
      "num_tokens": 8361608.0,
      "step": 3550
    },
    {
      "entropy": 5.55647463798523,
      "epoch": 3.620478858889455,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041296113727875346,
      "loss": 5.3338,
      "mean_token_accuracy": 0.18116967082023622,
      "num_tokens": 8374179.0,
      "step": 3555
    },
    {
      "entropy": 5.4600788116455075,
      "epoch": 3.6255731023942945,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041264401203166266,
      "loss": 5.2441,
      "mean_token_accuracy": 0.19064738601446152,
      "num_tokens": 8385715.0,
      "step": 3560
    },
    {
      "entropy": 5.5325511455535885,
      "epoch": 3.630667345899134,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041232644921698357,
      "loss": 5.2613,
      "mean_token_accuracy": 0.19022748470306397,
      "num_tokens": 8396948.0,
      "step": 3565
    },
    {
      "entropy": 5.5715700626373295,
      "epoch": 3.6357615894039736,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004120084498442425,
      "loss": 5.3647,
      "mean_token_accuracy": 0.18287853449583052,
      "num_tokens": 8408652.0,
      "step": 3570
    },
    {
      "entropy": 5.429012823104858,
      "epoch": 3.640855832908813,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041169001492435275,
      "loss": 5.186,
      "mean_token_accuracy": 0.19066958427429198,
      "num_tokens": 8419857.0,
      "step": 3575
    },
    {
      "entropy": 5.552859544754028,
      "epoch": 3.6459500764136523,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004113711454696128,
      "loss": 5.2511,
      "mean_token_accuracy": 0.18854512870311738,
      "num_tokens": 8432085.0,
      "step": 3580
    },
    {
      "entropy": 5.549443912506104,
      "epoch": 3.6510443199184923,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041105184249370236,
      "loss": 5.274,
      "mean_token_accuracy": 0.18390826433897017,
      "num_tokens": 8443435.0,
      "step": 3585
    },
    {
      "entropy": 5.45549635887146,
      "epoch": 3.6561385634233314,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004107321070116792,
      "loss": 5.2493,
      "mean_token_accuracy": 0.19264701008796692,
      "num_tokens": 8456049.0,
      "step": 3590
    },
    {
      "entropy": 5.533119344711304,
      "epoch": 3.6612328069281714,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041041194003997616,
      "loss": 5.3294,
      "mean_token_accuracy": 0.18397263586521148,
      "num_tokens": 8469699.0,
      "step": 3595
    },
    {
      "entropy": 5.493679094314575,
      "epoch": 3.6663270504330105,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041009134259639766,
      "loss": 5.3457,
      "mean_token_accuracy": 0.17645138651132583,
      "num_tokens": 8480354.0,
      "step": 3600
    },
    {
      "entropy": 5.590160226821899,
      "epoch": 3.67142129393785,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004097703157001166,
      "loss": 5.3437,
      "mean_token_accuracy": 0.18699588030576705,
      "num_tokens": 8492475.0,
      "step": 3605
    },
    {
      "entropy": 5.502712678909302,
      "epoch": 3.6765155374426897,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004094488603716712,
      "loss": 5.3277,
      "mean_token_accuracy": 0.1851692169904709,
      "num_tokens": 8504272.0,
      "step": 3610
    },
    {
      "entropy": 5.507415103912353,
      "epoch": 3.681609780947529,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004091269776329614,
      "loss": 5.2647,
      "mean_token_accuracy": 0.18889132291078567,
      "num_tokens": 8515803.0,
      "step": 3615
    },
    {
      "entropy": 5.52598466873169,
      "epoch": 3.6867040244523688,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004088046685072462,
      "loss": 5.3415,
      "mean_token_accuracy": 0.18917083591222764,
      "num_tokens": 8527781.0,
      "step": 3620
    },
    {
      "entropy": 5.556311321258545,
      "epoch": 3.6917982679572083,
      "grad_norm": 1.0625,
      "learning_rate": 0.00040848193401914,
      "loss": 5.4458,
      "mean_token_accuracy": 0.17584429681301117,
      "num_tokens": 8542380.0,
      "step": 3625
    },
    {
      "entropy": 5.516146087646485,
      "epoch": 3.696892511462048,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040815877519460923,
      "loss": 5.2554,
      "mean_token_accuracy": 0.18828494995832443,
      "num_tokens": 8553729.0,
      "step": 3630
    },
    {
      "entropy": 5.570184326171875,
      "epoch": 3.7019867549668874,
      "grad_norm": 1.078125,
      "learning_rate": 0.00040783519306096934,
      "loss": 5.2859,
      "mean_token_accuracy": 0.18657394796609877,
      "num_tokens": 8565193.0,
      "step": 3635
    },
    {
      "entropy": 5.563874673843384,
      "epoch": 3.707080998471727,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004075111886468817,
      "loss": 5.3169,
      "mean_token_accuracy": 0.18157622814178467,
      "num_tokens": 8577418.0,
      "step": 3640
    },
    {
      "entropy": 5.462403011322022,
      "epoch": 3.7121752419765666,
      "grad_norm": 1.125,
      "learning_rate": 0.00040718676298234975,
      "loss": 5.2203,
      "mean_token_accuracy": 0.19786756336688996,
      "num_tokens": 8589139.0,
      "step": 3645
    },
    {
      "entropy": 5.5289689064025875,
      "epoch": 3.717269485481406,
      "grad_norm": 1.03125,
      "learning_rate": 0.00040686191709871637,
      "loss": 5.2582,
      "mean_token_accuracy": 0.1845638245344162,
      "num_tokens": 8601093.0,
      "step": 3650
    },
    {
      "entropy": 5.518384075164795,
      "epoch": 3.7223637289862457,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004065366520286601,
      "loss": 5.2847,
      "mean_token_accuracy": 0.18624062985181808,
      "num_tokens": 8612969.0,
      "step": 3655
    },
    {
      "entropy": 5.513448286056518,
      "epoch": 3.7274579724910852,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004062109688061921,
      "loss": 5.2194,
      "mean_token_accuracy": 0.19028918743133544,
      "num_tokens": 8624991.0,
      "step": 3660
    },
    {
      "entropy": 5.491545009613037,
      "epoch": 3.7325522159959243,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040588486846665316,
      "loss": 5.2882,
      "mean_token_accuracy": 0.18945505619049072,
      "num_tokens": 8636124.0,
      "step": 3665
    },
    {
      "entropy": 5.455791282653808,
      "epoch": 3.7376464595007644,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004055583520467096,
      "loss": 5.3063,
      "mean_token_accuracy": 0.18936239778995514,
      "num_tokens": 8649263.0,
      "step": 3670
    },
    {
      "entropy": 5.480527353286743,
      "epoch": 3.7427407030056035,
      "grad_norm": 1.03125,
      "learning_rate": 0.00040523142058435064,
      "loss": 5.2793,
      "mean_token_accuracy": 0.1879584848880768,
      "num_tokens": 8661415.0,
      "step": 3675
    },
    {
      "entropy": 5.525414609909058,
      "epoch": 3.7478349465104435,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00040490407511888495,
      "loss": 5.2691,
      "mean_token_accuracy": 0.18875865042209625,
      "num_tokens": 8673037.0,
      "step": 3680
    },
    {
      "entropy": 5.45166130065918,
      "epoch": 3.7529291900152826,
      "grad_norm": 1.0,
      "learning_rate": 0.0004045763166909374,
      "loss": 5.2701,
      "mean_token_accuracy": 0.19220236241817473,
      "num_tokens": 8687500.0,
      "step": 3685
    },
    {
      "entropy": 5.446043300628662,
      "epoch": 3.758023433520122,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004042481463424453,
      "loss": 5.1983,
      "mean_token_accuracy": 0.19016106873750688,
      "num_tokens": 8699414.0,
      "step": 3690
    },
    {
      "entropy": 5.362189531326294,
      "epoch": 3.7631176770249617,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004039195651166558,
      "loss": 5.1579,
      "mean_token_accuracy": 0.19531481564044953,
      "num_tokens": 8711318.0,
      "step": 3695
    },
    {
      "entropy": 5.5171928882598875,
      "epoch": 3.7682119205298013,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004035905740581222,
      "loss": 5.3234,
      "mean_token_accuracy": 0.18783893436193466,
      "num_tokens": 8723418.0,
      "step": 3700
    },
    {
      "entropy": 5.449063539505005,
      "epoch": 3.773306164034641,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00040326117421270055,
      "loss": 5.193,
      "mean_token_accuracy": 0.1946106359362602,
      "num_tokens": 8735549.0,
      "step": 3705
    },
    {
      "entropy": 5.516118383407592,
      "epoch": 3.7784004075394804,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004029313666275463,
      "loss": 5.2972,
      "mean_token_accuracy": 0.18847466111183167,
      "num_tokens": 8746755.0,
      "step": 3710
    },
    {
      "entropy": 5.484922075271607,
      "epoch": 3.78349465104432,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004026011523511112,
      "loss": 5.3242,
      "mean_token_accuracy": 0.1802894502878189,
      "num_tokens": 8758630.0,
      "step": 3715
    },
    {
      "entropy": 5.625292253494263,
      "epoch": 3.7885888945491595,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00040227053243314015,
      "loss": 5.3395,
      "mean_token_accuracy": 0.1827728495001793,
      "num_tokens": 8770906.0,
      "step": 3720
    },
    {
      "entropy": 5.421060419082641,
      "epoch": 3.793683138053999,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004019395079246672,
      "loss": 5.1541,
      "mean_token_accuracy": 0.20187637507915496,
      "num_tokens": 8781666.0,
      "step": 3725
    },
    {
      "entropy": 5.441168022155762,
      "epoch": 3.7987773815588386,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040160807987801265,
      "loss": 5.265,
      "mean_token_accuracy": 0.18436054438352584,
      "num_tokens": 8793065.0,
      "step": 3730
    },
    {
      "entropy": 5.521347188949585,
      "epoch": 3.803871625063678,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00040127624934677966,
      "loss": 5.3,
      "mean_token_accuracy": 0.18140489161014556,
      "num_tokens": 8804877.0,
      "step": 3735
    },
    {
      "entropy": 5.494076776504516,
      "epoch": 3.8089658685685177,
      "grad_norm": 1.125,
      "learning_rate": 0.000400944017385851,
      "loss": 5.2571,
      "mean_token_accuracy": 0.18579984754323958,
      "num_tokens": 8816151.0,
      "step": 3740
    },
    {
      "entropy": 5.444279432296753,
      "epoch": 3.8140601120733573,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040061138505138546,
      "loss": 5.163,
      "mean_token_accuracy": 0.19901911467313765,
      "num_tokens": 8827465.0,
      "step": 3745
    },
    {
      "entropy": 5.500946569442749,
      "epoch": 3.8191543555781964,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004002783534008147,
      "loss": 5.3297,
      "mean_token_accuracy": 0.18310041576623917,
      "num_tokens": 8840068.0,
      "step": 3750
    },
    {
      "entropy": 5.490247821807861,
      "epoch": 3.8242485990830364,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039994492349283966,
      "loss": 5.2583,
      "mean_token_accuracy": 0.19094090312719345,
      "num_tokens": 8851775.0,
      "step": 3755
    },
    {
      "entropy": 5.454621934890747,
      "epoch": 3.8293428425878755,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003996110963874274,
      "loss": 5.298,
      "mean_token_accuracy": 0.18614607453346252,
      "num_tokens": 8863704.0,
      "step": 3760
    },
    {
      "entropy": 5.5926073551177975,
      "epoch": 3.8344370860927155,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00039927687314580783,
      "loss": 5.3727,
      "mean_token_accuracy": 0.17817466109991073,
      "num_tokens": 8876411.0,
      "step": 3765
    },
    {
      "entropy": 5.546767807006836,
      "epoch": 3.8395313295975546,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039894225483047,
      "loss": 5.2601,
      "mean_token_accuracy": 0.18924956321716307,
      "num_tokens": 8888104.0,
      "step": 3770
    },
    {
      "entropy": 5.5138482570648195,
      "epoch": 3.844625573102394,
      "grad_norm": 1.0,
      "learning_rate": 0.0003986072425051587,
      "loss": 5.3159,
      "mean_token_accuracy": 0.18107025921344758,
      "num_tokens": 8901227.0,
      "step": 3775
    },
    {
      "entropy": 5.508226776123047,
      "epoch": 3.8497198166072337,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00039827183723487156,
      "loss": 5.252,
      "mean_token_accuracy": 0.18441893458366393,
      "num_tokens": 8913052.0,
      "step": 3780
    },
    {
      "entropy": 5.479751396179199,
      "epoch": 3.8548140601120733,
      "grad_norm": 1.09375,
      "learning_rate": 0.00039793604008585536,
      "loss": 5.2429,
      "mean_token_accuracy": 0.19081486612558365,
      "num_tokens": 8924303.0,
      "step": 3785
    },
    {
      "entropy": 5.518397855758667,
      "epoch": 3.859908303616913,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003975998521256026,
      "loss": 5.2928,
      "mean_token_accuracy": 0.18275457620620728,
      "num_tokens": 8935920.0,
      "step": 3790
    },
    {
      "entropy": 5.432967185974121,
      "epoch": 3.8650025471217524,
      "grad_norm": 1.15625,
      "learning_rate": 0.000397263274422848,
      "loss": 5.2147,
      "mean_token_accuracy": 0.19071601778268815,
      "num_tokens": 8947181.0,
      "step": 3795
    },
    {
      "entropy": 5.470023155212402,
      "epoch": 3.870096790626592,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003969263080475657,
      "loss": 5.2832,
      "mean_token_accuracy": 0.19259316623210906,
      "num_tokens": 8958275.0,
      "step": 3800
    },
    {
      "entropy": 5.490440607070923,
      "epoch": 3.8751910341314315,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039658895407096483,
      "loss": 5.2007,
      "mean_token_accuracy": 0.1861083596944809,
      "num_tokens": 8969319.0,
      "step": 3805
    },
    {
      "entropy": 5.490753602981568,
      "epoch": 3.880285277636271,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003962512135654871,
      "loss": 5.3022,
      "mean_token_accuracy": 0.17727622985839844,
      "num_tokens": 8980710.0,
      "step": 3810
    },
    {
      "entropy": 5.460601043701172,
      "epoch": 3.8853795211411106,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039591308760480293,
      "loss": 5.2655,
      "mean_token_accuracy": 0.18398440331220628,
      "num_tokens": 8991779.0,
      "step": 3815
    },
    {
      "entropy": 5.474878931045533,
      "epoch": 3.89047376464595,
      "grad_norm": 1.125,
      "learning_rate": 0.00039557457726380803,
      "loss": 5.189,
      "mean_token_accuracy": 0.19069245606660842,
      "num_tokens": 9003351.0,
      "step": 3820
    },
    {
      "entropy": 5.464775323867798,
      "epoch": 3.8955680081507897,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039523568361862007,
      "loss": 5.2423,
      "mean_token_accuracy": 0.19078782796859742,
      "num_tokens": 9015113.0,
      "step": 3825
    },
    {
      "entropy": 5.439772367477417,
      "epoch": 3.9006622516556293,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039489640774657527,
      "loss": 5.3554,
      "mean_token_accuracy": 0.1853403180837631,
      "num_tokens": 9026529.0,
      "step": 3830
    },
    {
      "entropy": 5.478518342971801,
      "epoch": 3.9057564951604684,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039455675072622485,
      "loss": 5.2359,
      "mean_token_accuracy": 0.1945350557565689,
      "num_tokens": 9038007.0,
      "step": 3835
    },
    {
      "entropy": 5.495754814147949,
      "epoch": 3.9108507386653084,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003942167136373317,
      "loss": 5.2011,
      "mean_token_accuracy": 0.19397969394922257,
      "num_tokens": 9049284.0,
      "step": 3840
    },
    {
      "entropy": 5.371614646911621,
      "epoch": 3.9159449821701475,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003938762975608671,
      "loss": 5.2385,
      "mean_token_accuracy": 0.1894398584961891,
      "num_tokens": 9060678.0,
      "step": 3845
    },
    {
      "entropy": 5.534006118774414,
      "epoch": 3.9210392256749875,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003935355035790071,
      "loss": 5.3552,
      "mean_token_accuracy": 0.18180383294820784,
      "num_tokens": 9071864.0,
      "step": 3850
    },
    {
      "entropy": 5.423132085800171,
      "epoch": 3.9261334691798266,
      "grad_norm": 1.125,
      "learning_rate": 0.00039319433277512866,
      "loss": 5.2385,
      "mean_token_accuracy": 0.19444318115711212,
      "num_tokens": 9083578.0,
      "step": 3855
    },
    {
      "entropy": 5.5640869140625,
      "epoch": 3.931227712684666,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00039285278623380746,
      "loss": 5.3379,
      "mean_token_accuracy": 0.18280695229768754,
      "num_tokens": 9096148.0,
      "step": 3860
    },
    {
      "entropy": 5.51947283744812,
      "epoch": 3.9363219561895058,
      "grad_norm": 1.09375,
      "learning_rate": 0.00039251086504081285,
      "loss": 5.2901,
      "mean_token_accuracy": 0.18642651289701462,
      "num_tokens": 9108963.0,
      "step": 3865
    },
    {
      "entropy": 5.5186632633209225,
      "epoch": 3.9414161996943453,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039216857028310576,
      "loss": 5.4047,
      "mean_token_accuracy": 0.17489920407533646,
      "num_tokens": 9120865.0,
      "step": 3870
    },
    {
      "entropy": 5.490428495407104,
      "epoch": 3.946510443199185,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003918259030488342,
      "loss": 5.2533,
      "mean_token_accuracy": 0.188284932076931,
      "num_tokens": 9132401.0,
      "step": 3875
    },
    {
      "entropy": 5.5043436050415036,
      "epoch": 3.9516046867040244,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00039148286442733073,
      "loss": 5.2932,
      "mean_token_accuracy": 0.18889451026916504,
      "num_tokens": 9145291.0,
      "step": 3880
    },
    {
      "entropy": 5.442192077636719,
      "epoch": 3.956698930208864,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003911394555091083,
      "loss": 5.1637,
      "mean_token_accuracy": 0.20066449493169786,
      "num_tokens": 9155811.0,
      "step": 3885
    },
    {
      "entropy": 5.524414300918579,
      "epoch": 3.9617931737137035,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039079567738585703,
      "loss": 5.2415,
      "mean_token_accuracy": 0.17859885543584825,
      "num_tokens": 9166535.0,
      "step": 3890
    },
    {
      "entropy": 5.42221417427063,
      "epoch": 3.966887417218543,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003904515311504407,
      "loss": 5.1992,
      "mean_token_accuracy": 0.19670761227607728,
      "num_tokens": 9179751.0,
      "step": 3895
    },
    {
      "entropy": 5.392773056030274,
      "epoch": 3.9719816607233827,
      "grad_norm": 1.03125,
      "learning_rate": 0.00039010701789689344,
      "loss": 5.1947,
      "mean_token_accuracy": 0.193730428814888,
      "num_tokens": 9192266.0,
      "step": 3900
    },
    {
      "entropy": 5.410204219818115,
      "epoch": 3.977075904228222,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003897621387204161,
      "loss": 5.2099,
      "mean_token_accuracy": 0.19288454651832582,
      "num_tokens": 9204093.0,
      "step": 3905
    },
    {
      "entropy": 5.522636032104492,
      "epoch": 3.9821701477330618,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003894168947173727,
      "loss": 5.2381,
      "mean_token_accuracy": 0.1873869225382805,
      "num_tokens": 9214815.0,
      "step": 3910
    },
    {
      "entropy": 5.46841950416565,
      "epoch": 3.9872643912379013,
      "grad_norm": 1.1875,
      "learning_rate": 0.00038907128698528705,
      "loss": 5.2482,
      "mean_token_accuracy": 0.1925479918718338,
      "num_tokens": 9226770.0,
      "step": 3915
    },
    {
      "entropy": 5.3238434314727785,
      "epoch": 3.9923586347427404,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003887253166228394,
      "loss": 5.1694,
      "mean_token_accuracy": 0.2025023952126503,
      "num_tokens": 9237733.0,
      "step": 3920
    },
    {
      "entropy": 5.549081420898437,
      "epoch": 3.9974528782475804,
      "grad_norm": 0.96875,
      "learning_rate": 0.00038837898472986247,
      "loss": 5.313,
      "mean_token_accuracy": 0.18289730399847032,
      "num_tokens": 9250539.0,
      "step": 3925
    },
    {
      "entropy": 5.520198080274794,
      "epoch": 4.002037697401936,
      "grad_norm": 0.984375,
      "learning_rate": 0.00038803229240733886,
      "loss": 5.1717,
      "mean_token_accuracy": 0.1930143866274092,
      "num_tokens": 9261827.0,
      "step": 3930
    },
    {
      "entropy": 5.313441324234009,
      "epoch": 4.007131940906775,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003876852407573962,
      "loss": 4.8935,
      "mean_token_accuracy": 0.2142712488770485,
      "num_tokens": 9273351.0,
      "step": 3935
    },
    {
      "entropy": 5.4588303565979,
      "epoch": 4.012226184411615,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000387337830883305,
      "loss": 4.9282,
      "mean_token_accuracy": 0.21729503571987152,
      "num_tokens": 9285087.0,
      "step": 3940
    },
    {
      "entropy": 5.412793350219727,
      "epoch": 4.017320427916454,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003869900638894743,
      "loss": 4.9694,
      "mean_token_accuracy": 0.2091067314147949,
      "num_tokens": 9296416.0,
      "step": 3945
    },
    {
      "entropy": 5.473224544525147,
      "epoch": 4.022414671421294,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00038664194088144867,
      "loss": 4.8842,
      "mean_token_accuracy": 0.21790419965982438,
      "num_tokens": 9307163.0,
      "step": 3950
    },
    {
      "entropy": 5.439658546447754,
      "epoch": 4.027508914926133,
      "grad_norm": 1.171875,
      "learning_rate": 0.000386293462965904,
      "loss": 4.9667,
      "mean_token_accuracy": 0.2052128255367279,
      "num_tokens": 9319318.0,
      "step": 3955
    },
    {
      "entropy": 5.414942359924316,
      "epoch": 4.032603158430973,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003859446312506449,
      "loss": 4.986,
      "mean_token_accuracy": 0.20205750614404677,
      "num_tokens": 9331454.0,
      "step": 3960
    },
    {
      "entropy": 5.488663625717163,
      "epoch": 4.037697401935812,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003855954468446003,
      "loss": 5.0029,
      "mean_token_accuracy": 0.20646335035562516,
      "num_tokens": 9343881.0,
      "step": 3965
    },
    {
      "entropy": 5.500966262817383,
      "epoch": 4.042791645440652,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00038524591085782063,
      "loss": 5.0149,
      "mean_token_accuracy": 0.1964875265955925,
      "num_tokens": 9354659.0,
      "step": 3970
    },
    {
      "entropy": 5.4410261631011965,
      "epoch": 4.047885888945491,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003848960244014738,
      "loss": 4.9011,
      "mean_token_accuracy": 0.21452577263116837,
      "num_tokens": 9366176.0,
      "step": 3975
    },
    {
      "entropy": 5.366271448135376,
      "epoch": 4.052980132450331,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003845457885878421,
      "loss": 4.9804,
      "mean_token_accuracy": 0.21089834868907928,
      "num_tokens": 9379433.0,
      "step": 3980
    },
    {
      "entropy": 5.420285320281982,
      "epoch": 4.0580743759551705,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003841952045303179,
      "loss": 4.9275,
      "mean_token_accuracy": 0.2145117223262787,
      "num_tokens": 9391264.0,
      "step": 3985
    },
    {
      "entropy": 5.36814661026001,
      "epoch": 4.0631686194600105,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003838442733434013,
      "loss": 4.939,
      "mean_token_accuracy": 0.21000799536705017,
      "num_tokens": 9404114.0,
      "step": 3990
    },
    {
      "entropy": 5.348601484298706,
      "epoch": 4.06826286296485,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038349299614269544,
      "loss": 4.8332,
      "mean_token_accuracy": 0.2262760654091835,
      "num_tokens": 9415657.0,
      "step": 3995
    },
    {
      "entropy": 5.413491725921631,
      "epoch": 4.07335710646969,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003831413740449037,
      "loss": 4.9496,
      "mean_token_accuracy": 0.21081938445568085,
      "num_tokens": 9426154.0,
      "step": 4000
    },
    {
      "epoch": 4.07335710646969,
      "eval_entropy": 5.232787068684896,
      "eval_loss": 5.7487969398498535,
      "eval_mean_token_accuracy": 0.1739971618593475,
      "eval_num_tokens": 9426154.0,
      "eval_runtime": 2.6825,
      "eval_samples_per_second": 1207.838,
      "eval_steps_per_second": 150.98,
      "step": 4000
    },
    {
      "entropy": 5.449730253219604,
      "epoch": 4.078451349974529,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003827894081678255,
      "loss": 4.9953,
      "mean_token_accuracy": 0.20456648617982864,
      "num_tokens": 9438006.0,
      "step": 4005
    },
    {
      "entropy": 5.453718137741089,
      "epoch": 4.083545593479369,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003824370996303538,
      "loss": 4.9672,
      "mean_token_accuracy": 0.20590050220489503,
      "num_tokens": 9451331.0,
      "step": 4010
    },
    {
      "entropy": 5.460933685302734,
      "epoch": 4.088639836984208,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00038208444955247005,
      "loss": 5.0397,
      "mean_token_accuracy": 0.1994566723704338,
      "num_tokens": 9462907.0,
      "step": 4015
    },
    {
      "entropy": 5.331179285049439,
      "epoch": 4.093734080489047,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003817314590552421,
      "loss": 4.9537,
      "mean_token_accuracy": 0.21153350472450255,
      "num_tokens": 9474294.0,
      "step": 4020
    },
    {
      "entropy": 5.417905664443969,
      "epoch": 4.098828323993887,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00038137812926081986,
      "loss": 4.9491,
      "mean_token_accuracy": 0.20954519808292388,
      "num_tokens": 9486728.0,
      "step": 4025
    },
    {
      "entropy": 5.414187622070313,
      "epoch": 4.103922567498726,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003810244612924314,
      "loss": 5.0221,
      "mean_token_accuracy": 0.19837037175893785,
      "num_tokens": 9498433.0,
      "step": 4030
    },
    {
      "entropy": 5.354794311523437,
      "epoch": 4.109016811003566,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003806704562743805,
      "loss": 4.9155,
      "mean_token_accuracy": 0.19975878447294235,
      "num_tokens": 9510617.0,
      "step": 4035
    },
    {
      "entropy": 5.464527320861817,
      "epoch": 4.114111054508405,
      "grad_norm": 1.25,
      "learning_rate": 0.0003803161153320418,
      "loss": 4.935,
      "mean_token_accuracy": 0.20716629773378373,
      "num_tokens": 9521755.0,
      "step": 4040
    },
    {
      "entropy": 5.418537235260009,
      "epoch": 4.119205298013245,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00037996143959185847,
      "loss": 5.0653,
      "mean_token_accuracy": 0.20131232291460038,
      "num_tokens": 9534796.0,
      "step": 4045
    },
    {
      "entropy": 5.447728395462036,
      "epoch": 4.124299541518084,
      "grad_norm": 0.9375,
      "learning_rate": 0.00037960643018133743,
      "loss": 5.0501,
      "mean_token_accuracy": 0.19971159100532532,
      "num_tokens": 9546718.0,
      "step": 4050
    },
    {
      "entropy": 5.468939256668091,
      "epoch": 4.129393785022924,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00037925108822904667,
      "loss": 4.9641,
      "mean_token_accuracy": 0.2095114693045616,
      "num_tokens": 9559045.0,
      "step": 4055
    },
    {
      "entropy": 5.324898719787598,
      "epoch": 4.134488028527763,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003788954148646112,
      "loss": 4.9964,
      "mean_token_accuracy": 0.2083497539162636,
      "num_tokens": 9571170.0,
      "step": 4060
    },
    {
      "entropy": 5.503489923477173,
      "epoch": 4.139582272032603,
      "grad_norm": 1.234375,
      "learning_rate": 0.00037853941121870964,
      "loss": 5.0176,
      "mean_token_accuracy": 0.2058577135205269,
      "num_tokens": 9583080.0,
      "step": 4065
    },
    {
      "entropy": 5.436334466934204,
      "epoch": 4.1446765155374425,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003781830784230704,
      "loss": 5.0985,
      "mean_token_accuracy": 0.1944021388888359,
      "num_tokens": 9595474.0,
      "step": 4070
    },
    {
      "entropy": 5.413578653335572,
      "epoch": 4.1497707590422825,
      "grad_norm": 1.109375,
      "learning_rate": 0.00037782641761046856,
      "loss": 5.0465,
      "mean_token_accuracy": 0.19963150173425676,
      "num_tokens": 9606950.0,
      "step": 4075
    },
    {
      "entropy": 5.532321643829346,
      "epoch": 4.154865002547122,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00037746942991472157,
      "loss": 5.1,
      "mean_token_accuracy": 0.19192057251930236,
      "num_tokens": 9619199.0,
      "step": 4080
    },
    {
      "entropy": 5.37316403388977,
      "epoch": 4.159959246051962,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00037711211647068637,
      "loss": 4.9792,
      "mean_token_accuracy": 0.20839075446128846,
      "num_tokens": 9631596.0,
      "step": 4085
    },
    {
      "entropy": 5.4479673385620115,
      "epoch": 4.165053489556801,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00037675447841425535,
      "loss": 5.0216,
      "mean_token_accuracy": 0.19856827557086945,
      "num_tokens": 9644610.0,
      "step": 4090
    },
    {
      "entropy": 5.36942310333252,
      "epoch": 4.170147733061641,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00037639651688235276,
      "loss": 5.0126,
      "mean_token_accuracy": 0.2051938995718956,
      "num_tokens": 9657359.0,
      "step": 4095
    },
    {
      "entropy": 5.401572132110596,
      "epoch": 4.17524197656648,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003760382330129312,
      "loss": 5.0422,
      "mean_token_accuracy": 0.20518056750297547,
      "num_tokens": 9670732.0,
      "step": 4100
    },
    {
      "entropy": 5.424186897277832,
      "epoch": 4.180336220071319,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000375679627944968,
      "loss": 4.9475,
      "mean_token_accuracy": 0.2099715694785118,
      "num_tokens": 9683421.0,
      "step": 4105
    },
    {
      "entropy": 5.3791584968566895,
      "epoch": 4.185430463576159,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003753207028184617,
      "loss": 4.9705,
      "mean_token_accuracy": 0.2038587972521782,
      "num_tokens": 9696696.0,
      "step": 4110
    },
    {
      "entropy": 5.4151777744293215,
      "epoch": 4.190524707080998,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000374961458774428,
      "loss": 4.9663,
      "mean_token_accuracy": 0.20452841073274614,
      "num_tokens": 9707380.0,
      "step": 4115
    },
    {
      "entropy": 5.401227760314941,
      "epoch": 4.195618950585838,
      "grad_norm": 1.25,
      "learning_rate": 0.0003746018969548966,
      "loss": 5.0348,
      "mean_token_accuracy": 0.19933833926916122,
      "num_tokens": 9718404.0,
      "step": 4120
    },
    {
      "entropy": 5.372776222229004,
      "epoch": 4.200713194090677,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003742420185029075,
      "loss": 5.0073,
      "mean_token_accuracy": 0.2062012583017349,
      "num_tokens": 9731230.0,
      "step": 4125
    },
    {
      "entropy": 5.450782823562622,
      "epoch": 4.205807437595517,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003738818245625072,
      "loss": 4.9037,
      "mean_token_accuracy": 0.21417510211467744,
      "num_tokens": 9742490.0,
      "step": 4130
    },
    {
      "entropy": 5.313555145263672,
      "epoch": 4.210901681100356,
      "grad_norm": 1.140625,
      "learning_rate": 0.000373521316278745,
      "loss": 4.8541,
      "mean_token_accuracy": 0.21353932917118074,
      "num_tokens": 9753504.0,
      "step": 4135
    },
    {
      "entropy": 5.3078985691070555,
      "epoch": 4.215995924605196,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003731604947976695,
      "loss": 4.9994,
      "mean_token_accuracy": 0.21068064868450165,
      "num_tokens": 9765914.0,
      "step": 4140
    },
    {
      "entropy": 5.527010869979859,
      "epoch": 4.2210901681100355,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003727993612663253,
      "loss": 5.0674,
      "mean_token_accuracy": 0.2022989347577095,
      "num_tokens": 9778324.0,
      "step": 4145
    },
    {
      "entropy": 5.392856025695801,
      "epoch": 4.2261844116148755,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00037243791683274865,
      "loss": 5.0347,
      "mean_token_accuracy": 0.20320393443107604,
      "num_tokens": 9789325.0,
      "step": 4150
    },
    {
      "entropy": 5.356120347976685,
      "epoch": 4.231278655119715,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003720761626459641,
      "loss": 4.9256,
      "mean_token_accuracy": 0.2127709984779358,
      "num_tokens": 9800969.0,
      "step": 4155
    },
    {
      "entropy": 5.357995128631591,
      "epoch": 4.236372898624555,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003717140998559812,
      "loss": 5.0202,
      "mean_token_accuracy": 0.2027714356780052,
      "num_tokens": 9812835.0,
      "step": 4160
    },
    {
      "entropy": 5.466751003265381,
      "epoch": 4.241467142129394,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00037135172961379035,
      "loss": 5.0925,
      "mean_token_accuracy": 0.1969348520040512,
      "num_tokens": 9825103.0,
      "step": 4165
    },
    {
      "entropy": 5.435714626312256,
      "epoch": 4.246561385634234,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003709890530713594,
      "loss": 4.9546,
      "mean_token_accuracy": 0.20436008721590043,
      "num_tokens": 9836733.0,
      "step": 4170
    },
    {
      "entropy": 5.368697690963745,
      "epoch": 4.251655629139073,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003706260713816298,
      "loss": 5.0202,
      "mean_token_accuracy": 0.1989650994539261,
      "num_tokens": 9848472.0,
      "step": 4175
    },
    {
      "entropy": 5.33016037940979,
      "epoch": 4.256749872643912,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00037026278569851325,
      "loss": 4.904,
      "mean_token_accuracy": 0.21770911663770676,
      "num_tokens": 9860131.0,
      "step": 4180
    },
    {
      "entropy": 5.386573743820191,
      "epoch": 4.261844116148752,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003698991971768877,
      "loss": 4.9921,
      "mean_token_accuracy": 0.20340700894594194,
      "num_tokens": 9871996.0,
      "step": 4185
    },
    {
      "entropy": 5.3544686794281,
      "epoch": 4.266938359653591,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003695353069725939,
      "loss": 4.9447,
      "mean_token_accuracy": 0.20257433950901033,
      "num_tokens": 9883131.0,
      "step": 4190
    },
    {
      "entropy": 5.429681634902954,
      "epoch": 4.272032603158431,
      "grad_norm": 1.234375,
      "learning_rate": 0.00036917111624243143,
      "loss": 4.9706,
      "mean_token_accuracy": 0.20149180740118028,
      "num_tokens": 9894305.0,
      "step": 4195
    },
    {
      "entropy": 5.359818983078003,
      "epoch": 4.27712684666327,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00036880662614415574,
      "loss": 4.9652,
      "mean_token_accuracy": 0.20834836065769197,
      "num_tokens": 9908095.0,
      "step": 4200
    },
    {
      "entropy": 5.422232961654663,
      "epoch": 4.28222109016811,
      "grad_norm": 1.109375,
      "learning_rate": 0.00036844183783647333,
      "loss": 5.083,
      "mean_token_accuracy": 0.20519161820411683,
      "num_tokens": 9920777.0,
      "step": 4205
    },
    {
      "entropy": 5.370787000656128,
      "epoch": 4.287315333672949,
      "grad_norm": 1.265625,
      "learning_rate": 0.00036807675247903915,
      "loss": 4.9383,
      "mean_token_accuracy": 0.21085598766803743,
      "num_tokens": 9930555.0,
      "step": 4210
    },
    {
      "entropy": 5.397698783874512,
      "epoch": 4.292409577177789,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003677113712324522,
      "loss": 4.9552,
      "mean_token_accuracy": 0.20250182151794432,
      "num_tokens": 9942770.0,
      "step": 4215
    },
    {
      "entropy": 5.37546443939209,
      "epoch": 4.297503820682628,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00036734569525825243,
      "loss": 4.9911,
      "mean_token_accuracy": 0.2008635252714157,
      "num_tokens": 9953939.0,
      "step": 4220
    },
    {
      "entropy": 5.429863452911377,
      "epoch": 4.302598064187468,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003669797257189163,
      "loss": 4.976,
      "mean_token_accuracy": 0.20772750079631805,
      "num_tokens": 9965887.0,
      "step": 4225
    },
    {
      "entropy": 5.310213661193847,
      "epoch": 4.3076923076923075,
      "grad_norm": 1.25,
      "learning_rate": 0.00036661346377785397,
      "loss": 4.9872,
      "mean_token_accuracy": 0.2070951998233795,
      "num_tokens": 9976897.0,
      "step": 4230
    },
    {
      "entropy": 5.3396178722381595,
      "epoch": 4.3127865511971475,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003662469105994046,
      "loss": 4.9885,
      "mean_token_accuracy": 0.20979513823986054,
      "num_tokens": 9988592.0,
      "step": 4235
    },
    {
      "entropy": 5.429060792922973,
      "epoch": 4.317880794701987,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00036588006734883377,
      "loss": 4.9895,
      "mean_token_accuracy": 0.20164374858140946,
      "num_tokens": 10000014.0,
      "step": 4240
    },
    {
      "entropy": 5.40374641418457,
      "epoch": 4.322975038206827,
      "grad_norm": 1.21875,
      "learning_rate": 0.00036551293519232865,
      "loss": 4.9908,
      "mean_token_accuracy": 0.20438784658908843,
      "num_tokens": 10009867.0,
      "step": 4245
    },
    {
      "entropy": 5.340420484542847,
      "epoch": 4.328069281711666,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003651455152969954,
      "loss": 4.9687,
      "mean_token_accuracy": 0.20756426602602004,
      "num_tokens": 10021132.0,
      "step": 4250
    },
    {
      "entropy": 5.388458347320556,
      "epoch": 4.333163525216506,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00036477780883085423,
      "loss": 5.0096,
      "mean_token_accuracy": 0.20435259491205215,
      "num_tokens": 10033075.0,
      "step": 4255
    },
    {
      "entropy": 5.4534299850463865,
      "epoch": 4.338257768721345,
      "grad_norm": 1.328125,
      "learning_rate": 0.00036440981696283695,
      "loss": 4.9856,
      "mean_token_accuracy": 0.20024957805871962,
      "num_tokens": 10043043.0,
      "step": 4260
    },
    {
      "entropy": 5.3612730503082275,
      "epoch": 4.343352012226184,
      "grad_norm": 1.15625,
      "learning_rate": 0.00036404154086278236,
      "loss": 4.9679,
      "mean_token_accuracy": 0.21037161350250244,
      "num_tokens": 10055292.0,
      "step": 4265
    },
    {
      "entropy": 5.437903165817261,
      "epoch": 4.348446255731024,
      "grad_norm": 1.25,
      "learning_rate": 0.000363672981701433,
      "loss": 5.0337,
      "mean_token_accuracy": 0.20233087986707687,
      "num_tokens": 10066258.0,
      "step": 4270
    },
    {
      "entropy": 5.362940311431885,
      "epoch": 4.353540499235863,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00036330414065043087,
      "loss": 5.0386,
      "mean_token_accuracy": 0.20331443697214127,
      "num_tokens": 10079200.0,
      "step": 4275
    },
    {
      "entropy": 5.356106233596802,
      "epoch": 4.358634742740703,
      "grad_norm": 1.125,
      "learning_rate": 0.00036293501888231465,
      "loss": 4.9689,
      "mean_token_accuracy": 0.2081938624382019,
      "num_tokens": 10091600.0,
      "step": 4280
    },
    {
      "entropy": 5.398920392990112,
      "epoch": 4.363728986245542,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00036256561757051493,
      "loss": 4.9283,
      "mean_token_accuracy": 0.20747285187244416,
      "num_tokens": 10102700.0,
      "step": 4285
    },
    {
      "entropy": 5.35002818107605,
      "epoch": 4.368823229750382,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003621959378893514,
      "loss": 4.9063,
      "mean_token_accuracy": 0.20988034904003144,
      "num_tokens": 10112953.0,
      "step": 4290
    },
    {
      "entropy": 5.319896984100342,
      "epoch": 4.373917473255221,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003618259810140281,
      "loss": 4.9079,
      "mean_token_accuracy": 0.20971768647432326,
      "num_tokens": 10123027.0,
      "step": 4295
    },
    {
      "entropy": 5.32706389427185,
      "epoch": 4.379011716760061,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00036145574812063093,
      "loss": 4.9056,
      "mean_token_accuracy": 0.21398795247077942,
      "num_tokens": 10134796.0,
      "step": 4300
    },
    {
      "entropy": 5.429208135604858,
      "epoch": 4.3841059602649,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003610852403861227,
      "loss": 5.1327,
      "mean_token_accuracy": 0.19812282025814057,
      "num_tokens": 10147539.0,
      "step": 4305
    },
    {
      "entropy": 5.333648061752319,
      "epoch": 4.38920020376974,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00036071445898834043,
      "loss": 4.9463,
      "mean_token_accuracy": 0.2111382380127907,
      "num_tokens": 10158755.0,
      "step": 4310
    },
    {
      "entropy": 5.398270940780639,
      "epoch": 4.3942944472745795,
      "grad_norm": 1.203125,
      "learning_rate": 0.00036034340510599053,
      "loss": 5.0256,
      "mean_token_accuracy": 0.20548568069934844,
      "num_tokens": 10170800.0,
      "step": 4315
    },
    {
      "entropy": 5.342477130889892,
      "epoch": 4.3993886907794195,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00035997207991864604,
      "loss": 4.9986,
      "mean_token_accuracy": 0.20531322807073593,
      "num_tokens": 10182088.0,
      "step": 4320
    },
    {
      "entropy": 5.438752174377441,
      "epoch": 4.404482934284259,
      "grad_norm": 1.15625,
      "learning_rate": 0.00035960048460674244,
      "loss": 4.9925,
      "mean_token_accuracy": 0.20491478890180587,
      "num_tokens": 10193390.0,
      "step": 4325
    },
    {
      "entropy": 5.337182426452637,
      "epoch": 4.409577177789099,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00035922862035157394,
      "loss": 5.0071,
      "mean_token_accuracy": 0.2041212409734726,
      "num_tokens": 10205729.0,
      "step": 4330
    },
    {
      "entropy": 5.384648036956787,
      "epoch": 4.414671421293938,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003588564883352895,
      "loss": 5.0214,
      "mean_token_accuracy": 0.20601855516433715,
      "num_tokens": 10217744.0,
      "step": 4335
    },
    {
      "entropy": 5.33169264793396,
      "epoch": 4.419765664798778,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003584840897408895,
      "loss": 4.933,
      "mean_token_accuracy": 0.21310179978609084,
      "num_tokens": 10228444.0,
      "step": 4340
    },
    {
      "entropy": 5.406608963012696,
      "epoch": 4.424859908303617,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003581114257522216,
      "loss": 5.0764,
      "mean_token_accuracy": 0.20339810699224473,
      "num_tokens": 10240786.0,
      "step": 4345
    },
    {
      "entropy": 5.390498924255371,
      "epoch": 4.429954151808456,
      "grad_norm": 1.1875,
      "learning_rate": 0.00035773849755397737,
      "loss": 4.9134,
      "mean_token_accuracy": 0.21342567056417466,
      "num_tokens": 10252317.0,
      "step": 4350
    },
    {
      "entropy": 5.321036005020142,
      "epoch": 4.435048395313296,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00035736530633168805,
      "loss": 4.9665,
      "mean_token_accuracy": 0.21715394407510757,
      "num_tokens": 10264075.0,
      "step": 4355
    },
    {
      "entropy": 5.357851457595825,
      "epoch": 4.440142638818136,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003569918532717212,
      "loss": 4.9104,
      "mean_token_accuracy": 0.212430901825428,
      "num_tokens": 10274696.0,
      "step": 4360
    },
    {
      "entropy": 5.3572508811950685,
      "epoch": 4.445236882322975,
      "grad_norm": 1.203125,
      "learning_rate": 0.00035661813956127677,
      "loss": 5.0531,
      "mean_token_accuracy": 0.1989592954516411,
      "num_tokens": 10285579.0,
      "step": 4365
    },
    {
      "entropy": 5.347762107849121,
      "epoch": 4.450331125827814,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00035624416638838313,
      "loss": 4.9696,
      "mean_token_accuracy": 0.199952195584774,
      "num_tokens": 10296624.0,
      "step": 4370
    },
    {
      "entropy": 5.374545621871948,
      "epoch": 4.455425369332654,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003558699349418937,
      "loss": 5.0254,
      "mean_token_accuracy": 0.20042396187782288,
      "num_tokens": 10308086.0,
      "step": 4375
    },
    {
      "entropy": 5.409057855606079,
      "epoch": 4.460519612837493,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003554954464114827,
      "loss": 4.9886,
      "mean_token_accuracy": 0.20829138904809952,
      "num_tokens": 10320669.0,
      "step": 4380
    },
    {
      "entropy": 5.353666496276856,
      "epoch": 4.465613856342333,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00035512070198764185,
      "loss": 4.9883,
      "mean_token_accuracy": 0.20981950610876082,
      "num_tokens": 10331126.0,
      "step": 4385
    },
    {
      "entropy": 5.334564208984375,
      "epoch": 4.470708099847172,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00035474570286167626,
      "loss": 5.0549,
      "mean_token_accuracy": 0.20117305368185043,
      "num_tokens": 10343183.0,
      "step": 4390
    },
    {
      "entropy": 5.4526856422424315,
      "epoch": 4.475802343352012,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003543704502257006,
      "loss": 4.9909,
      "mean_token_accuracy": 0.2080983489751816,
      "num_tokens": 10354926.0,
      "step": 4395
    },
    {
      "entropy": 5.325784587860108,
      "epoch": 4.4808965868568515,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00035399494527263566,
      "loss": 4.974,
      "mean_token_accuracy": 0.20877518951892854,
      "num_tokens": 10366371.0,
      "step": 4400
    },
    {
      "entropy": 5.3800599575042725,
      "epoch": 4.4859908303616916,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003536191891962043,
      "loss": 4.9756,
      "mean_token_accuracy": 0.20993165224790572,
      "num_tokens": 10377207.0,
      "step": 4405
    },
    {
      "entropy": 5.374589252471924,
      "epoch": 4.491085073866531,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003532431831909276,
      "loss": 4.9279,
      "mean_token_accuracy": 0.2114395722746849,
      "num_tokens": 10388214.0,
      "step": 4410
    },
    {
      "entropy": 5.328938007354736,
      "epoch": 4.496179317371371,
      "grad_norm": 1.109375,
      "learning_rate": 0.00035286692845212114,
      "loss": 5.0178,
      "mean_token_accuracy": 0.20434480309486389,
      "num_tokens": 10400421.0,
      "step": 4415
    },
    {
      "entropy": 5.3085243701934814,
      "epoch": 4.50127356087621,
      "grad_norm": 1.09375,
      "learning_rate": 0.00035249042617589135,
      "loss": 5.0074,
      "mean_token_accuracy": 0.1984187737107277,
      "num_tokens": 10412693.0,
      "step": 4420
    },
    {
      "entropy": 5.449683380126953,
      "epoch": 4.50636780438105,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003521136775591315,
      "loss": 5.0287,
      "mean_token_accuracy": 0.19977275729179383,
      "num_tokens": 10424706.0,
      "step": 4425
    },
    {
      "entropy": 5.31883430480957,
      "epoch": 4.511462047885889,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000351736683799518,
      "loss": 4.937,
      "mean_token_accuracy": 0.2060149982571602,
      "num_tokens": 10435965.0,
      "step": 4430
    },
    {
      "entropy": 5.365731143951416,
      "epoch": 4.516556291390728,
      "grad_norm": 1.21875,
      "learning_rate": 0.00035135944609550654,
      "loss": 5.025,
      "mean_token_accuracy": 0.2066955655813217,
      "num_tokens": 10447347.0,
      "step": 4435
    },
    {
      "entropy": 5.350469875335693,
      "epoch": 4.521650534895568,
      "grad_norm": 1.125,
      "learning_rate": 0.00035098196564632825,
      "loss": 4.9045,
      "mean_token_accuracy": 0.20735154598951339,
      "num_tokens": 10459916.0,
      "step": 4440
    },
    {
      "entropy": 5.296450805664063,
      "epoch": 4.526744778400408,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00035060424365198603,
      "loss": 4.9285,
      "mean_token_accuracy": 0.20855508893728256,
      "num_tokens": 10471337.0,
      "step": 4445
    },
    {
      "entropy": 5.362928295135498,
      "epoch": 4.531839021905247,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003502262813132507,
      "loss": 4.9908,
      "mean_token_accuracy": 0.20859081894159318,
      "num_tokens": 10484130.0,
      "step": 4450
    },
    {
      "entropy": 5.327928590774536,
      "epoch": 4.536933265410086,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00034984807983165695,
      "loss": 4.9999,
      "mean_token_accuracy": 0.20783299058675767,
      "num_tokens": 10496812.0,
      "step": 4455
    },
    {
      "entropy": 5.426850605010986,
      "epoch": 4.542027508914926,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003494696404094998,
      "loss": 5.1099,
      "mean_token_accuracy": 0.1878078833222389,
      "num_tokens": 10509544.0,
      "step": 4460
    },
    {
      "entropy": 5.310198259353638,
      "epoch": 4.547121752419765,
      "grad_norm": 1.09375,
      "learning_rate": 0.00034909096424983073,
      "loss": 4.9929,
      "mean_token_accuracy": 0.20972779095172883,
      "num_tokens": 10522468.0,
      "step": 4465
    },
    {
      "entropy": 5.3340620517730715,
      "epoch": 4.552215995924605,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003487120525564537,
      "loss": 5.0465,
      "mean_token_accuracy": 0.2064715340733528,
      "num_tokens": 10534177.0,
      "step": 4470
    },
    {
      "entropy": 5.378270673751831,
      "epoch": 4.5573102394294445,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003483329065339216,
      "loss": 4.9626,
      "mean_token_accuracy": 0.20385078489780425,
      "num_tokens": 10546568.0,
      "step": 4475
    },
    {
      "entropy": 5.416376638412475,
      "epoch": 4.5624044829342845,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003479535273875319,
      "loss": 5.0282,
      "mean_token_accuracy": 0.1990809068083763,
      "num_tokens": 10558398.0,
      "step": 4480
    },
    {
      "entropy": 5.376024913787842,
      "epoch": 4.567498726439124,
      "grad_norm": 1.265625,
      "learning_rate": 0.00034757391632332364,
      "loss": 5.0466,
      "mean_token_accuracy": 0.20090211033821107,
      "num_tokens": 10570601.0,
      "step": 4485
    },
    {
      "entropy": 5.3625092029571535,
      "epoch": 4.572592969943964,
      "grad_norm": 1.28125,
      "learning_rate": 0.00034719407454807274,
      "loss": 4.9844,
      "mean_token_accuracy": 0.2020719975233078,
      "num_tokens": 10581745.0,
      "step": 4490
    },
    {
      "entropy": 5.3319947719573975,
      "epoch": 4.577687213448803,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003468140032692887,
      "loss": 4.9853,
      "mean_token_accuracy": 0.20609573423862457,
      "num_tokens": 10591881.0,
      "step": 4495
    },
    {
      "entropy": 5.436495113372803,
      "epoch": 4.582781456953643,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003464337036952105,
      "loss": 5.0683,
      "mean_token_accuracy": 0.2017616868019104,
      "num_tokens": 10602849.0,
      "step": 4500
    },
    {
      "epoch": 4.582781456953643,
      "eval_entropy": 5.257230782214506,
      "eval_loss": 5.712556838989258,
      "eval_mean_token_accuracy": 0.17621020935935738,
      "eval_num_tokens": 10602849.0,
      "eval_runtime": 2.6695,
      "eval_samples_per_second": 1213.691,
      "eval_steps_per_second": 151.711,
      "step": 4500
    },
    {
      "entropy": 5.30481481552124,
      "epoch": 4.587875700458482,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00034605317703480313,
      "loss": 4.9001,
      "mean_token_accuracy": 0.22328849583864213,
      "num_tokens": 10613616.0,
      "step": 4505
    },
    {
      "entropy": 5.399619293212891,
      "epoch": 4.592969943963322,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003456724244977533,
      "loss": 5.1089,
      "mean_token_accuracy": 0.19515899419784546,
      "num_tokens": 10625308.0,
      "step": 4510
    },
    {
      "entropy": 5.284378576278686,
      "epoch": 4.598064187468161,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003452914472944656,
      "loss": 4.9366,
      "mean_token_accuracy": 0.21729601919651031,
      "num_tokens": 10637071.0,
      "step": 4515
    },
    {
      "entropy": 5.328333139419556,
      "epoch": 4.603158430973,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003449102466360591,
      "loss": 4.8413,
      "mean_token_accuracy": 0.21691457331180572,
      "num_tokens": 10647681.0,
      "step": 4520
    },
    {
      "entropy": 5.312632608413696,
      "epoch": 4.60825267447784,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003445288237343632,
      "loss": 5.0102,
      "mean_token_accuracy": 0.20289668887853624,
      "num_tokens": 10659789.0,
      "step": 4525
    },
    {
      "entropy": 5.369935655593872,
      "epoch": 4.61334691798268,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00034414717980191375,
      "loss": 5.0222,
      "mean_token_accuracy": 0.20245738029479982,
      "num_tokens": 10671435.0,
      "step": 4530
    },
    {
      "entropy": 5.496319961547852,
      "epoch": 4.618441161487519,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00034376531605194915,
      "loss": 5.1359,
      "mean_token_accuracy": 0.19581755101680756,
      "num_tokens": 10684693.0,
      "step": 4535
    },
    {
      "entropy": 5.290979528427124,
      "epoch": 4.623535404992358,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003433832336984068,
      "loss": 4.8805,
      "mean_token_accuracy": 0.21655526310205458,
      "num_tokens": 10695069.0,
      "step": 4540
    },
    {
      "entropy": 5.277418994903565,
      "epoch": 4.628629648497198,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003430009339559189,
      "loss": 5.0104,
      "mean_token_accuracy": 0.20975907593965532,
      "num_tokens": 10706108.0,
      "step": 4545
    },
    {
      "entropy": 5.390916204452514,
      "epoch": 4.633723892002037,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003426184180398087,
      "loss": 4.9765,
      "mean_token_accuracy": 0.21222971975803376,
      "num_tokens": 10717167.0,
      "step": 4550
    },
    {
      "entropy": 5.335337114334107,
      "epoch": 4.638818135506877,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003422356871660868,
      "loss": 4.9979,
      "mean_token_accuracy": 0.2155989021062851,
      "num_tokens": 10729376.0,
      "step": 4555
    },
    {
      "entropy": 5.378046464920044,
      "epoch": 4.6439123790117165,
      "grad_norm": 1.1875,
      "learning_rate": 0.00034185274255144705,
      "loss": 4.9972,
      "mean_token_accuracy": 0.20768585950136184,
      "num_tokens": 10741147.0,
      "step": 4560
    },
    {
      "entropy": 5.357807207107544,
      "epoch": 4.6490066225165565,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003414695854132629,
      "loss": 5.0147,
      "mean_token_accuracy": 0.2081143468618393,
      "num_tokens": 10752890.0,
      "step": 4565
    },
    {
      "entropy": 5.428834342956543,
      "epoch": 4.654100866021396,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003410862169695831,
      "loss": 5.0153,
      "mean_token_accuracy": 0.20666087716817855,
      "num_tokens": 10764681.0,
      "step": 4570
    },
    {
      "entropy": 5.366861009597779,
      "epoch": 4.659195109526236,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003407026384391284,
      "loss": 5.0226,
      "mean_token_accuracy": 0.20604213923215867,
      "num_tokens": 10776328.0,
      "step": 4575
    },
    {
      "entropy": 5.343188095092773,
      "epoch": 4.664289353031075,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003403188510412874,
      "loss": 5.0712,
      "mean_token_accuracy": 0.20427639037370682,
      "num_tokens": 10788488.0,
      "step": 4580
    },
    {
      "entropy": 5.431555891036988,
      "epoch": 4.669383596535915,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00033993485599611266,
      "loss": 5.0498,
      "mean_token_accuracy": 0.20569992661476136,
      "num_tokens": 10800232.0,
      "step": 4585
    },
    {
      "entropy": 5.412730884552002,
      "epoch": 4.674477840040754,
      "grad_norm": 1.046875,
      "learning_rate": 0.00033955065452431684,
      "loss": 5.102,
      "mean_token_accuracy": 0.1947254553437233,
      "num_tokens": 10813382.0,
      "step": 4590
    },
    {
      "entropy": 5.387947082519531,
      "epoch": 4.679572083545594,
      "grad_norm": 1.171875,
      "learning_rate": 0.00033916624784726865,
      "loss": 5.0743,
      "mean_token_accuracy": 0.20092709362506866,
      "num_tokens": 10825897.0,
      "step": 4595
    },
    {
      "entropy": 5.332760334014893,
      "epoch": 4.684666327050433,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00033878163718698945,
      "loss": 4.9532,
      "mean_token_accuracy": 0.20995809584856034,
      "num_tokens": 10837798.0,
      "step": 4600
    },
    {
      "entropy": 5.38763689994812,
      "epoch": 4.689760570555272,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00033839682376614875,
      "loss": 5.0398,
      "mean_token_accuracy": 0.20329337418079377,
      "num_tokens": 10850874.0,
      "step": 4605
    },
    {
      "entropy": 5.436527919769287,
      "epoch": 4.694854814060112,
      "grad_norm": 1.109375,
      "learning_rate": 0.00033801180880806115,
      "loss": 5.0176,
      "mean_token_accuracy": 0.203446926176548,
      "num_tokens": 10862964.0,
      "step": 4610
    },
    {
      "entropy": 5.444200563430786,
      "epoch": 4.699949057564952,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003376265935366812,
      "loss": 5.0335,
      "mean_token_accuracy": 0.20009576082229613,
      "num_tokens": 10876168.0,
      "step": 4615
    },
    {
      "entropy": 5.419836711883545,
      "epoch": 4.705043301069791,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003372411791766008,
      "loss": 4.9794,
      "mean_token_accuracy": 0.20532263368368148,
      "num_tokens": 10887990.0,
      "step": 4620
    },
    {
      "entropy": 5.336600494384766,
      "epoch": 4.71013754457463,
      "grad_norm": 1.15625,
      "learning_rate": 0.00033685556695304446,
      "loss": 4.9363,
      "mean_token_accuracy": 0.20956223756074904,
      "num_tokens": 10899315.0,
      "step": 4625
    },
    {
      "entropy": 5.301969718933106,
      "epoch": 4.71523178807947,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00033646975809186597,
      "loss": 4.9917,
      "mean_token_accuracy": 0.20807337164878845,
      "num_tokens": 10910704.0,
      "step": 4630
    },
    {
      "entropy": 5.3665611267089846,
      "epoch": 4.720326031584309,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003360837538195439,
      "loss": 5.0213,
      "mean_token_accuracy": 0.20517987608909607,
      "num_tokens": 10922503.0,
      "step": 4635
    },
    {
      "entropy": 5.389917469024658,
      "epoch": 4.725420275089149,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003356975553631782,
      "loss": 5.0473,
      "mean_token_accuracy": 0.20488539934158326,
      "num_tokens": 10933872.0,
      "step": 4640
    },
    {
      "entropy": 5.334551668167114,
      "epoch": 4.7305145185939885,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00033531116395048614,
      "loss": 4.9865,
      "mean_token_accuracy": 0.20436448305845262,
      "num_tokens": 10945268.0,
      "step": 4645
    },
    {
      "entropy": 5.3949168682098385,
      "epoch": 4.7356087620988285,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003349245808097985,
      "loss": 5.0593,
      "mean_token_accuracy": 0.2046230301260948,
      "num_tokens": 10956714.0,
      "step": 4650
    },
    {
      "entropy": 5.426679801940918,
      "epoch": 4.740703005603668,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003345378071700553,
      "loss": 5.034,
      "mean_token_accuracy": 0.20087760835886,
      "num_tokens": 10969845.0,
      "step": 4655
    },
    {
      "entropy": 5.41066198348999,
      "epoch": 4.745797249108508,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003341508442608022,
      "loss": 4.978,
      "mean_token_accuracy": 0.2062778115272522,
      "num_tokens": 10981103.0,
      "step": 4660
    },
    {
      "entropy": 5.337995290756226,
      "epoch": 4.750891492613347,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003337636933121867,
      "loss": 5.043,
      "mean_token_accuracy": 0.2054070994257927,
      "num_tokens": 10992328.0,
      "step": 4665
    },
    {
      "entropy": 5.452453660964966,
      "epoch": 4.755985736118187,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003333763555549541,
      "loss": 5.1345,
      "mean_token_accuracy": 0.18243547976017,
      "num_tokens": 11002696.0,
      "step": 4670
    },
    {
      "entropy": 5.3764472007751465,
      "epoch": 4.761079979623026,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003329888322204434,
      "loss": 4.9994,
      "mean_token_accuracy": 0.2046897977590561,
      "num_tokens": 11014786.0,
      "step": 4675
    },
    {
      "entropy": 5.312127685546875,
      "epoch": 4.766174223127866,
      "grad_norm": 0.953125,
      "learning_rate": 0.00033260112454058344,
      "loss": 4.969,
      "mean_token_accuracy": 0.2068356156349182,
      "num_tokens": 11026586.0,
      "step": 4680
    },
    {
      "entropy": 5.383845233917237,
      "epoch": 4.771268466632705,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00033221323374788946,
      "loss": 4.9885,
      "mean_token_accuracy": 0.20858259052038192,
      "num_tokens": 11037234.0,
      "step": 4685
    },
    {
      "entropy": 5.411475372314453,
      "epoch": 4.776362710137544,
      "grad_norm": 1.09375,
      "learning_rate": 0.00033182516107545864,
      "loss": 5.0494,
      "mean_token_accuracy": 0.20333594381809234,
      "num_tokens": 11049963.0,
      "step": 4690
    },
    {
      "entropy": 5.43092360496521,
      "epoch": 4.781456953642384,
      "grad_norm": 1.203125,
      "learning_rate": 0.00033143690775696606,
      "loss": 5.0755,
      "mean_token_accuracy": 0.1943735212087631,
      "num_tokens": 11062679.0,
      "step": 4695
    },
    {
      "entropy": 5.398403167724609,
      "epoch": 4.786551197147224,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003310484750266616,
      "loss": 4.9886,
      "mean_token_accuracy": 0.20380942672491073,
      "num_tokens": 11073806.0,
      "step": 4700
    },
    {
      "entropy": 5.398754978179932,
      "epoch": 4.791645440652063,
      "grad_norm": 1.140625,
      "learning_rate": 0.00033065986411936506,
      "loss": 5.036,
      "mean_token_accuracy": 0.2010859191417694,
      "num_tokens": 11086091.0,
      "step": 4705
    },
    {
      "entropy": 5.358682584762573,
      "epoch": 4.796739684156902,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003302710762704629,
      "loss": 5.0152,
      "mean_token_accuracy": 0.19859752506017686,
      "num_tokens": 11098223.0,
      "step": 4710
    },
    {
      "entropy": 5.332942342758178,
      "epoch": 4.801833927661742,
      "grad_norm": 1.046875,
      "learning_rate": 0.000329882112715904,
      "loss": 5.0101,
      "mean_token_accuracy": 0.20655299723148346,
      "num_tokens": 11111402.0,
      "step": 4715
    },
    {
      "entropy": 5.295500373840332,
      "epoch": 4.806928171166581,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00032949297469219563,
      "loss": 4.9171,
      "mean_token_accuracy": 0.21271281987428664,
      "num_tokens": 11122449.0,
      "step": 4720
    },
    {
      "entropy": 5.3289635181427,
      "epoch": 4.8120224146714214,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00032910366343639997,
      "loss": 5.0022,
      "mean_token_accuracy": 0.20524270832538605,
      "num_tokens": 11134149.0,
      "step": 4725
    },
    {
      "entropy": 5.368810701370239,
      "epoch": 4.817116658176261,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00032871418018612993,
      "loss": 4.9849,
      "mean_token_accuracy": 0.2120379701256752,
      "num_tokens": 11145951.0,
      "step": 4730
    },
    {
      "entropy": 5.3084629535675045,
      "epoch": 4.822210901681101,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003283245261795448,
      "loss": 4.9472,
      "mean_token_accuracy": 0.20657918453216553,
      "num_tokens": 11156204.0,
      "step": 4735
    },
    {
      "entropy": 5.408049058914185,
      "epoch": 4.82730514518594,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003279347026553472,
      "loss": 4.971,
      "mean_token_accuracy": 0.2075286939740181,
      "num_tokens": 11167021.0,
      "step": 4740
    },
    {
      "entropy": 5.421921730041504,
      "epoch": 4.83239938869078,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003275447108527782,
      "loss": 5.0152,
      "mean_token_accuracy": 0.20384640991687775,
      "num_tokens": 11179433.0,
      "step": 4745
    },
    {
      "entropy": 5.300720453262329,
      "epoch": 4.837493632195619,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003271545520116141,
      "loss": 4.9732,
      "mean_token_accuracy": 0.20813059508800508,
      "num_tokens": 11190076.0,
      "step": 4750
    },
    {
      "entropy": 5.4480509757995605,
      "epoch": 4.842587875700459,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003267642273721621,
      "loss": 5.1318,
      "mean_token_accuracy": 0.1924700230360031,
      "num_tokens": 11201838.0,
      "step": 4755
    },
    {
      "entropy": 5.4072741031646725,
      "epoch": 4.847682119205298,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00032637373817525657,
      "loss": 5.0069,
      "mean_token_accuracy": 0.205011048913002,
      "num_tokens": 11214640.0,
      "step": 4760
    },
    {
      "entropy": 5.326920795440674,
      "epoch": 4.852776362710138,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003259830856622547,
      "loss": 5.0557,
      "mean_token_accuracy": 0.20030544102191924,
      "num_tokens": 11227447.0,
      "step": 4765
    },
    {
      "entropy": 5.416737079620361,
      "epoch": 4.857870606214977,
      "grad_norm": 1.28125,
      "learning_rate": 0.00032559227107503344,
      "loss": 5.0282,
      "mean_token_accuracy": 0.20018671900033952,
      "num_tokens": 11237681.0,
      "step": 4770
    },
    {
      "entropy": 5.367226791381836,
      "epoch": 4.862964849719816,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00032520129565598435,
      "loss": 5.0765,
      "mean_token_accuracy": 0.1967522084712982,
      "num_tokens": 11249672.0,
      "step": 4775
    },
    {
      "entropy": 5.410008430480957,
      "epoch": 4.868059093224656,
      "grad_norm": 0.9375,
      "learning_rate": 0.0003248101606480106,
      "loss": 5.0361,
      "mean_token_accuracy": 0.20303404629230498,
      "num_tokens": 11262378.0,
      "step": 4780
    },
    {
      "entropy": 5.372464370727539,
      "epoch": 4.873153336729496,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003244188672945228,
      "loss": 5.033,
      "mean_token_accuracy": 0.2014129027724266,
      "num_tokens": 11275024.0,
      "step": 4785
    },
    {
      "entropy": 5.34934630393982,
      "epoch": 4.878247580234335,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00032402741683943456,
      "loss": 5.0875,
      "mean_token_accuracy": 0.19534129798412322,
      "num_tokens": 11286312.0,
      "step": 4790
    },
    {
      "entropy": 5.315851449966431,
      "epoch": 4.883341823739174,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003236358105271593,
      "loss": 4.9527,
      "mean_token_accuracy": 0.21532251089811325,
      "num_tokens": 11297905.0,
      "step": 4795
    },
    {
      "entropy": 5.351438808441162,
      "epoch": 4.888436067244014,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003232440496026055,
      "loss": 4.962,
      "mean_token_accuracy": 0.20979333519935608,
      "num_tokens": 11311209.0,
      "step": 4800
    },
    {
      "entropy": 5.347570896148682,
      "epoch": 4.8935303107488535,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003228521353111736,
      "loss": 4.9601,
      "mean_token_accuracy": 0.20920209884643554,
      "num_tokens": 11321345.0,
      "step": 4805
    },
    {
      "entropy": 5.2763307094573975,
      "epoch": 4.8986245542536935,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003224600688987513,
      "loss": 4.9135,
      "mean_token_accuracy": 0.21334250122308732,
      "num_tokens": 11332720.0,
      "step": 4810
    },
    {
      "entropy": 5.345761823654175,
      "epoch": 4.903718797758533,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003220678516117099,
      "loss": 4.991,
      "mean_token_accuracy": 0.2128927379846573,
      "num_tokens": 11344662.0,
      "step": 4815
    },
    {
      "entropy": 5.3524449348449705,
      "epoch": 4.908813041263373,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00032167548469690045,
      "loss": 5.0513,
      "mean_token_accuracy": 0.19952254146337509,
      "num_tokens": 11356807.0,
      "step": 4820
    },
    {
      "entropy": 5.380684423446655,
      "epoch": 4.913907284768212,
      "grad_norm": 1.25,
      "learning_rate": 0.0003212829694016496,
      "loss": 5.0881,
      "mean_token_accuracy": 0.20185190439224243,
      "num_tokens": 11368853.0,
      "step": 4825
    },
    {
      "entropy": 5.440235376358032,
      "epoch": 4.919001528273052,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003208903069737557,
      "loss": 5.0794,
      "mean_token_accuracy": 0.20006583780050277,
      "num_tokens": 11381062.0,
      "step": 4830
    },
    {
      "entropy": 5.3525145053863525,
      "epoch": 4.924095771777891,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0003204974986614848,
      "loss": 5.0152,
      "mean_token_accuracy": 0.2055067762732506,
      "num_tokens": 11393831.0,
      "step": 4835
    },
    {
      "entropy": 5.274740409851074,
      "epoch": 4.929190015282731,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00032010454571356677,
      "loss": 5.003,
      "mean_token_accuracy": 0.2139187440276146,
      "num_tokens": 11405125.0,
      "step": 4840
    },
    {
      "entropy": 5.367819547653198,
      "epoch": 4.93428425878757,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003197114493791913,
      "loss": 4.955,
      "mean_token_accuracy": 0.21093281954526902,
      "num_tokens": 11416318.0,
      "step": 4845
    },
    {
      "entropy": 5.348317432403564,
      "epoch": 4.93937850229241,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00031931821090800365,
      "loss": 5.0436,
      "mean_token_accuracy": 0.20141559094190598,
      "num_tokens": 11428360.0,
      "step": 4850
    },
    {
      "entropy": 5.362545537948608,
      "epoch": 4.944472745797249,
      "grad_norm": 1.234375,
      "learning_rate": 0.00031892483155010145,
      "loss": 4.9665,
      "mean_token_accuracy": 0.2089117258787155,
      "num_tokens": 11440274.0,
      "step": 4855
    },
    {
      "entropy": 5.318439579010009,
      "epoch": 4.949566989302088,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003185313125560295,
      "loss": 4.9805,
      "mean_token_accuracy": 0.21013960540294646,
      "num_tokens": 11453393.0,
      "step": 4860
    },
    {
      "entropy": 5.382112789154053,
      "epoch": 4.954661232806928,
      "grad_norm": 1.046875,
      "learning_rate": 0.00031813765517677727,
      "loss": 5.0682,
      "mean_token_accuracy": 0.19686328619718552,
      "num_tokens": 11465944.0,
      "step": 4865
    },
    {
      "entropy": 5.309226751327515,
      "epoch": 4.959755476311768,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003177438606637736,
      "loss": 4.9072,
      "mean_token_accuracy": 0.2188994526863098,
      "num_tokens": 11478312.0,
      "step": 4870
    },
    {
      "entropy": 5.397652101516724,
      "epoch": 4.964849719816607,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003173499302688834,
      "loss": 5.0619,
      "mean_token_accuracy": 0.19806985855102538,
      "num_tokens": 11489807.0,
      "step": 4875
    },
    {
      "entropy": 5.351362323760986,
      "epoch": 4.969943963321446,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00031695586524440364,
      "loss": 5.0297,
      "mean_token_accuracy": 0.20302100479602814,
      "num_tokens": 11501348.0,
      "step": 4880
    },
    {
      "entropy": 5.313244581222534,
      "epoch": 4.975038206826286,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003165616668430592,
      "loss": 4.9449,
      "mean_token_accuracy": 0.21231851428747178,
      "num_tokens": 11513561.0,
      "step": 4885
    },
    {
      "entropy": 5.307394647598267,
      "epoch": 4.9801324503311255,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00031616733631799907,
      "loss": 4.954,
      "mean_token_accuracy": 0.21287191659212112,
      "num_tokens": 11524368.0,
      "step": 4890
    },
    {
      "entropy": 5.343631219863892,
      "epoch": 4.9852266938359655,
      "grad_norm": 1.078125,
      "learning_rate": 0.00031577287492279217,
      "loss": 5.0242,
      "mean_token_accuracy": 0.1955944761633873,
      "num_tokens": 11536557.0,
      "step": 4895
    },
    {
      "entropy": 5.37512903213501,
      "epoch": 4.990320937340805,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003153782839114233,
      "loss": 5.0318,
      "mean_token_accuracy": 0.20586279183626174,
      "num_tokens": 11549138.0,
      "step": 4900
    },
    {
      "entropy": 5.366048860549927,
      "epoch": 4.995415180845645,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00031498356453828967,
      "loss": 5.0526,
      "mean_token_accuracy": 0.19955099523067474,
      "num_tokens": 11560887.0,
      "step": 4905
    },
    {
      "entropy": 5.38076729244656,
      "epoch": 5.0,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0003145887180581963,
      "loss": 5.0476,
      "mean_token_accuracy": 0.2061317414045334,
      "num_tokens": 11570380.0,
      "step": 4910
    },
    {
      "entropy": 5.367516136169433,
      "epoch": 5.005094243504839,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003141937457263523,
      "loss": 4.7385,
      "mean_token_accuracy": 0.2294103816151619,
      "num_tokens": 11581749.0,
      "step": 4915
    },
    {
      "entropy": 5.320555067062378,
      "epoch": 5.010188487009679,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003137986487983668,
      "loss": 4.79,
      "mean_token_accuracy": 0.22036812454462051,
      "num_tokens": 11593812.0,
      "step": 4920
    },
    {
      "entropy": 5.346633768081665,
      "epoch": 5.015282730514518,
      "grad_norm": 1.171875,
      "learning_rate": 0.00031340342853024533,
      "loss": 4.7639,
      "mean_token_accuracy": 0.21964269131422043,
      "num_tokens": 11604734.0,
      "step": 4925
    },
    {
      "entropy": 5.349350500106811,
      "epoch": 5.020376974019358,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003130080861783851,
      "loss": 4.7373,
      "mean_token_accuracy": 0.22793440967798234,
      "num_tokens": 11616919.0,
      "step": 4930
    },
    {
      "entropy": 5.393327713012695,
      "epoch": 5.025471217524197,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003126126229995717,
      "loss": 4.8153,
      "mean_token_accuracy": 0.21688217669725418,
      "num_tokens": 11628114.0,
      "step": 4935
    },
    {
      "entropy": 5.343525791168213,
      "epoch": 5.030565461029037,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003122170402509747,
      "loss": 4.7746,
      "mean_token_accuracy": 0.22237068712711333,
      "num_tokens": 11640614.0,
      "step": 4940
    },
    {
      "entropy": 5.400874185562134,
      "epoch": 5.0356597045338765,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003118213391901439,
      "loss": 4.7812,
      "mean_token_accuracy": 0.22263894677162172,
      "num_tokens": 11653182.0,
      "step": 4945
    },
    {
      "entropy": 5.333602428436279,
      "epoch": 5.0407539480387165,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000311425521075005,
      "loss": 4.7123,
      "mean_token_accuracy": 0.22871391624212264,
      "num_tokens": 11665152.0,
      "step": 4950
    },
    {
      "entropy": 5.300535106658936,
      "epoch": 5.045848191543556,
      "grad_norm": 1.25,
      "learning_rate": 0.0003110295871638561,
      "loss": 4.788,
      "mean_token_accuracy": 0.21997157037258147,
      "num_tokens": 11676149.0,
      "step": 4955
    },
    {
      "entropy": 5.337724828720093,
      "epoch": 5.050942435048396,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003106335387153632,
      "loss": 4.705,
      "mean_token_accuracy": 0.23031598329544067,
      "num_tokens": 11686138.0,
      "step": 4960
    },
    {
      "entropy": 5.3073831558227536,
      "epoch": 5.056036678553235,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003102373769885565,
      "loss": 4.7139,
      "mean_token_accuracy": 0.21811883002519608,
      "num_tokens": 11697192.0,
      "step": 4965
    },
    {
      "entropy": 5.241533422470093,
      "epoch": 5.061130922058075,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003098411032428262,
      "loss": 4.6686,
      "mean_token_accuracy": 0.23379482924938202,
      "num_tokens": 11708959.0,
      "step": 4970
    },
    {
      "entropy": 5.30102391242981,
      "epoch": 5.066225165562914,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00030944471873791876,
      "loss": 4.7686,
      "mean_token_accuracy": 0.22368180006742477,
      "num_tokens": 11720507.0,
      "step": 4975
    },
    {
      "entropy": 5.341554021835327,
      "epoch": 5.071319409067754,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003090482247339326,
      "loss": 4.7572,
      "mean_token_accuracy": 0.2216231867671013,
      "num_tokens": 11733775.0,
      "step": 4980
    },
    {
      "entropy": 5.412316942214966,
      "epoch": 5.076413652572593,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003086516224913144,
      "loss": 4.8284,
      "mean_token_accuracy": 0.21866228878498079,
      "num_tokens": 11746383.0,
      "step": 4985
    },
    {
      "entropy": 5.262031888961792,
      "epoch": 5.081507896077433,
      "grad_norm": 1.125,
      "learning_rate": 0.0003082549132708548,
      "loss": 4.7861,
      "mean_token_accuracy": 0.21758369952440262,
      "num_tokens": 11758232.0,
      "step": 4990
    },
    {
      "entropy": 5.261825132369995,
      "epoch": 5.086602139582272,
      "grad_norm": 1.109375,
      "learning_rate": 0.00030785809833368456,
      "loss": 4.6823,
      "mean_token_accuracy": 0.22409756779670714,
      "num_tokens": 11769636.0,
      "step": 4995
    },
    {
      "entropy": 5.3370648384094235,
      "epoch": 5.091696383087111,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00030746117894127047,
      "loss": 4.7771,
      "mean_token_accuracy": 0.221085587143898,
      "num_tokens": 11781494.0,
      "step": 5000
    },
    {
      "epoch": 5.091696383087111,
      "eval_entropy": 5.116202018879078,
      "eval_loss": 5.697664737701416,
      "eval_mean_token_accuracy": 0.1785265859262443,
      "eval_num_tokens": 11781494.0,
      "eval_runtime": 2.6715,
      "eval_samples_per_second": 1212.807,
      "eval_steps_per_second": 151.601,
      "step": 5000
    },
    {
      "entropy": 5.321157503128052,
      "epoch": 5.096790626591951,
      "grad_norm": 1.171875,
      "learning_rate": 0.00030706415635541143,
      "loss": 4.7579,
      "mean_token_accuracy": 0.22442465722560884,
      "num_tokens": 11793698.0,
      "step": 5005
    },
    {
      "entropy": 5.302441596984863,
      "epoch": 5.10188487009679,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003066670318382344,
      "loss": 4.7611,
      "mean_token_accuracy": 0.21992678493261336,
      "num_tokens": 11804556.0,
      "step": 5010
    },
    {
      "entropy": 5.280402994155883,
      "epoch": 5.10697911360163,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003062698066521903,
      "loss": 4.7367,
      "mean_token_accuracy": 0.22201977521181107,
      "num_tokens": 11816369.0,
      "step": 5015
    },
    {
      "entropy": 5.300079345703125,
      "epoch": 5.112073357106469,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003058724820600503,
      "loss": 4.7648,
      "mean_token_accuracy": 0.21335089057683945,
      "num_tokens": 11828573.0,
      "step": 5020
    },
    {
      "entropy": 5.310085010528565,
      "epoch": 5.117167600611309,
      "grad_norm": 1.171875,
      "learning_rate": 0.00030547505932490115,
      "loss": 4.7579,
      "mean_token_accuracy": 0.22165835946798323,
      "num_tokens": 11839941.0,
      "step": 5025
    },
    {
      "entropy": 5.389428806304932,
      "epoch": 5.1222618441161485,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003050775397101421,
      "loss": 4.8428,
      "mean_token_accuracy": 0.2137691006064415,
      "num_tokens": 11854858.0,
      "step": 5030
    },
    {
      "entropy": 5.26011905670166,
      "epoch": 5.1273560876209885,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003046799244794799,
      "loss": 4.6306,
      "mean_token_accuracy": 0.23221800327301026,
      "num_tokens": 11866774.0,
      "step": 5035
    },
    {
      "entropy": 5.2602417945861815,
      "epoch": 5.132450331125828,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003042822148969257,
      "loss": 4.7221,
      "mean_token_accuracy": 0.22927424311637878,
      "num_tokens": 11878247.0,
      "step": 5040
    },
    {
      "entropy": 5.223009634017944,
      "epoch": 5.137544574630668,
      "grad_norm": 1.125,
      "learning_rate": 0.00030388441222679017,
      "loss": 4.7368,
      "mean_token_accuracy": 0.22680188566446305,
      "num_tokens": 11891290.0,
      "step": 5045
    },
    {
      "entropy": 5.37976565361023,
      "epoch": 5.142638818135507,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00030348651773368047,
      "loss": 4.8373,
      "mean_token_accuracy": 0.21878013908863067,
      "num_tokens": 11905227.0,
      "step": 5050
    },
    {
      "entropy": 5.318962621688843,
      "epoch": 5.147733061640347,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00030308853268249487,
      "loss": 4.7535,
      "mean_token_accuracy": 0.22587480247020722,
      "num_tokens": 11917819.0,
      "step": 5055
    },
    {
      "entropy": 5.2430112838745115,
      "epoch": 5.152827305145186,
      "grad_norm": 1.390625,
      "learning_rate": 0.00030269045833842027,
      "loss": 4.7254,
      "mean_token_accuracy": 0.231517930328846,
      "num_tokens": 11929325.0,
      "step": 5060
    },
    {
      "entropy": 5.318389797210694,
      "epoch": 5.157921548650026,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00030229229596692703,
      "loss": 4.75,
      "mean_token_accuracy": 0.22693974375724793,
      "num_tokens": 11941298.0,
      "step": 5065
    },
    {
      "entropy": 5.350164413452148,
      "epoch": 5.163015792154865,
      "grad_norm": 1.203125,
      "learning_rate": 0.00030189404683376547,
      "loss": 4.7936,
      "mean_token_accuracy": 0.2182493805885315,
      "num_tokens": 11952808.0,
      "step": 5070
    },
    {
      "entropy": 5.338929462432861,
      "epoch": 5.168110035659705,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003014957122049617,
      "loss": 4.8738,
      "mean_token_accuracy": 0.2155437782406807,
      "num_tokens": 11967208.0,
      "step": 5075
    },
    {
      "entropy": 5.333193254470825,
      "epoch": 5.173204279164544,
      "grad_norm": 1.25,
      "learning_rate": 0.0003010972933468136,
      "loss": 4.8295,
      "mean_token_accuracy": 0.21691921204328538,
      "num_tokens": 11979218.0,
      "step": 5080
    },
    {
      "entropy": 5.310906171798706,
      "epoch": 5.178298522669383,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003006987915258867,
      "loss": 4.7751,
      "mean_token_accuracy": 0.22493956238031387,
      "num_tokens": 11990809.0,
      "step": 5085
    },
    {
      "entropy": 5.367737674713135,
      "epoch": 5.183392766174223,
      "grad_norm": 1.40625,
      "learning_rate": 0.00030030020800901066,
      "loss": 4.732,
      "mean_token_accuracy": 0.22175245732069016,
      "num_tokens": 12001999.0,
      "step": 5090
    },
    {
      "entropy": 5.3675394535064695,
      "epoch": 5.188487009679062,
      "grad_norm": 1.3125,
      "learning_rate": 0.00029990154406327435,
      "loss": 4.8187,
      "mean_token_accuracy": 0.2177831918001175,
      "num_tokens": 12014507.0,
      "step": 5095
    },
    {
      "entropy": 5.31121301651001,
      "epoch": 5.193581253183902,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002995028009560228,
      "loss": 4.6868,
      "mean_token_accuracy": 0.23257064074277878,
      "num_tokens": 12026000.0,
      "step": 5100
    },
    {
      "entropy": 5.269972467422486,
      "epoch": 5.198675496688741,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00029910397995485213,
      "loss": 4.725,
      "mean_token_accuracy": 0.22587687969207765,
      "num_tokens": 12036632.0,
      "step": 5105
    },
    {
      "entropy": 5.309423303604126,
      "epoch": 5.203769740193581,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00029870508232760667,
      "loss": 4.8184,
      "mean_token_accuracy": 0.2162817671895027,
      "num_tokens": 12047605.0,
      "step": 5110
    },
    {
      "entropy": 5.239737033843994,
      "epoch": 5.2088639836984205,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00029830610934237394,
      "loss": 4.6852,
      "mean_token_accuracy": 0.23151226043701173,
      "num_tokens": 12059124.0,
      "step": 5115
    },
    {
      "entropy": 5.35202808380127,
      "epoch": 5.2139582272032605,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00029790706226748124,
      "loss": 4.7101,
      "mean_token_accuracy": 0.22337489873170852,
      "num_tokens": 12069651.0,
      "step": 5120
    },
    {
      "entropy": 5.272590208053589,
      "epoch": 5.2190524707081,
      "grad_norm": 1.140625,
      "learning_rate": 0.00029750794237149125,
      "loss": 4.7968,
      "mean_token_accuracy": 0.2151492416858673,
      "num_tokens": 12082639.0,
      "step": 5125
    },
    {
      "entropy": 5.283179092407226,
      "epoch": 5.22414671421294,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00029710875092319827,
      "loss": 4.7343,
      "mean_token_accuracy": 0.22324326634407043,
      "num_tokens": 12094667.0,
      "step": 5130
    },
    {
      "entropy": 5.295515727996826,
      "epoch": 5.229240957717779,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000296709489191624,
      "loss": 4.7649,
      "mean_token_accuracy": 0.22441953122615815,
      "num_tokens": 12106567.0,
      "step": 5135
    },
    {
      "entropy": 5.329156494140625,
      "epoch": 5.234335201222619,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002963101584460136,
      "loss": 4.7693,
      "mean_token_accuracy": 0.22204919755458832,
      "num_tokens": 12117505.0,
      "step": 5140
    },
    {
      "entropy": 5.26026725769043,
      "epoch": 5.239429444727458,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002959107599558315,
      "loss": 4.7368,
      "mean_token_accuracy": 0.23440717309713363,
      "num_tokens": 12129670.0,
      "step": 5145
    },
    {
      "entropy": 5.271461009979248,
      "epoch": 5.244523688232298,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002955112949907576,
      "loss": 4.7468,
      "mean_token_accuracy": 0.22598524689674376,
      "num_tokens": 12141227.0,
      "step": 5150
    },
    {
      "entropy": 5.327522897720337,
      "epoch": 5.249617931737137,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00029511176482068327,
      "loss": 4.7715,
      "mean_token_accuracy": 0.22456395626068115,
      "num_tokens": 12153299.0,
      "step": 5155
    },
    {
      "entropy": 5.296003293991089,
      "epoch": 5.254712175241977,
      "grad_norm": 1.125,
      "learning_rate": 0.00029471217071570685,
      "loss": 4.7237,
      "mean_token_accuracy": 0.23002535551786424,
      "num_tokens": 12165337.0,
      "step": 5160
    },
    {
      "entropy": 5.292874813079834,
      "epoch": 5.259806418746816,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002943125139461301,
      "loss": 4.7096,
      "mean_token_accuracy": 0.23363901525735856,
      "num_tokens": 12176916.0,
      "step": 5165
    },
    {
      "entropy": 5.251583528518677,
      "epoch": 5.264900662251655,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002939127957824541,
      "loss": 4.8218,
      "mean_token_accuracy": 0.22326641529798508,
      "num_tokens": 12189448.0,
      "step": 5170
    },
    {
      "entropy": 5.2524182319641115,
      "epoch": 5.269994905756495,
      "grad_norm": 1.125,
      "learning_rate": 0.00029351301749537485,
      "loss": 4.7688,
      "mean_token_accuracy": 0.22802286148071288,
      "num_tokens": 12200761.0,
      "step": 5175
    },
    {
      "entropy": 5.255531930923462,
      "epoch": 5.275089149261334,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002931131803557796,
      "loss": 4.6961,
      "mean_token_accuracy": 0.23549867570400237,
      "num_tokens": 12212616.0,
      "step": 5180
    },
    {
      "entropy": 5.357332754135132,
      "epoch": 5.280183392766174,
      "grad_norm": 1.34375,
      "learning_rate": 0.00029271328563474275,
      "loss": 4.8817,
      "mean_token_accuracy": 0.21308680325746537,
      "num_tokens": 12224559.0,
      "step": 5185
    },
    {
      "entropy": 5.284110975265503,
      "epoch": 5.285277636271013,
      "grad_norm": 1.34375,
      "learning_rate": 0.00029231333460352175,
      "loss": 4.7487,
      "mean_token_accuracy": 0.22766771614551545,
      "num_tokens": 12236803.0,
      "step": 5190
    },
    {
      "entropy": 5.280096006393433,
      "epoch": 5.290371879775853,
      "grad_norm": 1.125,
      "learning_rate": 0.0002919133285335529,
      "loss": 4.7365,
      "mean_token_accuracy": 0.22699289470911027,
      "num_tokens": 12247983.0,
      "step": 5195
    },
    {
      "entropy": 5.247777700424194,
      "epoch": 5.2954661232806925,
      "grad_norm": 1.25,
      "learning_rate": 0.0002915132686964476,
      "loss": 4.7278,
      "mean_token_accuracy": 0.22666896134614944,
      "num_tokens": 12259650.0,
      "step": 5200
    },
    {
      "entropy": 5.281350946426391,
      "epoch": 5.3005603667855326,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002911131563639883,
      "loss": 4.7736,
      "mean_token_accuracy": 0.22422062307596208,
      "num_tokens": 12272016.0,
      "step": 5205
    },
    {
      "entropy": 5.285686349868774,
      "epoch": 5.305654610290372,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00029071299280812395,
      "loss": 4.7565,
      "mean_token_accuracy": 0.22544667720794678,
      "num_tokens": 12283102.0,
      "step": 5210
    },
    {
      "entropy": 5.276587963104248,
      "epoch": 5.310748853795212,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002903127793009667,
      "loss": 4.7323,
      "mean_token_accuracy": 0.23656094074249268,
      "num_tokens": 12294409.0,
      "step": 5215
    },
    {
      "entropy": 5.313216972351074,
      "epoch": 5.315843097300051,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002899125171147872,
      "loss": 4.8343,
      "mean_token_accuracy": 0.2126247614622116,
      "num_tokens": 12305880.0,
      "step": 5220
    },
    {
      "entropy": 5.248071956634521,
      "epoch": 5.320937340804891,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002895122075220113,
      "loss": 4.6553,
      "mean_token_accuracy": 0.2346448853611946,
      "num_tokens": 12317258.0,
      "step": 5225
    },
    {
      "entropy": 5.278992319107056,
      "epoch": 5.32603158430973,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002891118517952151,
      "loss": 4.7868,
      "mean_token_accuracy": 0.2300610974431038,
      "num_tokens": 12328214.0,
      "step": 5230
    },
    {
      "entropy": 5.261919355392456,
      "epoch": 5.33112582781457,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002887114512071215,
      "loss": 4.8469,
      "mean_token_accuracy": 0.21920417696237565,
      "num_tokens": 12341333.0,
      "step": 5235
    },
    {
      "entropy": 5.3255870819091795,
      "epoch": 5.336220071319409,
      "grad_norm": 1.25,
      "learning_rate": 0.000288311007030596,
      "loss": 4.8301,
      "mean_token_accuracy": 0.2216414600610733,
      "num_tokens": 12352864.0,
      "step": 5240
    },
    {
      "entropy": 5.359915065765381,
      "epoch": 5.341314314824249,
      "grad_norm": 1.125,
      "learning_rate": 0.0002879105205386429,
      "loss": 4.8744,
      "mean_token_accuracy": 0.21428250074386596,
      "num_tokens": 12364795.0,
      "step": 5245
    },
    {
      "entropy": 5.388999032974243,
      "epoch": 5.346408558329088,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002875099930044007,
      "loss": 4.8074,
      "mean_token_accuracy": 0.21171821653842926,
      "num_tokens": 12376859.0,
      "step": 5250
    },
    {
      "entropy": 5.318556833267212,
      "epoch": 5.351502801833927,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00028710942570113853,
      "loss": 4.7672,
      "mean_token_accuracy": 0.22188491523265838,
      "num_tokens": 12388928.0,
      "step": 5255
    },
    {
      "entropy": 5.2329676151275635,
      "epoch": 5.356597045338767,
      "grad_norm": 1.234375,
      "learning_rate": 0.00028670881990225177,
      "loss": 4.6703,
      "mean_token_accuracy": 0.2326484277844429,
      "num_tokens": 12400371.0,
      "step": 5260
    },
    {
      "entropy": 5.25079607963562,
      "epoch": 5.361691288843606,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002863081768812585,
      "loss": 4.751,
      "mean_token_accuracy": 0.22226710021495819,
      "num_tokens": 12412034.0,
      "step": 5265
    },
    {
      "entropy": 5.310902500152588,
      "epoch": 5.366785532348446,
      "grad_norm": 1.125,
      "learning_rate": 0.0002859074979117949,
      "loss": 4.8983,
      "mean_token_accuracy": 0.21635923236608506,
      "num_tokens": 12424785.0,
      "step": 5270
    },
    {
      "entropy": 5.2841705799102785,
      "epoch": 5.3718797758532855,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002855067842676115,
      "loss": 4.7137,
      "mean_token_accuracy": 0.2331419751048088,
      "num_tokens": 12435254.0,
      "step": 5275
    },
    {
      "entropy": 5.308323001861572,
      "epoch": 5.3769740193581255,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00028510603722256897,
      "loss": 4.8007,
      "mean_token_accuracy": 0.21709874421358108,
      "num_tokens": 12447710.0,
      "step": 5280
    },
    {
      "entropy": 5.275171327590942,
      "epoch": 5.382068262862965,
      "grad_norm": 1.25,
      "learning_rate": 0.00028470525805063425,
      "loss": 4.7027,
      "mean_token_accuracy": 0.23219434916973114,
      "num_tokens": 12458622.0,
      "step": 5285
    },
    {
      "entropy": 5.371753406524658,
      "epoch": 5.387162506367805,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00028430444802587647,
      "loss": 4.9114,
      "mean_token_accuracy": 0.20741007328033448,
      "num_tokens": 12472114.0,
      "step": 5290
    },
    {
      "entropy": 5.275445032119751,
      "epoch": 5.392256749872644,
      "grad_norm": 1.3125,
      "learning_rate": 0.00028390360842246265,
      "loss": 4.7512,
      "mean_token_accuracy": 0.22168818563222886,
      "num_tokens": 12483927.0,
      "step": 5295
    },
    {
      "entropy": 5.223222780227661,
      "epoch": 5.397350993377484,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000283502740514654,
      "loss": 4.6982,
      "mean_token_accuracy": 0.22570544928312303,
      "num_tokens": 12494928.0,
      "step": 5300
    },
    {
      "entropy": 5.191484594345093,
      "epoch": 5.402445236882323,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002831018455768016,
      "loss": 4.7192,
      "mean_token_accuracy": 0.23438584208488464,
      "num_tokens": 12506418.0,
      "step": 5305
    },
    {
      "entropy": 5.2769598960876465,
      "epoch": 5.407539480387163,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002827009248833427,
      "loss": 4.7752,
      "mean_token_accuracy": 0.22103542387485503,
      "num_tokens": 12519099.0,
      "step": 5310
    },
    {
      "entropy": 5.2742729663848875,
      "epoch": 5.412633723892002,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00028229997970879604,
      "loss": 4.7798,
      "mean_token_accuracy": 0.22923927158117294,
      "num_tokens": 12529678.0,
      "step": 5315
    },
    {
      "entropy": 5.2699745178222654,
      "epoch": 5.417727967396842,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002818990113277584,
      "loss": 4.7941,
      "mean_token_accuracy": 0.21754174679517746,
      "num_tokens": 12542072.0,
      "step": 5320
    },
    {
      "entropy": 5.33921537399292,
      "epoch": 5.422822210901681,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002814980210149005,
      "loss": 4.7749,
      "mean_token_accuracy": 0.22073635309934617,
      "num_tokens": 12554694.0,
      "step": 5325
    },
    {
      "entropy": 5.30570421218872,
      "epoch": 5.427916454406521,
      "grad_norm": 1.203125,
      "learning_rate": 0.00028109701004496254,
      "loss": 4.804,
      "mean_token_accuracy": 0.22337957769632338,
      "num_tokens": 12566449.0,
      "step": 5330
    },
    {
      "entropy": 5.320533514022827,
      "epoch": 5.43301069791136,
      "grad_norm": 1.15625,
      "learning_rate": 0.00028069597969275023,
      "loss": 4.7871,
      "mean_token_accuracy": 0.2216341108083725,
      "num_tokens": 12578853.0,
      "step": 5335
    },
    {
      "entropy": 5.395033836364746,
      "epoch": 5.438104941416199,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002802949312331314,
      "loss": 4.8779,
      "mean_token_accuracy": 0.20491388142108918,
      "num_tokens": 12590689.0,
      "step": 5340
    },
    {
      "entropy": 5.2268500328063965,
      "epoch": 5.443199184921039,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000279893865941031,
      "loss": 4.761,
      "mean_token_accuracy": 0.22644856721162795,
      "num_tokens": 12602646.0,
      "step": 5345
    },
    {
      "entropy": 5.348932933807373,
      "epoch": 5.448293428425878,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002794927850914278,
      "loss": 4.801,
      "mean_token_accuracy": 0.2171991065144539,
      "num_tokens": 12615371.0,
      "step": 5350
    },
    {
      "entropy": 5.334804677963257,
      "epoch": 5.453387671930718,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002790916899593496,
      "loss": 4.8126,
      "mean_token_accuracy": 0.21606216430664063,
      "num_tokens": 12627472.0,
      "step": 5355
    },
    {
      "entropy": 5.353354549407959,
      "epoch": 5.4584819154355575,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002786905818198702,
      "loss": 4.8873,
      "mean_token_accuracy": 0.2086475074291229,
      "num_tokens": 12640151.0,
      "step": 5360
    },
    {
      "entropy": 5.319568920135498,
      "epoch": 5.4635761589403975,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002782894619481042,
      "loss": 4.851,
      "mean_token_accuracy": 0.21386671513319017,
      "num_tokens": 12653578.0,
      "step": 5365
    },
    {
      "entropy": 5.238232517242432,
      "epoch": 5.468670402445237,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000277888331619204,
      "loss": 4.6817,
      "mean_token_accuracy": 0.2315113827586174,
      "num_tokens": 12664158.0,
      "step": 5370
    },
    {
      "entropy": 5.3355857849121096,
      "epoch": 5.473764645950077,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002774871921083546,
      "loss": 4.7272,
      "mean_token_accuracy": 0.22962037473917007,
      "num_tokens": 12674165.0,
      "step": 5375
    },
    {
      "entropy": 5.193838262557984,
      "epoch": 5.478858889454916,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002770860446907708,
      "loss": 4.6401,
      "mean_token_accuracy": 0.2418966546654701,
      "num_tokens": 12684787.0,
      "step": 5380
    },
    {
      "entropy": 5.290788078308106,
      "epoch": 5.483953132959756,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002766848906416922,
      "loss": 4.8328,
      "mean_token_accuracy": 0.2179897040128708,
      "num_tokens": 12695846.0,
      "step": 5385
    },
    {
      "entropy": 5.21913514137268,
      "epoch": 5.489047376464595,
      "grad_norm": 1.078125,
      "learning_rate": 0.00027628373123637977,
      "loss": 4.7755,
      "mean_token_accuracy": 0.22442578077316283,
      "num_tokens": 12708643.0,
      "step": 5390
    },
    {
      "entropy": 5.33020601272583,
      "epoch": 5.494141619969435,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000275882567750111,
      "loss": 4.8823,
      "mean_token_accuracy": 0.21489386409521102,
      "num_tokens": 12721599.0,
      "step": 5395
    },
    {
      "entropy": 5.387035989761353,
      "epoch": 5.499235863474274,
      "grad_norm": 1.25,
      "learning_rate": 0.0002754814014581769,
      "loss": 4.8307,
      "mean_token_accuracy": 0.2242463320493698,
      "num_tokens": 12734902.0,
      "step": 5400
    },
    {
      "entropy": 5.339293193817139,
      "epoch": 5.504330106979114,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002750802336358771,
      "loss": 4.9162,
      "mean_token_accuracy": 0.20969029217958451,
      "num_tokens": 12746087.0,
      "step": 5405
    },
    {
      "entropy": 5.265274715423584,
      "epoch": 5.509424350483953,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00027467906555851614,
      "loss": 4.7933,
      "mean_token_accuracy": 0.22152003645896912,
      "num_tokens": 12759366.0,
      "step": 5410
    },
    {
      "entropy": 5.240903663635254,
      "epoch": 5.514518593988793,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002742778985013994,
      "loss": 4.7051,
      "mean_token_accuracy": 0.2264609232544899,
      "num_tokens": 12771691.0,
      "step": 5415
    },
    {
      "entropy": 5.3040179252624515,
      "epoch": 5.519612837493632,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002738767337398292,
      "loss": 4.8153,
      "mean_token_accuracy": 0.22320718616247176,
      "num_tokens": 12783342.0,
      "step": 5420
    },
    {
      "entropy": 5.299612283706665,
      "epoch": 5.524707080998471,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002734755725491,
      "loss": 4.7665,
      "mean_token_accuracy": 0.2172636419534683,
      "num_tokens": 12794207.0,
      "step": 5425
    },
    {
      "entropy": 5.342070722579956,
      "epoch": 5.529801324503311,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002730744162044955,
      "loss": 4.8489,
      "mean_token_accuracy": 0.21600721627473832,
      "num_tokens": 12806153.0,
      "step": 5430
    },
    {
      "entropy": 5.292962980270386,
      "epoch": 5.53489556800815,
      "grad_norm": 1.3125,
      "learning_rate": 0.00027267326598128373,
      "loss": 4.8298,
      "mean_token_accuracy": 0.21777139902114867,
      "num_tokens": 12817524.0,
      "step": 5435
    },
    {
      "entropy": 5.3304208278656,
      "epoch": 5.53998981151299,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00027227212315471337,
      "loss": 4.7914,
      "mean_token_accuracy": 0.221538008749485,
      "num_tokens": 12830224.0,
      "step": 5440
    },
    {
      "entropy": 5.349244165420532,
      "epoch": 5.5450840550178295,
      "grad_norm": 1.140625,
      "learning_rate": 0.00027187098900000945,
      "loss": 4.8013,
      "mean_token_accuracy": 0.2163992404937744,
      "num_tokens": 12843404.0,
      "step": 5445
    },
    {
      "entropy": 5.229301023483276,
      "epoch": 5.5501782985226695,
      "grad_norm": 1.125,
      "learning_rate": 0.00027146986479236954,
      "loss": 4.7079,
      "mean_token_accuracy": 0.2317512586712837,
      "num_tokens": 12854917.0,
      "step": 5450
    },
    {
      "entropy": 5.289776992797852,
      "epoch": 5.555272542027509,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002710687518069595,
      "loss": 4.7862,
      "mean_token_accuracy": 0.21520272195339202,
      "num_tokens": 12866452.0,
      "step": 5455
    },
    {
      "entropy": 5.255536270141602,
      "epoch": 5.560366785532349,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002706676513189097,
      "loss": 4.7505,
      "mean_token_accuracy": 0.223416568338871,
      "num_tokens": 12879439.0,
      "step": 5460
    },
    {
      "entropy": 5.285934829711914,
      "epoch": 5.565461029037188,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002702665646033105,
      "loss": 4.7956,
      "mean_token_accuracy": 0.21677534133195878,
      "num_tokens": 12892223.0,
      "step": 5465
    },
    {
      "entropy": 5.295909690856933,
      "epoch": 5.570555272542028,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002698654929352087,
      "loss": 4.7251,
      "mean_token_accuracy": 0.2292933478951454,
      "num_tokens": 12903814.0,
      "step": 5470
    },
    {
      "entropy": 5.281788778305054,
      "epoch": 5.575649516046867,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002694644375896031,
      "loss": 4.812,
      "mean_token_accuracy": 0.21538807898759843,
      "num_tokens": 12915767.0,
      "step": 5475
    },
    {
      "entropy": 5.319742965698242,
      "epoch": 5.580743759551707,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00026906339984144077,
      "loss": 4.7822,
      "mean_token_accuracy": 0.21814385205507278,
      "num_tokens": 12926716.0,
      "step": 5480
    },
    {
      "entropy": 5.300481510162354,
      "epoch": 5.585838003056546,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002686623809656128,
      "loss": 4.8381,
      "mean_token_accuracy": 0.21908631771802903,
      "num_tokens": 12938617.0,
      "step": 5485
    },
    {
      "entropy": 5.25345778465271,
      "epoch": 5.590932246561386,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00026826138223695026,
      "loss": 4.8251,
      "mean_token_accuracy": 0.21725087910890578,
      "num_tokens": 12951333.0,
      "step": 5490
    },
    {
      "entropy": 5.2960920333862305,
      "epoch": 5.596026490066225,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002678604049302201,
      "loss": 4.78,
      "mean_token_accuracy": 0.2233548179268837,
      "num_tokens": 12962633.0,
      "step": 5495
    },
    {
      "entropy": 5.303360652923584,
      "epoch": 5.601120733571065,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00026745945032012126,
      "loss": 4.8014,
      "mean_token_accuracy": 0.2187095910310745,
      "num_tokens": 12974013.0,
      "step": 5500
    },
    {
      "epoch": 5.601120733571065,
      "eval_entropy": 5.160082993683992,
      "eval_loss": 5.663180351257324,
      "eval_mean_token_accuracy": 0.18091766878410623,
      "eval_num_tokens": 12974013.0,
      "eval_runtime": 2.6844,
      "eval_samples_per_second": 1206.979,
      "eval_steps_per_second": 150.872,
      "step": 5500
    },
    {
      "entropy": 5.356230974197388,
      "epoch": 5.606214977075904,
      "grad_norm": 1.171875,
      "learning_rate": 0.00026705851968128046,
      "loss": 4.8194,
      "mean_token_accuracy": 0.21552616357803345,
      "num_tokens": 12985292.0,
      "step": 5505
    },
    {
      "entropy": 5.245092153549194,
      "epoch": 5.611309220580743,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00026665761428824837,
      "loss": 4.7137,
      "mean_token_accuracy": 0.22757291346788405,
      "num_tokens": 12996914.0,
      "step": 5510
    },
    {
      "entropy": 5.330206108093262,
      "epoch": 5.616403464085583,
      "grad_norm": 1.234375,
      "learning_rate": 0.00026625673541549525,
      "loss": 4.8329,
      "mean_token_accuracy": 0.22164387702941896,
      "num_tokens": 13008990.0,
      "step": 5515
    },
    {
      "entropy": 5.34192910194397,
      "epoch": 5.621497707590422,
      "grad_norm": 1.359375,
      "learning_rate": 0.00026585588433740697,
      "loss": 4.8313,
      "mean_token_accuracy": 0.21830009967088698,
      "num_tokens": 13018944.0,
      "step": 5520
    },
    {
      "entropy": 5.261390829086304,
      "epoch": 5.6265919510952624,
      "grad_norm": 1.125,
      "learning_rate": 0.0002654550623282813,
      "loss": 4.7679,
      "mean_token_accuracy": 0.22438600212335585,
      "num_tokens": 13029765.0,
      "step": 5525
    },
    {
      "entropy": 5.3112249851226805,
      "epoch": 5.631686194600102,
      "grad_norm": 1.3125,
      "learning_rate": 0.00026505427066232353,
      "loss": 4.8367,
      "mean_token_accuracy": 0.22343890517950057,
      "num_tokens": 13040296.0,
      "step": 5530
    },
    {
      "entropy": 5.28897066116333,
      "epoch": 5.636780438104942,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002646535106136423,
      "loss": 4.7445,
      "mean_token_accuracy": 0.23197436034679414,
      "num_tokens": 13050242.0,
      "step": 5535
    },
    {
      "entropy": 5.3211840152740475,
      "epoch": 5.641874681609781,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002642527834562459,
      "loss": 4.8234,
      "mean_token_accuracy": 0.2141679733991623,
      "num_tokens": 13061171.0,
      "step": 5540
    },
    {
      "entropy": 5.3132641315460205,
      "epoch": 5.646968925114621,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002638520904640381,
      "loss": 4.8001,
      "mean_token_accuracy": 0.21395914256572723,
      "num_tokens": 13073316.0,
      "step": 5545
    },
    {
      "entropy": 5.321552038192749,
      "epoch": 5.65206316861946,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002634514329108138,
      "loss": 4.8659,
      "mean_token_accuracy": 0.21299288868904115,
      "num_tokens": 13086491.0,
      "step": 5550
    },
    {
      "entropy": 5.2759730339050295,
      "epoch": 5.6571574121243,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002630508120702557,
      "loss": 4.7672,
      "mean_token_accuracy": 0.2235550880432129,
      "num_tokens": 13098427.0,
      "step": 5555
    },
    {
      "entropy": 5.268409967422485,
      "epoch": 5.662251655629139,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002626502292159292,
      "loss": 4.8086,
      "mean_token_accuracy": 0.22338040471076964,
      "num_tokens": 13110465.0,
      "step": 5560
    },
    {
      "entropy": 5.345301914215088,
      "epoch": 5.667345899133979,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002622496856212795,
      "loss": 4.8357,
      "mean_token_accuracy": 0.2162660613656044,
      "num_tokens": 13121930.0,
      "step": 5565
    },
    {
      "entropy": 5.254309368133545,
      "epoch": 5.672440142638818,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002618491825596267,
      "loss": 4.7062,
      "mean_token_accuracy": 0.2261454001069069,
      "num_tokens": 13132015.0,
      "step": 5570
    },
    {
      "entropy": 5.24613127708435,
      "epoch": 5.677534386143658,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002614487213041621,
      "loss": 4.7243,
      "mean_token_accuracy": 0.22474730610847474,
      "num_tokens": 13142752.0,
      "step": 5575
    },
    {
      "entropy": 5.266914749145508,
      "epoch": 5.682628629648497,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002610483031279441,
      "loss": 4.7991,
      "mean_token_accuracy": 0.22486677020788193,
      "num_tokens": 13154760.0,
      "step": 5580
    },
    {
      "entropy": 5.261200952529907,
      "epoch": 5.687722873153337,
      "grad_norm": 1.234375,
      "learning_rate": 0.00026064792930389406,
      "loss": 4.7558,
      "mean_token_accuracy": 0.22247630655765532,
      "num_tokens": 13166102.0,
      "step": 5585
    },
    {
      "entropy": 5.241149520874023,
      "epoch": 5.692817116658176,
      "grad_norm": 1.125,
      "learning_rate": 0.0002602476011047926,
      "loss": 4.734,
      "mean_token_accuracy": 0.22420804053545,
      "num_tokens": 13177794.0,
      "step": 5590
    },
    {
      "entropy": 5.296810150146484,
      "epoch": 5.697911360163015,
      "grad_norm": 1.234375,
      "learning_rate": 0.00025984731980327496,
      "loss": 4.7949,
      "mean_token_accuracy": 0.2232470154762268,
      "num_tokens": 13188514.0,
      "step": 5595
    },
    {
      "entropy": 5.289203643798828,
      "epoch": 5.703005603667855,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00025944708667182754,
      "loss": 4.799,
      "mean_token_accuracy": 0.22321783006191254,
      "num_tokens": 13199627.0,
      "step": 5600
    },
    {
      "entropy": 5.293942928314209,
      "epoch": 5.7080998471726945,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002590469029827836,
      "loss": 4.8302,
      "mean_token_accuracy": 0.21186018735170364,
      "num_tokens": 13211645.0,
      "step": 5605
    },
    {
      "entropy": 5.358622884750366,
      "epoch": 5.7131940906775345,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002586467700083191,
      "loss": 4.8568,
      "mean_token_accuracy": 0.2134514644742012,
      "num_tokens": 13224251.0,
      "step": 5610
    },
    {
      "entropy": 5.265875244140625,
      "epoch": 5.718288334182374,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00025824668902044883,
      "loss": 4.7932,
      "mean_token_accuracy": 0.21695260107517242,
      "num_tokens": 13236639.0,
      "step": 5615
    },
    {
      "entropy": 5.258214902877808,
      "epoch": 5.723382577687214,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00025784666129102254,
      "loss": 4.7903,
      "mean_token_accuracy": 0.21794215440750123,
      "num_tokens": 13247923.0,
      "step": 5620
    },
    {
      "entropy": 5.293379402160644,
      "epoch": 5.728476821192053,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002574466880917201,
      "loss": 4.7892,
      "mean_token_accuracy": 0.22777802795171737,
      "num_tokens": 13260727.0,
      "step": 5625
    },
    {
      "entropy": 5.350016355514526,
      "epoch": 5.733571064696893,
      "grad_norm": 1.21875,
      "learning_rate": 0.00025704677069404854,
      "loss": 4.8148,
      "mean_token_accuracy": 0.21531449854373932,
      "num_tokens": 13270911.0,
      "step": 5630
    },
    {
      "entropy": 5.217638158798218,
      "epoch": 5.738665308201732,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00025664691036933745,
      "loss": 4.769,
      "mean_token_accuracy": 0.22838869094848632,
      "num_tokens": 13282499.0,
      "step": 5635
    },
    {
      "entropy": 5.302360391616821,
      "epoch": 5.743759551706572,
      "grad_norm": 1.171875,
      "learning_rate": 0.00025624710838873476,
      "loss": 4.8244,
      "mean_token_accuracy": 0.21737925708293915,
      "num_tokens": 13294274.0,
      "step": 5640
    },
    {
      "entropy": 5.3408896923065186,
      "epoch": 5.748853795211411,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002558473660232031,
      "loss": 4.753,
      "mean_token_accuracy": 0.22132869809865952,
      "num_tokens": 13305689.0,
      "step": 5645
    },
    {
      "entropy": 5.238768196105957,
      "epoch": 5.753948038716251,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00025544768454351547,
      "loss": 4.6792,
      "mean_token_accuracy": 0.2332421898841858,
      "num_tokens": 13316244.0,
      "step": 5650
    },
    {
      "entropy": 5.213983583450317,
      "epoch": 5.75904228222109,
      "grad_norm": 1.15625,
      "learning_rate": 0.00025504806522025147,
      "loss": 4.8307,
      "mean_token_accuracy": 0.21796021312475206,
      "num_tokens": 13328227.0,
      "step": 5655
    },
    {
      "entropy": 5.288554954528808,
      "epoch": 5.76413652572593,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00025464850932379305,
      "loss": 4.7829,
      "mean_token_accuracy": 0.2132395997643471,
      "num_tokens": 13340185.0,
      "step": 5660
    },
    {
      "entropy": 5.258321237564087,
      "epoch": 5.769230769230769,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002542490181243203,
      "loss": 4.8224,
      "mean_token_accuracy": 0.2161949023604393,
      "num_tokens": 13352519.0,
      "step": 5665
    },
    {
      "entropy": 5.240520429611206,
      "epoch": 5.774325012735609,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000253849592891808,
      "loss": 4.7701,
      "mean_token_accuracy": 0.2158770442008972,
      "num_tokens": 13364532.0,
      "step": 5670
    },
    {
      "entropy": 5.318478393554687,
      "epoch": 5.779419256240448,
      "grad_norm": 1.25,
      "learning_rate": 0.00025345023489602104,
      "loss": 4.8089,
      "mean_token_accuracy": 0.2176717609167099,
      "num_tokens": 13375566.0,
      "step": 5675
    },
    {
      "entropy": 5.163907194137574,
      "epoch": 5.784513499745287,
      "grad_norm": 1.265625,
      "learning_rate": 0.00025305094540651045,
      "loss": 4.5866,
      "mean_token_accuracy": 0.24697982966899873,
      "num_tokens": 13387149.0,
      "step": 5680
    },
    {
      "entropy": 5.20381875038147,
      "epoch": 5.789607743250127,
      "grad_norm": 1.1875,
      "learning_rate": 0.00025265172569260974,
      "loss": 4.7036,
      "mean_token_accuracy": 0.23056611865758897,
      "num_tokens": 13398945.0,
      "step": 5685
    },
    {
      "entropy": 5.230354928970337,
      "epoch": 5.7947019867549665,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002522525770234304,
      "loss": 4.7754,
      "mean_token_accuracy": 0.227412249147892,
      "num_tokens": 13411706.0,
      "step": 5690
    },
    {
      "entropy": 5.258226776123047,
      "epoch": 5.7997962302598065,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002518535006678582,
      "loss": 4.7165,
      "mean_token_accuracy": 0.23163807094097139,
      "num_tokens": 13423680.0,
      "step": 5695
    },
    {
      "entropy": 5.350976991653442,
      "epoch": 5.804890473764646,
      "grad_norm": 1.203125,
      "learning_rate": 0.00025145449789454907,
      "loss": 4.879,
      "mean_token_accuracy": 0.21114764362573624,
      "num_tokens": 13435569.0,
      "step": 5700
    },
    {
      "entropy": 5.289439725875854,
      "epoch": 5.809984717269486,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002510555699719246,
      "loss": 4.8002,
      "mean_token_accuracy": 0.21897833198308944,
      "num_tokens": 13446618.0,
      "step": 5705
    },
    {
      "entropy": 5.256340265274048,
      "epoch": 5.815078960774325,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002506567181681691,
      "loss": 4.7822,
      "mean_token_accuracy": 0.227226060628891,
      "num_tokens": 13458205.0,
      "step": 5710
    },
    {
      "entropy": 5.259782600402832,
      "epoch": 5.820173204279165,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002502579437512243,
      "loss": 4.7395,
      "mean_token_accuracy": 0.21904097944498063,
      "num_tokens": 13469185.0,
      "step": 5715
    },
    {
      "entropy": 5.286248445510864,
      "epoch": 5.825267447784004,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002498592479887863,
      "loss": 4.7673,
      "mean_token_accuracy": 0.21838839948177338,
      "num_tokens": 13480436.0,
      "step": 5720
    },
    {
      "entropy": 5.342421245574951,
      "epoch": 5.830361691288844,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00024946063214830106,
      "loss": 4.8162,
      "mean_token_accuracy": 0.22092859148979188,
      "num_tokens": 13491952.0,
      "step": 5725
    },
    {
      "entropy": 5.313463973999023,
      "epoch": 5.835455934793683,
      "grad_norm": 1.265625,
      "learning_rate": 0.00024906209749696036,
      "loss": 4.8117,
      "mean_token_accuracy": 0.22274213433265685,
      "num_tokens": 13502986.0,
      "step": 5730
    },
    {
      "entropy": 5.3031449794769285,
      "epoch": 5.840550178298523,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00024866364530169797,
      "loss": 4.7847,
      "mean_token_accuracy": 0.22466183453798294,
      "num_tokens": 13514137.0,
      "step": 5735
    },
    {
      "entropy": 5.2109987258911135,
      "epoch": 5.845644421803362,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002482652768291857,
      "loss": 4.7858,
      "mean_token_accuracy": 0.22091449648141862,
      "num_tokens": 13525927.0,
      "step": 5740
    },
    {
      "entropy": 5.2220865249633786,
      "epoch": 5.850738665308202,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002478669933458288,
      "loss": 4.7836,
      "mean_token_accuracy": 0.2217986986041069,
      "num_tokens": 13536564.0,
      "step": 5745
    },
    {
      "entropy": 5.292818164825439,
      "epoch": 5.855832908813041,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002474687961177626,
      "loss": 4.7617,
      "mean_token_accuracy": 0.2256689339876175,
      "num_tokens": 13548179.0,
      "step": 5750
    },
    {
      "entropy": 5.354149007797242,
      "epoch": 5.860927152317881,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00024707068641084836,
      "loss": 4.8788,
      "mean_token_accuracy": 0.20837071537971497,
      "num_tokens": 13561036.0,
      "step": 5755
    },
    {
      "entropy": 5.26949725151062,
      "epoch": 5.86602139582272,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002466726654906688,
      "loss": 4.7025,
      "mean_token_accuracy": 0.22802980095148087,
      "num_tokens": 13573027.0,
      "step": 5760
    },
    {
      "entropy": 5.185589933395386,
      "epoch": 5.871115639327559,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002462747346225246,
      "loss": 4.6396,
      "mean_token_accuracy": 0.2361936867237091,
      "num_tokens": 13583599.0,
      "step": 5765
    },
    {
      "entropy": 5.237279462814331,
      "epoch": 5.876209882832399,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002458768950714302,
      "loss": 4.7721,
      "mean_token_accuracy": 0.2247619777917862,
      "num_tokens": 13594110.0,
      "step": 5770
    },
    {
      "entropy": 5.195794773101807,
      "epoch": 5.8813041263372385,
      "grad_norm": 1.3125,
      "learning_rate": 0.00024547914810210946,
      "loss": 4.7941,
      "mean_token_accuracy": 0.2207624152302742,
      "num_tokens": 13606041.0,
      "step": 5775
    },
    {
      "entropy": 5.329722690582275,
      "epoch": 5.8863983698420785,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00024508149497899233,
      "loss": 4.7904,
      "mean_token_accuracy": 0.22202976644039155,
      "num_tokens": 13618757.0,
      "step": 5780
    },
    {
      "entropy": 5.383864974975586,
      "epoch": 5.891492613346918,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00024468393696621006,
      "loss": 4.9145,
      "mean_token_accuracy": 0.21322237253189086,
      "num_tokens": 13631346.0,
      "step": 5785
    },
    {
      "entropy": 5.302651214599609,
      "epoch": 5.896586856851758,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00024428647532759177,
      "loss": 4.8076,
      "mean_token_accuracy": 0.21845850348472595,
      "num_tokens": 13642759.0,
      "step": 5790
    },
    {
      "entropy": 5.294602155685425,
      "epoch": 5.901681100356597,
      "grad_norm": 1.28125,
      "learning_rate": 0.00024388911132666003,
      "loss": 4.7258,
      "mean_token_accuracy": 0.22733961641788483,
      "num_tokens": 13653338.0,
      "step": 5795
    },
    {
      "entropy": 5.294568872451782,
      "epoch": 5.906775343861437,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002434918462266273,
      "loss": 4.8515,
      "mean_token_accuracy": 0.21644322723150253,
      "num_tokens": 13667416.0,
      "step": 5800
    },
    {
      "entropy": 5.200868892669678,
      "epoch": 5.911869587366276,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00024309468129039125,
      "loss": 4.7596,
      "mean_token_accuracy": 0.22705030888319017,
      "num_tokens": 13679579.0,
      "step": 5805
    },
    {
      "entropy": 5.302225780487061,
      "epoch": 5.916963830871116,
      "grad_norm": 1.25,
      "learning_rate": 0.00024269761778053156,
      "loss": 4.8844,
      "mean_token_accuracy": 0.2103213593363762,
      "num_tokens": 13690074.0,
      "step": 5810
    },
    {
      "entropy": 5.4166850566864015,
      "epoch": 5.922058074375955,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002423006569593052,
      "loss": 4.8745,
      "mean_token_accuracy": 0.2118265300989151,
      "num_tokens": 13703593.0,
      "step": 5815
    },
    {
      "entropy": 5.32852692604065,
      "epoch": 5.927152317880795,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00024190380008864276,
      "loss": 4.8851,
      "mean_token_accuracy": 0.21185795962810516,
      "num_tokens": 13717667.0,
      "step": 5820
    },
    {
      "entropy": 5.219115734100342,
      "epoch": 5.932246561385634,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00024150704843014427,
      "loss": 4.6794,
      "mean_token_accuracy": 0.23125241845846176,
      "num_tokens": 13728759.0,
      "step": 5825
    },
    {
      "entropy": 5.3164019107818605,
      "epoch": 5.937340804890474,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002411104032450755,
      "loss": 4.8123,
      "mean_token_accuracy": 0.22004642486572265,
      "num_tokens": 13741711.0,
      "step": 5830
    },
    {
      "entropy": 5.308833408355713,
      "epoch": 5.942435048395313,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00024071386579436354,
      "loss": 4.862,
      "mean_token_accuracy": 0.21536983102560042,
      "num_tokens": 13754312.0,
      "step": 5835
    },
    {
      "entropy": 5.281599473953247,
      "epoch": 5.947529291900153,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00024031743733859325,
      "loss": 4.9021,
      "mean_token_accuracy": 0.21471806168556212,
      "num_tokens": 13765464.0,
      "step": 5840
    },
    {
      "entropy": 5.268782615661621,
      "epoch": 5.952623535404992,
      "grad_norm": 1.25,
      "learning_rate": 0.0002399211191380027,
      "loss": 4.7553,
      "mean_token_accuracy": 0.21858601421117782,
      "num_tokens": 13777131.0,
      "step": 5845
    },
    {
      "entropy": 5.217960691452026,
      "epoch": 5.9577177789098315,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002395249124524797,
      "loss": 4.7817,
      "mean_token_accuracy": 0.2162828728556633,
      "num_tokens": 13788637.0,
      "step": 5850
    },
    {
      "entropy": 5.336798429489136,
      "epoch": 5.9628120224146715,
      "grad_norm": 1.265625,
      "learning_rate": 0.00023912881854155733,
      "loss": 4.8261,
      "mean_token_accuracy": 0.21757743060588836,
      "num_tokens": 13799511.0,
      "step": 5855
    },
    {
      "entropy": 5.2992956161499025,
      "epoch": 5.967906265919511,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00023873283866441048,
      "loss": 4.745,
      "mean_token_accuracy": 0.2258385106921196,
      "num_tokens": 13810320.0,
      "step": 5860
    },
    {
      "entropy": 5.326899576187134,
      "epoch": 5.973000509424351,
      "grad_norm": 1.296875,
      "learning_rate": 0.00023833697407985117,
      "loss": 4.7854,
      "mean_token_accuracy": 0.22156974375247956,
      "num_tokens": 13822749.0,
      "step": 5865
    },
    {
      "entropy": 5.235223484039307,
      "epoch": 5.97809475292919,
      "grad_norm": 1.140625,
      "learning_rate": 0.00023794122604632505,
      "loss": 4.7886,
      "mean_token_accuracy": 0.2262983053922653,
      "num_tokens": 13834768.0,
      "step": 5870
    },
    {
      "entropy": 5.340647649765015,
      "epoch": 5.98318899643403,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00023754559582190744,
      "loss": 4.8409,
      "mean_token_accuracy": 0.21348068565130235,
      "num_tokens": 13846468.0,
      "step": 5875
    },
    {
      "entropy": 5.285129594802856,
      "epoch": 5.988283239938869,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002371500846642989,
      "loss": 4.723,
      "mean_token_accuracy": 0.2301824003458023,
      "num_tokens": 13858005.0,
      "step": 5880
    },
    {
      "entropy": 5.28271713256836,
      "epoch": 5.993377483443709,
      "grad_norm": 1.359375,
      "learning_rate": 0.00023675469383082148,
      "loss": 4.7971,
      "mean_token_accuracy": 0.22174069583415984,
      "num_tokens": 13868246.0,
      "step": 5885
    },
    {
      "entropy": 5.287574386596679,
      "epoch": 5.998471726948548,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00023635942457841497,
      "loss": 4.8062,
      "mean_token_accuracy": 0.22599100917577744,
      "num_tokens": 13880799.0,
      "step": 5890
    },
    {
      "entropy": 5.2326202392578125,
      "epoch": 6.003056546102903,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00023596427816363242,
      "loss": 4.5916,
      "mean_token_accuracy": 0.23881403439574772,
      "num_tokens": 13891936.0,
      "step": 5895
    },
    {
      "entropy": 5.322227525711059,
      "epoch": 6.008150789607743,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002355692558426365,
      "loss": 4.5724,
      "mean_token_accuracy": 0.23615602254867554,
      "num_tokens": 13903123.0,
      "step": 5900
    },
    {
      "entropy": 5.325913381576538,
      "epoch": 6.013245033112582,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00023517435887119526,
      "loss": 4.5874,
      "mean_token_accuracy": 0.23714303523302077,
      "num_tokens": 13914095.0,
      "step": 5905
    },
    {
      "entropy": 5.257232999801635,
      "epoch": 6.018339276617422,
      "grad_norm": 1.15625,
      "learning_rate": 0.00023477958850467835,
      "loss": 4.5867,
      "mean_token_accuracy": 0.23693043738603592,
      "num_tokens": 13926776.0,
      "step": 5910
    },
    {
      "entropy": 5.272213649749756,
      "epoch": 6.0234335201222615,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002343849459980531,
      "loss": 4.6287,
      "mean_token_accuracy": 0.23120100051164627,
      "num_tokens": 13938860.0,
      "step": 5915
    },
    {
      "entropy": 5.239475584030151,
      "epoch": 6.0285277636271015,
      "grad_norm": 1.203125,
      "learning_rate": 0.00023399043260588,
      "loss": 4.5493,
      "mean_token_accuracy": 0.24075593054294586,
      "num_tokens": 13950288.0,
      "step": 5920
    },
    {
      "entropy": 5.317878103256225,
      "epoch": 6.033622007131941,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00023359604958230945,
      "loss": 4.5927,
      "mean_token_accuracy": 0.23403551131486894,
      "num_tokens": 13961660.0,
      "step": 5925
    },
    {
      "entropy": 5.300948190689087,
      "epoch": 6.038716250636781,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00023320179818107735,
      "loss": 4.6834,
      "mean_token_accuracy": 0.22785708010196687,
      "num_tokens": 13975252.0,
      "step": 5930
    },
    {
      "entropy": 5.261181592941284,
      "epoch": 6.04381049414162,
      "grad_norm": 1.34375,
      "learning_rate": 0.00023280767965550087,
      "loss": 4.5341,
      "mean_token_accuracy": 0.24236685931682586,
      "num_tokens": 13986830.0,
      "step": 5935
    },
    {
      "entropy": 5.282742977142334,
      "epoch": 6.04890473764646,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00023241369525847522,
      "loss": 4.7051,
      "mean_token_accuracy": 0.22701049149036406,
      "num_tokens": 13999172.0,
      "step": 5940
    },
    {
      "entropy": 5.261336851119995,
      "epoch": 6.053998981151299,
      "grad_norm": 1.328125,
      "learning_rate": 0.00023201984624246887,
      "loss": 4.5983,
      "mean_token_accuracy": 0.23404674530029296,
      "num_tokens": 14009569.0,
      "step": 5945
    },
    {
      "entropy": 5.293906593322754,
      "epoch": 6.059093224656139,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00023162613385951998,
      "loss": 4.6449,
      "mean_token_accuracy": 0.23827873319387435,
      "num_tokens": 14021973.0,
      "step": 5950
    },
    {
      "entropy": 5.284627866744995,
      "epoch": 6.064187468160978,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00023123255936123254,
      "loss": 4.579,
      "mean_token_accuracy": 0.23358042538166046,
      "num_tokens": 14032288.0,
      "step": 5955
    },
    {
      "entropy": 5.289303207397461,
      "epoch": 6.069281711665818,
      "grad_norm": 1.15625,
      "learning_rate": 0.00023083912399877204,
      "loss": 4.6508,
      "mean_token_accuracy": 0.23085738271474837,
      "num_tokens": 14046567.0,
      "step": 5960
    },
    {
      "entropy": 5.191253709793091,
      "epoch": 6.074375955170657,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00023044582902286177,
      "loss": 4.5196,
      "mean_token_accuracy": 0.24215824604034425,
      "num_tokens": 14058368.0,
      "step": 5965
    },
    {
      "entropy": 5.268327760696411,
      "epoch": 6.079470198675497,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00023005267568377847,
      "loss": 4.6069,
      "mean_token_accuracy": 0.24315980523824693,
      "num_tokens": 14071174.0,
      "step": 5970
    },
    {
      "entropy": 5.300138854980469,
      "epoch": 6.084564442180336,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00022965966523134912,
      "loss": 4.5887,
      "mean_token_accuracy": 0.2303079307079315,
      "num_tokens": 14082817.0,
      "step": 5975
    },
    {
      "entropy": 5.225939989089966,
      "epoch": 6.089658685685175,
      "grad_norm": 1.09375,
      "learning_rate": 0.000229266798914946,
      "loss": 4.5987,
      "mean_token_accuracy": 0.2360409051179886,
      "num_tokens": 14094920.0,
      "step": 5980
    },
    {
      "entropy": 5.287862968444824,
      "epoch": 6.094752929190015,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002288740779834833,
      "loss": 4.675,
      "mean_token_accuracy": 0.23116572499275206,
      "num_tokens": 14106771.0,
      "step": 5985
    },
    {
      "entropy": 5.236420726776123,
      "epoch": 6.099847172694854,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00022848150368541314,
      "loss": 4.5167,
      "mean_token_accuracy": 0.24289385080337525,
      "num_tokens": 14117454.0,
      "step": 5990
    },
    {
      "entropy": 5.249839496612549,
      "epoch": 6.104941416199694,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002280890772687214,
      "loss": 4.5438,
      "mean_token_accuracy": 0.2436761274933815,
      "num_tokens": 14127861.0,
      "step": 5995
    },
    {
      "entropy": 5.143825531005859,
      "epoch": 6.1100356597045336,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00022769679998092398,
      "loss": 4.4812,
      "mean_token_accuracy": 0.24846928268671037,
      "num_tokens": 14139647.0,
      "step": 6000
    },
    {
      "epoch": 6.1100356597045336,
      "eval_entropy": 5.069519849471104,
      "eval_loss": 5.66260290145874,
      "eval_mean_token_accuracy": 0.1823776615070708,
      "eval_num_tokens": 14139647.0,
      "eval_runtime": 2.6691,
      "eval_samples_per_second": 1213.883,
      "eval_steps_per_second": 151.735,
      "step": 6000
    }
  ],
  "logging_steps": 5,
  "max_steps": 9810,
  "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": 2.2596763557888e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}