nld-latn-10mb-after-ppt-shuff-dyck-10mb-ckpt500_seed3407/checkpoint-9500/trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 8.628805088596092,
  "eval_steps": 500,
  "global_step": 9500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 4.856400108337402,
      "epoch": 0.004543389368468878,
      "grad_norm": 21.625,
      "learning_rate": 2e-06,
      "loss": 14.2106,
      "mean_token_accuracy": 0.0006149356486275792,
      "num_tokens": 10156.0,
      "step": 5
    },
    {
      "entropy": 4.860177421569825,
      "epoch": 0.009086778736937756,
      "grad_norm": 21.375,
      "learning_rate": 4.5e-06,
      "loss": 14.2151,
      "mean_token_accuracy": 0.0003687591990455985,
      "num_tokens": 20669.0,
      "step": 10
    },
    {
      "entropy": 4.876665878295898,
      "epoch": 0.013630168105406633,
      "grad_norm": 22.5,
      "learning_rate": 7e-06,
      "loss": 14.0374,
      "mean_token_accuracy": 0.000468046945752576,
      "num_tokens": 30952.0,
      "step": 15
    },
    {
      "entropy": 4.957574272155762,
      "epoch": 0.01817355747387551,
      "grad_norm": 27.75,
      "learning_rate": 9.5e-06,
      "loss": 13.7887,
      "mean_token_accuracy": 0.0006278944842051714,
      "num_tokens": 41429.0,
      "step": 20
    },
    {
      "entropy": 5.298445510864258,
      "epoch": 0.02271694684234439,
      "grad_norm": 35.0,
      "learning_rate": 1.2e-05,
      "loss": 13.1575,
      "mean_token_accuracy": 0.0006482779106590896,
      "num_tokens": 52166.0,
      "step": 25
    },
    {
      "entropy": 6.667004203796386,
      "epoch": 0.027260336210813266,
      "grad_norm": 36.25,
      "learning_rate": 1.4500000000000002e-05,
      "loss": 12.0099,
      "mean_token_accuracy": 0.00022935778833925723,
      "num_tokens": 61670.0,
      "step": 30
    },
    {
      "entropy": 10.085978126525879,
      "epoch": 0.03180372557928214,
      "grad_norm": 4.0,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 10.8036,
      "mean_token_accuracy": 0.006481384776998311,
      "num_tokens": 71702.0,
      "step": 35
    },
    {
      "entropy": 10.696837425231934,
      "epoch": 0.03634711494775102,
      "grad_norm": 3.5625,
      "learning_rate": 1.95e-05,
      "loss": 10.629,
      "mean_token_accuracy": 0.014822260290384293,
      "num_tokens": 82840.0,
      "step": 40
    },
    {
      "entropy": 10.674850368499756,
      "epoch": 0.0408905043162199,
      "grad_norm": 3.09375,
      "learning_rate": 2.2e-05,
      "loss": 10.4551,
      "mean_token_accuracy": 0.017661335971206428,
      "num_tokens": 92290.0,
      "step": 45
    },
    {
      "entropy": 10.59873275756836,
      "epoch": 0.04543389368468878,
      "grad_norm": 2.390625,
      "learning_rate": 2.4500000000000003e-05,
      "loss": 10.2805,
      "mean_token_accuracy": 0.017723128758370878,
      "num_tokens": 101842.0,
      "step": 50
    },
    {
      "entropy": 10.64423999786377,
      "epoch": 0.049977283053157656,
      "grad_norm": 2.140625,
      "learning_rate": 2.7e-05,
      "loss": 10.2044,
      "mean_token_accuracy": 0.013444735389202832,
      "num_tokens": 114030.0,
      "step": 55
    },
    {
      "entropy": 10.725485229492188,
      "epoch": 0.05452067242162653,
      "grad_norm": 2.0,
      "learning_rate": 2.95e-05,
      "loss": 10.0243,
      "mean_token_accuracy": 0.02563867699354887,
      "num_tokens": 124349.0,
      "step": 60
    },
    {
      "entropy": 10.725618076324462,
      "epoch": 0.059064061790095414,
      "grad_norm": 1.9453125,
      "learning_rate": 3.2e-05,
      "loss": 9.8587,
      "mean_token_accuracy": 0.03280576281249523,
      "num_tokens": 133534.0,
      "step": 65
    },
    {
      "entropy": 10.69280939102173,
      "epoch": 0.06360745115856428,
      "grad_norm": 2.15625,
      "learning_rate": 3.4500000000000005e-05,
      "loss": 9.8074,
      "mean_token_accuracy": 0.019261055253446104,
      "num_tokens": 144383.0,
      "step": 70
    },
    {
      "entropy": 10.707190322875977,
      "epoch": 0.06815084052703317,
      "grad_norm": 1.796875,
      "learning_rate": 3.7e-05,
      "loss": 9.7725,
      "mean_token_accuracy": 0.028312317933887242,
      "num_tokens": 155467.0,
      "step": 75
    },
    {
      "entropy": 10.719540405273438,
      "epoch": 0.07269422989550205,
      "grad_norm": 1.9375,
      "learning_rate": 3.95e-05,
      "loss": 9.6826,
      "mean_token_accuracy": 0.05074650049209595,
      "num_tokens": 165903.0,
      "step": 80
    },
    {
      "entropy": 10.707062530517579,
      "epoch": 0.07723761926397092,
      "grad_norm": 2.078125,
      "learning_rate": 4.2000000000000004e-05,
      "loss": 9.5739,
      "mean_token_accuracy": 0.04917050153017044,
      "num_tokens": 175812.0,
      "step": 85
    },
    {
      "entropy": 10.685445880889892,
      "epoch": 0.0817810086324398,
      "grad_norm": 1.859375,
      "learning_rate": 4.45e-05,
      "loss": 9.4801,
      "mean_token_accuracy": 0.04583706185221672,
      "num_tokens": 186357.0,
      "step": 90
    },
    {
      "entropy": 10.684220504760741,
      "epoch": 0.08632439800090867,
      "grad_norm": 1.9296875,
      "learning_rate": 4.7000000000000004e-05,
      "loss": 9.3766,
      "mean_token_accuracy": 0.04345137663185596,
      "num_tokens": 196124.0,
      "step": 95
    },
    {
      "entropy": 10.67547779083252,
      "epoch": 0.09086778736937756,
      "grad_norm": 1.9140625,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 9.2652,
      "mean_token_accuracy": 0.04514526352286339,
      "num_tokens": 205655.0,
      "step": 100
    },
    {
      "entropy": 10.653054904937743,
      "epoch": 0.09541117673784644,
      "grad_norm": 1.8125,
      "learning_rate": 5.2e-05,
      "loss": 9.1821,
      "mean_token_accuracy": 0.050565244257450105,
      "num_tokens": 215870.0,
      "step": 105
    },
    {
      "entropy": 10.627846431732177,
      "epoch": 0.09995456610631531,
      "grad_norm": 1.828125,
      "learning_rate": 5.45e-05,
      "loss": 9.0638,
      "mean_token_accuracy": 0.05385896116495133,
      "num_tokens": 225027.0,
      "step": 110
    },
    {
      "entropy": 10.588682270050048,
      "epoch": 0.10449795547478419,
      "grad_norm": 1.78125,
      "learning_rate": 5.7e-05,
      "loss": 8.9803,
      "mean_token_accuracy": 0.05429651960730553,
      "num_tokens": 236352.0,
      "step": 115
    },
    {
      "entropy": 10.560179710388184,
      "epoch": 0.10904134484325306,
      "grad_norm": 1.75,
      "learning_rate": 5.9499999999999996e-05,
      "loss": 8.8435,
      "mean_token_accuracy": 0.0539657898247242,
      "num_tokens": 246713.0,
      "step": 120
    },
    {
      "entropy": 10.498886871337891,
      "epoch": 0.11358473421172194,
      "grad_norm": 1.7109375,
      "learning_rate": 6.2e-05,
      "loss": 8.6823,
      "mean_token_accuracy": 0.05030508264899254,
      "num_tokens": 256950.0,
      "step": 125
    },
    {
      "entropy": 10.440382480621338,
      "epoch": 0.11812812358019083,
      "grad_norm": 1.6484375,
      "learning_rate": 6.450000000000001e-05,
      "loss": 8.6331,
      "mean_token_accuracy": 0.05017514526844025,
      "num_tokens": 267345.0,
      "step": 130
    },
    {
      "entropy": 10.351335620880127,
      "epoch": 0.1226715129486597,
      "grad_norm": 2.0,
      "learning_rate": 6.7e-05,
      "loss": 8.4422,
      "mean_token_accuracy": 0.054022491723299024,
      "num_tokens": 276782.0,
      "step": 135
    },
    {
      "entropy": 10.266972255706786,
      "epoch": 0.12721490231712856,
      "grad_norm": 1.59375,
      "learning_rate": 6.950000000000001e-05,
      "loss": 8.3194,
      "mean_token_accuracy": 0.05632990375161171,
      "num_tokens": 287024.0,
      "step": 140
    },
    {
      "entropy": 10.132027339935302,
      "epoch": 0.13175829168559747,
      "grad_norm": 1.359375,
      "learning_rate": 7.2e-05,
      "loss": 8.2954,
      "mean_token_accuracy": 0.05156242474913597,
      "num_tokens": 296597.0,
      "step": 145
    },
    {
      "entropy": 10.018408966064452,
      "epoch": 0.13630168105406634,
      "grad_norm": 1.453125,
      "learning_rate": 7.45e-05,
      "loss": 8.1037,
      "mean_token_accuracy": 0.05491906628012657,
      "num_tokens": 307709.0,
      "step": 150
    },
    {
      "entropy": 9.880233860015869,
      "epoch": 0.14084507042253522,
      "grad_norm": 1.25,
      "learning_rate": 7.7e-05,
      "loss": 8.0694,
      "mean_token_accuracy": 0.05425432473421097,
      "num_tokens": 317976.0,
      "step": 155
    },
    {
      "entropy": 9.71201810836792,
      "epoch": 0.1453884597910041,
      "grad_norm": 1.203125,
      "learning_rate": 7.950000000000001e-05,
      "loss": 7.8984,
      "mean_token_accuracy": 0.054642430320382115,
      "num_tokens": 327319.0,
      "step": 160
    },
    {
      "entropy": 9.520874309539796,
      "epoch": 0.14993184915947297,
      "grad_norm": 1.1796875,
      "learning_rate": 8.2e-05,
      "loss": 7.8208,
      "mean_token_accuracy": 0.05670260861515999,
      "num_tokens": 337337.0,
      "step": 165
    },
    {
      "entropy": 9.264504623413085,
      "epoch": 0.15447523852794184,
      "grad_norm": 1.140625,
      "learning_rate": 8.450000000000001e-05,
      "loss": 7.7862,
      "mean_token_accuracy": 0.0514026153832674,
      "num_tokens": 347779.0,
      "step": 170
    },
    {
      "entropy": 9.049597454071044,
      "epoch": 0.15901862789641072,
      "grad_norm": 0.90234375,
      "learning_rate": 8.7e-05,
      "loss": 7.7529,
      "mean_token_accuracy": 0.06217045783996582,
      "num_tokens": 358088.0,
      "step": 175
    },
    {
      "entropy": 8.819415187835693,
      "epoch": 0.1635620172648796,
      "grad_norm": 0.9921875,
      "learning_rate": 8.95e-05,
      "loss": 7.6422,
      "mean_token_accuracy": 0.059666012227535245,
      "num_tokens": 368314.0,
      "step": 180
    },
    {
      "entropy": 8.560069179534912,
      "epoch": 0.16810540663334847,
      "grad_norm": 0.828125,
      "learning_rate": 9.2e-05,
      "loss": 7.5159,
      "mean_token_accuracy": 0.07182336524128914,
      "num_tokens": 378406.0,
      "step": 185
    },
    {
      "entropy": 8.396781253814698,
      "epoch": 0.17264879600181735,
      "grad_norm": 0.78125,
      "learning_rate": 9.45e-05,
      "loss": 7.5248,
      "mean_token_accuracy": 0.07036102451384067,
      "num_tokens": 388912.0,
      "step": 190
    },
    {
      "entropy": 8.250075912475586,
      "epoch": 0.17719218537028622,
      "grad_norm": 0.828125,
      "learning_rate": 9.7e-05,
      "loss": 7.5042,
      "mean_token_accuracy": 0.07041607536375523,
      "num_tokens": 399290.0,
      "step": 195
    },
    {
      "entropy": 8.157392597198486,
      "epoch": 0.18173557473875512,
      "grad_norm": 0.96484375,
      "learning_rate": 9.95e-05,
      "loss": 7.4089,
      "mean_token_accuracy": 0.07651857286691666,
      "num_tokens": 409757.0,
      "step": 200
    },
    {
      "entropy": 8.099771499633789,
      "epoch": 0.186278964107224,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000102,
      "loss": 7.5407,
      "mean_token_accuracy": 0.07969405055046082,
      "num_tokens": 420182.0,
      "step": 205
    },
    {
      "entropy": 8.01094584465027,
      "epoch": 0.19082235347569287,
      "grad_norm": 0.90625,
      "learning_rate": 0.00010449999999999999,
      "loss": 7.4483,
      "mean_token_accuracy": 0.07581862695515156,
      "num_tokens": 429897.0,
      "step": 210
    },
    {
      "entropy": 7.986596918106079,
      "epoch": 0.19536574284416175,
      "grad_norm": 0.77734375,
      "learning_rate": 0.000107,
      "loss": 7.3481,
      "mean_token_accuracy": 0.07829344719648361,
      "num_tokens": 441131.0,
      "step": 215
    },
    {
      "entropy": 7.942036485671997,
      "epoch": 0.19990913221263062,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0001095,
      "loss": 7.3161,
      "mean_token_accuracy": 0.0775233581662178,
      "num_tokens": 451894.0,
      "step": 220
    },
    {
      "entropy": 7.821108770370484,
      "epoch": 0.2044525215810995,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000112,
      "loss": 7.2593,
      "mean_token_accuracy": 0.07877892255783081,
      "num_tokens": 462613.0,
      "step": 225
    },
    {
      "entropy": 7.904732894897461,
      "epoch": 0.20899591094956838,
      "grad_norm": 0.953125,
      "learning_rate": 0.0001145,
      "loss": 7.3536,
      "mean_token_accuracy": 0.07972405850887299,
      "num_tokens": 471606.0,
      "step": 230
    },
    {
      "entropy": 7.824862718582153,
      "epoch": 0.21353930031803725,
      "grad_norm": 0.81640625,
      "learning_rate": 0.00011700000000000001,
      "loss": 7.2458,
      "mean_token_accuracy": 0.08337627723813057,
      "num_tokens": 481559.0,
      "step": 235
    },
    {
      "entropy": 7.791320323944092,
      "epoch": 0.21808268968650613,
      "grad_norm": 1.03125,
      "learning_rate": 0.00011949999999999999,
      "loss": 7.2611,
      "mean_token_accuracy": 0.08033445328474045,
      "num_tokens": 492766.0,
      "step": 240
    },
    {
      "entropy": 7.805171489715576,
      "epoch": 0.222626079054975,
      "grad_norm": 0.80859375,
      "learning_rate": 0.000122,
      "loss": 7.3194,
      "mean_token_accuracy": 0.08032750971615314,
      "num_tokens": 503357.0,
      "step": 245
    },
    {
      "entropy": 7.816235494613648,
      "epoch": 0.22716946842344388,
      "grad_norm": 0.77734375,
      "learning_rate": 0.0001245,
      "loss": 7.1556,
      "mean_token_accuracy": 0.08248411491513252,
      "num_tokens": 513098.0,
      "step": 250
    },
    {
      "entropy": 7.719041204452514,
      "epoch": 0.23171285779191278,
      "grad_norm": 1.046875,
      "learning_rate": 0.000127,
      "loss": 7.2404,
      "mean_token_accuracy": 0.08086299896240234,
      "num_tokens": 522807.0,
      "step": 255
    },
    {
      "entropy": 7.781749677658081,
      "epoch": 0.23625624716038165,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001295,
      "loss": 7.259,
      "mean_token_accuracy": 0.08047396913170815,
      "num_tokens": 532670.0,
      "step": 260
    },
    {
      "entropy": 7.758613204956054,
      "epoch": 0.24079963652885053,
      "grad_norm": 0.71875,
      "learning_rate": 0.000132,
      "loss": 7.2064,
      "mean_token_accuracy": 0.07659858912229538,
      "num_tokens": 543647.0,
      "step": 265
    },
    {
      "entropy": 7.7667632579803465,
      "epoch": 0.2453430258973194,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00013450000000000002,
      "loss": 7.2166,
      "mean_token_accuracy": 0.08752154409885407,
      "num_tokens": 553047.0,
      "step": 270
    },
    {
      "entropy": 7.720886993408203,
      "epoch": 0.24988641526578828,
      "grad_norm": 0.79296875,
      "learning_rate": 0.00013700000000000002,
      "loss": 7.2571,
      "mean_token_accuracy": 0.0864359587430954,
      "num_tokens": 563413.0,
      "step": 275
    },
    {
      "entropy": 7.675430965423584,
      "epoch": 0.25442980463425713,
      "grad_norm": 0.921875,
      "learning_rate": 0.0001395,
      "loss": 7.2182,
      "mean_token_accuracy": 0.08417232409119606,
      "num_tokens": 573790.0,
      "step": 280
    },
    {
      "entropy": 7.737106895446777,
      "epoch": 0.25897319400272606,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00014199999999999998,
      "loss": 7.1452,
      "mean_token_accuracy": 0.08157877773046493,
      "num_tokens": 582958.0,
      "step": 285
    },
    {
      "entropy": 7.618109798431396,
      "epoch": 0.26351658337119493,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0001445,
      "loss": 7.0887,
      "mean_token_accuracy": 0.09229333996772766,
      "num_tokens": 592886.0,
      "step": 290
    },
    {
      "entropy": 7.615639877319336,
      "epoch": 0.2680599727396638,
      "grad_norm": 0.921875,
      "learning_rate": 0.000147,
      "loss": 7.1444,
      "mean_token_accuracy": 0.08253698348999024,
      "num_tokens": 603692.0,
      "step": 295
    },
    {
      "entropy": 7.630686283111572,
      "epoch": 0.2726033621081327,
      "grad_norm": 1.015625,
      "learning_rate": 0.0001495,
      "loss": 7.1779,
      "mean_token_accuracy": 0.08081393018364906,
      "num_tokens": 613534.0,
      "step": 300
    },
    {
      "entropy": 7.684085559844971,
      "epoch": 0.27714675147660156,
      "grad_norm": 1.0625,
      "learning_rate": 0.000152,
      "loss": 7.1265,
      "mean_token_accuracy": 0.0869300678372383,
      "num_tokens": 623862.0,
      "step": 305
    },
    {
      "entropy": 7.708980464935303,
      "epoch": 0.28169014084507044,
      "grad_norm": 1.21875,
      "learning_rate": 0.00015450000000000001,
      "loss": 7.2041,
      "mean_token_accuracy": 0.0886037103831768,
      "num_tokens": 633759.0,
      "step": 310
    },
    {
      "entropy": 7.514765357971191,
      "epoch": 0.2862335302135393,
      "grad_norm": 0.859375,
      "learning_rate": 0.000157,
      "loss": 7.0843,
      "mean_token_accuracy": 0.0886026717722416,
      "num_tokens": 643553.0,
      "step": 315
    },
    {
      "entropy": 7.49160099029541,
      "epoch": 0.2907769195820082,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0001595,
      "loss": 7.0034,
      "mean_token_accuracy": 0.08772724494338036,
      "num_tokens": 653714.0,
      "step": 320
    },
    {
      "entropy": 7.63023567199707,
      "epoch": 0.29532030895047706,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000162,
      "loss": 7.033,
      "mean_token_accuracy": 0.09123715907335281,
      "num_tokens": 662817.0,
      "step": 325
    },
    {
      "entropy": 7.539647436141967,
      "epoch": 0.29986369831894594,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00016450000000000001,
      "loss": 7.0547,
      "mean_token_accuracy": 0.0879720188677311,
      "num_tokens": 672381.0,
      "step": 330
    },
    {
      "entropy": 7.440394639968872,
      "epoch": 0.3044070876874148,
      "grad_norm": 0.984375,
      "learning_rate": 0.00016700000000000002,
      "loss": 6.9355,
      "mean_token_accuracy": 0.08899887949228287,
      "num_tokens": 682525.0,
      "step": 335
    },
    {
      "entropy": 7.532028865814209,
      "epoch": 0.3089504770558837,
      "grad_norm": 1.0625,
      "learning_rate": 0.00016950000000000003,
      "loss": 7.0283,
      "mean_token_accuracy": 0.08837513625621796,
      "num_tokens": 691826.0,
      "step": 340
    },
    {
      "entropy": 7.473609542846679,
      "epoch": 0.31349386642435256,
      "grad_norm": 1.1875,
      "learning_rate": 0.00017199999999999998,
      "loss": 7.007,
      "mean_token_accuracy": 0.0950367122888565,
      "num_tokens": 701319.0,
      "step": 345
    },
    {
      "entropy": 7.459249973297119,
      "epoch": 0.31803725579282144,
      "grad_norm": 1.453125,
      "learning_rate": 0.00017449999999999999,
      "loss": 7.0947,
      "mean_token_accuracy": 0.08535790592432022,
      "num_tokens": 710888.0,
      "step": 350
    },
    {
      "entropy": 7.56165566444397,
      "epoch": 0.3225806451612903,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000177,
      "loss": 7.0342,
      "mean_token_accuracy": 0.09012823775410653,
      "num_tokens": 722516.0,
      "step": 355
    },
    {
      "entropy": 7.527815628051758,
      "epoch": 0.3271240345297592,
      "grad_norm": 1.65625,
      "learning_rate": 0.0001795,
      "loss": 7.0169,
      "mean_token_accuracy": 0.08682614788413048,
      "num_tokens": 733058.0,
      "step": 360
    },
    {
      "entropy": 7.45044322013855,
      "epoch": 0.33166742389822806,
      "grad_norm": 1.15625,
      "learning_rate": 0.000182,
      "loss": 7.0592,
      "mean_token_accuracy": 0.09242936447262765,
      "num_tokens": 744832.0,
      "step": 365
    },
    {
      "entropy": 7.463720369338989,
      "epoch": 0.33621081326669694,
      "grad_norm": 1.078125,
      "learning_rate": 0.0001845,
      "loss": 7.0049,
      "mean_token_accuracy": 0.09356781840324402,
      "num_tokens": 755387.0,
      "step": 370
    },
    {
      "entropy": 7.43138747215271,
      "epoch": 0.3407542026351658,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000187,
      "loss": 6.9355,
      "mean_token_accuracy": 0.09147335067391396,
      "num_tokens": 764455.0,
      "step": 375
    },
    {
      "entropy": 7.4005152702331545,
      "epoch": 0.3452975920036347,
      "grad_norm": 1.203125,
      "learning_rate": 0.0001895,
      "loss": 6.98,
      "mean_token_accuracy": 0.09236079826951027,
      "num_tokens": 774513.0,
      "step": 380
    },
    {
      "entropy": 7.440145206451416,
      "epoch": 0.34984098137210357,
      "grad_norm": 1.0,
      "learning_rate": 0.000192,
      "loss": 7.0819,
      "mean_token_accuracy": 0.08978344351053238,
      "num_tokens": 785540.0,
      "step": 385
    },
    {
      "entropy": 7.4839293479919435,
      "epoch": 0.35438437074057244,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0001945,
      "loss": 6.947,
      "mean_token_accuracy": 0.09177564308047295,
      "num_tokens": 795233.0,
      "step": 390
    },
    {
      "entropy": 7.431085395812988,
      "epoch": 0.35892776010904137,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00019700000000000002,
      "loss": 6.9969,
      "mean_token_accuracy": 0.09167078733444214,
      "num_tokens": 804855.0,
      "step": 395
    },
    {
      "entropy": 7.446485424041748,
      "epoch": 0.36347114947751025,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00019950000000000002,
      "loss": 6.9593,
      "mean_token_accuracy": 0.0934673197567463,
      "num_tokens": 814818.0,
      "step": 400
    },
    {
      "entropy": 7.352738714218139,
      "epoch": 0.3680145388459791,
      "grad_norm": 1.125,
      "learning_rate": 0.000202,
      "loss": 6.9751,
      "mean_token_accuracy": 0.09469850584864617,
      "num_tokens": 824622.0,
      "step": 405
    },
    {
      "entropy": 7.417457246780396,
      "epoch": 0.372557928214448,
      "grad_norm": 1.40625,
      "learning_rate": 0.00020449999999999998,
      "loss": 6.8664,
      "mean_token_accuracy": 0.09812097549438477,
      "num_tokens": 834813.0,
      "step": 410
    },
    {
      "entropy": 7.318056726455689,
      "epoch": 0.3771013175829169,
      "grad_norm": 0.84765625,
      "learning_rate": 0.000207,
      "loss": 6.9516,
      "mean_token_accuracy": 0.09538300186395646,
      "num_tokens": 844887.0,
      "step": 415
    },
    {
      "entropy": 7.351348257064819,
      "epoch": 0.38164470695138575,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0002095,
      "loss": 6.9529,
      "mean_token_accuracy": 0.0924244724214077,
      "num_tokens": 855430.0,
      "step": 420
    },
    {
      "entropy": 7.276928663253784,
      "epoch": 0.3861880963198546,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000212,
      "loss": 6.794,
      "mean_token_accuracy": 0.09646925926208497,
      "num_tokens": 866052.0,
      "step": 425
    },
    {
      "entropy": 7.296294832229615,
      "epoch": 0.3907314856883235,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002145,
      "loss": 6.9297,
      "mean_token_accuracy": 0.09176526665687561,
      "num_tokens": 877961.0,
      "step": 430
    },
    {
      "entropy": 7.363173007965088,
      "epoch": 0.3952748750567924,
      "grad_norm": 0.87890625,
      "learning_rate": 0.00021700000000000002,
      "loss": 6.8984,
      "mean_token_accuracy": 0.1007479526102543,
      "num_tokens": 888386.0,
      "step": 435
    },
    {
      "entropy": 7.295883321762085,
      "epoch": 0.39981826442526125,
      "grad_norm": 1.125,
      "learning_rate": 0.0002195,
      "loss": 6.9439,
      "mean_token_accuracy": 0.09971338957548141,
      "num_tokens": 898272.0,
      "step": 440
    },
    {
      "entropy": 7.353458261489868,
      "epoch": 0.4043616537937301,
      "grad_norm": 1.03125,
      "learning_rate": 0.000222,
      "loss": 6.9055,
      "mean_token_accuracy": 0.10149330794811248,
      "num_tokens": 908317.0,
      "step": 445
    },
    {
      "entropy": 7.281500577926636,
      "epoch": 0.408905043162199,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002245,
      "loss": 6.867,
      "mean_token_accuracy": 0.09293231442570686,
      "num_tokens": 917970.0,
      "step": 450
    },
    {
      "entropy": 7.282706642150879,
      "epoch": 0.4134484325306679,
      "grad_norm": 1.171875,
      "learning_rate": 0.00022700000000000002,
      "loss": 6.9115,
      "mean_token_accuracy": 0.0950246959924698,
      "num_tokens": 928132.0,
      "step": 455
    },
    {
      "entropy": 7.30927529335022,
      "epoch": 0.41799182189913675,
      "grad_norm": 1.21875,
      "learning_rate": 0.00022950000000000002,
      "loss": 6.8424,
      "mean_token_accuracy": 0.09944264888763428,
      "num_tokens": 938092.0,
      "step": 460
    },
    {
      "entropy": 7.37966022491455,
      "epoch": 0.4225352112676056,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00023200000000000003,
      "loss": 6.9599,
      "mean_token_accuracy": 0.09508244097232818,
      "num_tokens": 948650.0,
      "step": 465
    },
    {
      "entropy": 7.15450644493103,
      "epoch": 0.4270786006360745,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00023449999999999998,
      "loss": 6.868,
      "mean_token_accuracy": 0.10101077854633331,
      "num_tokens": 960195.0,
      "step": 470
    },
    {
      "entropy": 7.254466104507446,
      "epoch": 0.4316219900045434,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000237,
      "loss": 6.8956,
      "mean_token_accuracy": 0.10054716616868972,
      "num_tokens": 970608.0,
      "step": 475
    },
    {
      "entropy": 7.369555997848511,
      "epoch": 0.43616537937301225,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002395,
      "loss": 6.8689,
      "mean_token_accuracy": 0.09615805298089981,
      "num_tokens": 981053.0,
      "step": 480
    },
    {
      "entropy": 7.163026285171509,
      "epoch": 0.4407087687414811,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000242,
      "loss": 6.7827,
      "mean_token_accuracy": 0.10568929314613343,
      "num_tokens": 991560.0,
      "step": 485
    },
    {
      "entropy": 7.18547511100769,
      "epoch": 0.44525215810995,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002445,
      "loss": 6.8472,
      "mean_token_accuracy": 0.09352557733654976,
      "num_tokens": 1002008.0,
      "step": 490
    },
    {
      "entropy": 7.173881673812867,
      "epoch": 0.4497955474784189,
      "grad_norm": 1.109375,
      "learning_rate": 0.000247,
      "loss": 6.8306,
      "mean_token_accuracy": 0.10517893880605697,
      "num_tokens": 1012349.0,
      "step": 495
    },
    {
      "entropy": 7.266806602478027,
      "epoch": 0.45433893684688775,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002495,
      "loss": 6.8159,
      "mean_token_accuracy": 0.10154995843768119,
      "num_tokens": 1022886.0,
      "step": 500
    },
    {
      "epoch": 0.45433893684688775,
      "eval_entropy": 7.165732025545696,
      "eval_loss": 6.923176288604736,
      "eval_mean_token_accuracy": 0.10160443299385004,
      "eval_num_tokens": 1022886.0,
      "eval_runtime": 2.056,
      "eval_samples_per_second": 1670.696,
      "eval_steps_per_second": 209.141,
      "step": 500
    },
    {
      "entropy": 7.297193765640259,
      "epoch": 0.45888232621535663,
      "grad_norm": 1.03125,
      "learning_rate": 0.000252,
      "loss": 6.8619,
      "mean_token_accuracy": 0.09589311704039574,
      "num_tokens": 1033129.0,
      "step": 505
    },
    {
      "entropy": 7.214840221405029,
      "epoch": 0.46342571558382556,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002545,
      "loss": 6.7604,
      "mean_token_accuracy": 0.10675142779946327,
      "num_tokens": 1043069.0,
      "step": 510
    },
    {
      "entropy": 7.178305196762085,
      "epoch": 0.46796910495229443,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000257,
      "loss": 6.785,
      "mean_token_accuracy": 0.10303612574934959,
      "num_tokens": 1054049.0,
      "step": 515
    },
    {
      "entropy": 7.15740852355957,
      "epoch": 0.4725124943207633,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002595,
      "loss": 6.772,
      "mean_token_accuracy": 0.09831079840660095,
      "num_tokens": 1065155.0,
      "step": 520
    },
    {
      "entropy": 7.140080070495605,
      "epoch": 0.4770558836892322,
      "grad_norm": 1.09375,
      "learning_rate": 0.000262,
      "loss": 6.7699,
      "mean_token_accuracy": 0.1037091501057148,
      "num_tokens": 1075510.0,
      "step": 525
    },
    {
      "entropy": 7.111771297454834,
      "epoch": 0.48159927305770106,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00026450000000000003,
      "loss": 6.7653,
      "mean_token_accuracy": 0.10416601449251175,
      "num_tokens": 1086494.0,
      "step": 530
    },
    {
      "entropy": 7.242225551605225,
      "epoch": 0.48614266242616994,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00026700000000000004,
      "loss": 6.8098,
      "mean_token_accuracy": 0.10401513352990151,
      "num_tokens": 1097311.0,
      "step": 535
    },
    {
      "entropy": 7.118653678894043,
      "epoch": 0.4906860517946388,
      "grad_norm": 1.375,
      "learning_rate": 0.00026950000000000005,
      "loss": 6.7668,
      "mean_token_accuracy": 0.10299587845802308,
      "num_tokens": 1107757.0,
      "step": 540
    },
    {
      "entropy": 7.0831122398376465,
      "epoch": 0.4952294411631077,
      "grad_norm": 0.953125,
      "learning_rate": 0.00027200000000000005,
      "loss": 6.6761,
      "mean_token_accuracy": 0.10623887702822685,
      "num_tokens": 1118889.0,
      "step": 545
    },
    {
      "entropy": 7.184380054473877,
      "epoch": 0.49977283053157656,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002745,
      "loss": 6.8628,
      "mean_token_accuracy": 0.0998254045844078,
      "num_tokens": 1128951.0,
      "step": 550
    },
    {
      "entropy": 7.096487092971802,
      "epoch": 0.5043162199000454,
      "grad_norm": 1.234375,
      "learning_rate": 0.000277,
      "loss": 6.6669,
      "mean_token_accuracy": 0.10797309428453446,
      "num_tokens": 1138583.0,
      "step": 555
    },
    {
      "entropy": 7.1161802291870115,
      "epoch": 0.5088596092685143,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002795,
      "loss": 6.7456,
      "mean_token_accuracy": 0.103419478982687,
      "num_tokens": 1149572.0,
      "step": 560
    },
    {
      "entropy": 7.167421865463257,
      "epoch": 0.5134029986369832,
      "grad_norm": 0.953125,
      "learning_rate": 0.00028199999999999997,
      "loss": 6.7063,
      "mean_token_accuracy": 0.10864525139331818,
      "num_tokens": 1160532.0,
      "step": 565
    },
    {
      "entropy": 7.055079984664917,
      "epoch": 0.5179463880054521,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002845,
      "loss": 6.7034,
      "mean_token_accuracy": 0.10719505548477173,
      "num_tokens": 1170915.0,
      "step": 570
    },
    {
      "entropy": 7.016624212265015,
      "epoch": 0.522489777373921,
      "grad_norm": 1.109375,
      "learning_rate": 0.000287,
      "loss": 6.6622,
      "mean_token_accuracy": 0.1151500552892685,
      "num_tokens": 1180822.0,
      "step": 575
    },
    {
      "entropy": 7.104376125335693,
      "epoch": 0.5270331667423899,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002895,
      "loss": 6.7169,
      "mean_token_accuracy": 0.10622723251581193,
      "num_tokens": 1190447.0,
      "step": 580
    },
    {
      "entropy": 7.050690603256226,
      "epoch": 0.5315765561108587,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000292,
      "loss": 6.745,
      "mean_token_accuracy": 0.10602018609642982,
      "num_tokens": 1200951.0,
      "step": 585
    },
    {
      "entropy": 7.107309436798095,
      "epoch": 0.5361199454793276,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002945,
      "loss": 6.6341,
      "mean_token_accuracy": 0.11055795922875404,
      "num_tokens": 1211846.0,
      "step": 590
    },
    {
      "entropy": 6.894921445846558,
      "epoch": 0.5406633348477965,
      "grad_norm": 1.015625,
      "learning_rate": 0.000297,
      "loss": 6.5876,
      "mean_token_accuracy": 0.1088622160255909,
      "num_tokens": 1222074.0,
      "step": 595
    },
    {
      "entropy": 6.963123083114624,
      "epoch": 0.5452067242162654,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002995,
      "loss": 6.6548,
      "mean_token_accuracy": 0.10943065285682678,
      "num_tokens": 1231741.0,
      "step": 600
    },
    {
      "entropy": 6.993402624130249,
      "epoch": 0.5497501135847342,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000302,
      "loss": 6.5906,
      "mean_token_accuracy": 0.10523484498262406,
      "num_tokens": 1241078.0,
      "step": 605
    },
    {
      "entropy": 7.046900510787964,
      "epoch": 0.5542935029532031,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003045,
      "loss": 6.7859,
      "mean_token_accuracy": 0.10277008563280106,
      "num_tokens": 1251685.0,
      "step": 610
    },
    {
      "entropy": 6.997891521453857,
      "epoch": 0.558836892321672,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000307,
      "loss": 6.6293,
      "mean_token_accuracy": 0.11614568382501603,
      "num_tokens": 1262044.0,
      "step": 615
    },
    {
      "entropy": 6.896301555633545,
      "epoch": 0.5633802816901409,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003095,
      "loss": 6.6362,
      "mean_token_accuracy": 0.10905065536499023,
      "num_tokens": 1273647.0,
      "step": 620
    },
    {
      "entropy": 7.149247694015503,
      "epoch": 0.5679236710586097,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000312,
      "loss": 6.6994,
      "mean_token_accuracy": 0.10467459112405778,
      "num_tokens": 1284396.0,
      "step": 625
    },
    {
      "entropy": 6.961115026473999,
      "epoch": 0.5724670604270786,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003145,
      "loss": 6.6689,
      "mean_token_accuracy": 0.10689816027879714,
      "num_tokens": 1294339.0,
      "step": 630
    },
    {
      "entropy": 7.002821254730224,
      "epoch": 0.5770104497955475,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000317,
      "loss": 6.597,
      "mean_token_accuracy": 0.11088685020804405,
      "num_tokens": 1304572.0,
      "step": 635
    },
    {
      "entropy": 6.956146478652954,
      "epoch": 0.5815538391640164,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003195,
      "loss": 6.6421,
      "mean_token_accuracy": 0.11130892485380173,
      "num_tokens": 1314780.0,
      "step": 640
    },
    {
      "entropy": 6.921744728088379,
      "epoch": 0.5860972285324852,
      "grad_norm": 1.125,
      "learning_rate": 0.000322,
      "loss": 6.64,
      "mean_token_accuracy": 0.10691023096442223,
      "num_tokens": 1325215.0,
      "step": 645
    },
    {
      "entropy": 6.905644273757934,
      "epoch": 0.5906406179009541,
      "grad_norm": 1.171875,
      "learning_rate": 0.00032450000000000003,
      "loss": 6.5648,
      "mean_token_accuracy": 0.11088442802429199,
      "num_tokens": 1334745.0,
      "step": 650
    },
    {
      "entropy": 7.0158445835113525,
      "epoch": 0.595184007269423,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00032700000000000003,
      "loss": 6.6651,
      "mean_token_accuracy": 0.11198936477303505,
      "num_tokens": 1345150.0,
      "step": 655
    },
    {
      "entropy": 6.971886682510376,
      "epoch": 0.5997273966378919,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00032950000000000004,
      "loss": 6.6206,
      "mean_token_accuracy": 0.10981507077813149,
      "num_tokens": 1355619.0,
      "step": 660
    },
    {
      "entropy": 6.895886754989624,
      "epoch": 0.6042707860063607,
      "grad_norm": 1.078125,
      "learning_rate": 0.00033200000000000005,
      "loss": 6.5486,
      "mean_token_accuracy": 0.11562549024820327,
      "num_tokens": 1365933.0,
      "step": 665
    },
    {
      "entropy": 6.931118965148926,
      "epoch": 0.6088141753748296,
      "grad_norm": 0.953125,
      "learning_rate": 0.00033450000000000005,
      "loss": 6.5762,
      "mean_token_accuracy": 0.11562452912330627,
      "num_tokens": 1376867.0,
      "step": 670
    },
    {
      "entropy": 6.992132759094238,
      "epoch": 0.6133575647432985,
      "grad_norm": 1.28125,
      "learning_rate": 0.000337,
      "loss": 6.7267,
      "mean_token_accuracy": 0.11105038151144982,
      "num_tokens": 1386739.0,
      "step": 675
    },
    {
      "entropy": 6.933910131454468,
      "epoch": 0.6179009541117674,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003395,
      "loss": 6.5837,
      "mean_token_accuracy": 0.11368328630924225,
      "num_tokens": 1397138.0,
      "step": 680
    },
    {
      "entropy": 7.001461124420166,
      "epoch": 0.6224443434802363,
      "grad_norm": 1.140625,
      "learning_rate": 0.000342,
      "loss": 6.6873,
      "mean_token_accuracy": 0.10977592095732688,
      "num_tokens": 1408171.0,
      "step": 685
    },
    {
      "entropy": 6.838484525680542,
      "epoch": 0.6269877328487051,
      "grad_norm": 1.15625,
      "learning_rate": 0.00034449999999999997,
      "loss": 6.6101,
      "mean_token_accuracy": 0.10415276065468788,
      "num_tokens": 1418259.0,
      "step": 690
    },
    {
      "entropy": 6.87859959602356,
      "epoch": 0.631531122217174,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000347,
      "loss": 6.6169,
      "mean_token_accuracy": 0.11740297675132752,
      "num_tokens": 1428080.0,
      "step": 695
    },
    {
      "entropy": 6.9925439834594725,
      "epoch": 0.6360745115856429,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003495,
      "loss": 6.6374,
      "mean_token_accuracy": 0.11313816457986832,
      "num_tokens": 1438502.0,
      "step": 700
    },
    {
      "entropy": 6.828886651992798,
      "epoch": 0.6406179009541118,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000352,
      "loss": 6.5904,
      "mean_token_accuracy": 0.11328546851873397,
      "num_tokens": 1448396.0,
      "step": 705
    },
    {
      "entropy": 6.961853837966919,
      "epoch": 0.6451612903225806,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003545,
      "loss": 6.6356,
      "mean_token_accuracy": 0.10769107192754745,
      "num_tokens": 1459728.0,
      "step": 710
    },
    {
      "entropy": 6.863912916183471,
      "epoch": 0.6497046796910495,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000357,
      "loss": 6.5312,
      "mean_token_accuracy": 0.1191827192902565,
      "num_tokens": 1469733.0,
      "step": 715
    },
    {
      "entropy": 6.790990257263184,
      "epoch": 0.6542480690595184,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003595,
      "loss": 6.5308,
      "mean_token_accuracy": 0.11443287208676338,
      "num_tokens": 1479004.0,
      "step": 720
    },
    {
      "entropy": 6.854830598831176,
      "epoch": 0.6587914584279873,
      "grad_norm": 1.1875,
      "learning_rate": 0.000362,
      "loss": 6.4956,
      "mean_token_accuracy": 0.11455681398510933,
      "num_tokens": 1489848.0,
      "step": 725
    },
    {
      "entropy": 6.813479566574097,
      "epoch": 0.6633348477964561,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003645,
      "loss": 6.4824,
      "mean_token_accuracy": 0.11726864874362945,
      "num_tokens": 1500877.0,
      "step": 730
    },
    {
      "entropy": 6.935261297225952,
      "epoch": 0.667878237164925,
      "grad_norm": 1.234375,
      "learning_rate": 0.000367,
      "loss": 6.6316,
      "mean_token_accuracy": 0.11274069175124168,
      "num_tokens": 1511840.0,
      "step": 735
    },
    {
      "entropy": 6.835966539382935,
      "epoch": 0.6724216265333939,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003695,
      "loss": 6.5425,
      "mean_token_accuracy": 0.11053763255476952,
      "num_tokens": 1521839.0,
      "step": 740
    },
    {
      "entropy": 6.827030801773072,
      "epoch": 0.6769650159018628,
      "grad_norm": 1.34375,
      "learning_rate": 0.000372,
      "loss": 6.4601,
      "mean_token_accuracy": 0.11425546035170556,
      "num_tokens": 1531285.0,
      "step": 745
    },
    {
      "entropy": 6.73953218460083,
      "epoch": 0.6815084052703316,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003745,
      "loss": 6.4915,
      "mean_token_accuracy": 0.11177220493555069,
      "num_tokens": 1542027.0,
      "step": 750
    },
    {
      "entropy": 6.826250886917114,
      "epoch": 0.6860517946388005,
      "grad_norm": 1.171875,
      "learning_rate": 0.000377,
      "loss": 6.5947,
      "mean_token_accuracy": 0.11794949546456338,
      "num_tokens": 1552527.0,
      "step": 755
    },
    {
      "entropy": 6.8967725276947025,
      "epoch": 0.6905951840072694,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0003795,
      "loss": 6.639,
      "mean_token_accuracy": 0.11244681403040886,
      "num_tokens": 1562605.0,
      "step": 760
    },
    {
      "entropy": 6.803361415863037,
      "epoch": 0.6951385733757383,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000382,
      "loss": 6.5235,
      "mean_token_accuracy": 0.11243330240249634,
      "num_tokens": 1572128.0,
      "step": 765
    },
    {
      "entropy": 6.795600032806396,
      "epoch": 0.6996819627442071,
      "grad_norm": 1.515625,
      "learning_rate": 0.0003845,
      "loss": 6.4591,
      "mean_token_accuracy": 0.11609070152044296,
      "num_tokens": 1582352.0,
      "step": 770
    },
    {
      "entropy": 6.812645721435547,
      "epoch": 0.704225352112676,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00038700000000000003,
      "loss": 6.5547,
      "mean_token_accuracy": 0.11612385958433151,
      "num_tokens": 1592887.0,
      "step": 775
    },
    {
      "entropy": 6.746767950057984,
      "epoch": 0.7087687414811449,
      "grad_norm": 0.953125,
      "learning_rate": 0.00038950000000000003,
      "loss": 6.4939,
      "mean_token_accuracy": 0.11659797877073289,
      "num_tokens": 1603564.0,
      "step": 780
    },
    {
      "entropy": 6.6716784000396725,
      "epoch": 0.7133121308496138,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039200000000000004,
      "loss": 6.3529,
      "mean_token_accuracy": 0.12544001638889313,
      "num_tokens": 1613270.0,
      "step": 785
    },
    {
      "entropy": 6.728840160369873,
      "epoch": 0.7178555202180827,
      "grad_norm": 1.546875,
      "learning_rate": 0.00039450000000000005,
      "loss": 6.4293,
      "mean_token_accuracy": 0.1263292007148266,
      "num_tokens": 1623009.0,
      "step": 790
    },
    {
      "entropy": 6.746360015869141,
      "epoch": 0.7223989095865516,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00039700000000000005,
      "loss": 6.4635,
      "mean_token_accuracy": 0.11441954895853997,
      "num_tokens": 1633209.0,
      "step": 795
    },
    {
      "entropy": 6.779658365249634,
      "epoch": 0.7269422989550205,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003995,
      "loss": 6.4894,
      "mean_token_accuracy": 0.1203268215060234,
      "num_tokens": 1643708.0,
      "step": 800
    },
    {
      "entropy": 6.690767478942871,
      "epoch": 0.7314856883234894,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000402,
      "loss": 6.5492,
      "mean_token_accuracy": 0.11524337530136108,
      "num_tokens": 1653973.0,
      "step": 805
    },
    {
      "entropy": 6.796014213562012,
      "epoch": 0.7360290776919582,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004045,
      "loss": 6.5308,
      "mean_token_accuracy": 0.11834607645869255,
      "num_tokens": 1664856.0,
      "step": 810
    },
    {
      "entropy": 6.693472051620484,
      "epoch": 0.7405724670604271,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040699999999999997,
      "loss": 6.4518,
      "mean_token_accuracy": 0.11877465546131134,
      "num_tokens": 1675935.0,
      "step": 815
    },
    {
      "entropy": 6.736241674423217,
      "epoch": 0.745115856428896,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004095,
      "loss": 6.488,
      "mean_token_accuracy": 0.11770976707339287,
      "num_tokens": 1686597.0,
      "step": 820
    },
    {
      "entropy": 6.7434159278869625,
      "epoch": 0.7496592457973649,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000412,
      "loss": 6.4418,
      "mean_token_accuracy": 0.11995128318667411,
      "num_tokens": 1697385.0,
      "step": 825
    },
    {
      "entropy": 6.87018084526062,
      "epoch": 0.7542026351658337,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004145,
      "loss": 6.4871,
      "mean_token_accuracy": 0.11747646629810334,
      "num_tokens": 1706680.0,
      "step": 830
    },
    {
      "entropy": 6.659475994110108,
      "epoch": 0.7587460245343026,
      "grad_norm": 1.078125,
      "learning_rate": 0.000417,
      "loss": 6.4874,
      "mean_token_accuracy": 0.1250874675810337,
      "num_tokens": 1716130.0,
      "step": 835
    },
    {
      "entropy": 6.664006137847901,
      "epoch": 0.7632894139027715,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004195,
      "loss": 6.4137,
      "mean_token_accuracy": 0.11410242542624474,
      "num_tokens": 1726835.0,
      "step": 840
    },
    {
      "entropy": 6.7605010032653805,
      "epoch": 0.7678328032712404,
      "grad_norm": 1.09375,
      "learning_rate": 0.000422,
      "loss": 6.4559,
      "mean_token_accuracy": 0.12729908376932145,
      "num_tokens": 1737708.0,
      "step": 845
    },
    {
      "entropy": 6.726518821716309,
      "epoch": 0.7723761926397092,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004245,
      "loss": 6.4323,
      "mean_token_accuracy": 0.12231917381286621,
      "num_tokens": 1747649.0,
      "step": 850
    },
    {
      "entropy": 6.653599405288697,
      "epoch": 0.7769195820081781,
      "grad_norm": 1.09375,
      "learning_rate": 0.000427,
      "loss": 6.4593,
      "mean_token_accuracy": 0.11891063153743744,
      "num_tokens": 1758327.0,
      "step": 855
    },
    {
      "entropy": 6.721684741973877,
      "epoch": 0.781462971376647,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004295,
      "loss": 6.4156,
      "mean_token_accuracy": 0.11827657297253609,
      "num_tokens": 1769767.0,
      "step": 860
    },
    {
      "entropy": 6.6785807609558105,
      "epoch": 0.7860063607451159,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000432,
      "loss": 6.5214,
      "mean_token_accuracy": 0.12091457024216652,
      "num_tokens": 1780139.0,
      "step": 865
    },
    {
      "entropy": 6.760772800445556,
      "epoch": 0.7905497501135847,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004345,
      "loss": 6.4172,
      "mean_token_accuracy": 0.12346908524632454,
      "num_tokens": 1790783.0,
      "step": 870
    },
    {
      "entropy": 6.653431367874146,
      "epoch": 0.7950931394820536,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000437,
      "loss": 6.4131,
      "mean_token_accuracy": 0.11938240155577659,
      "num_tokens": 1801349.0,
      "step": 875
    },
    {
      "entropy": 6.589143133163452,
      "epoch": 0.7996365288505225,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004395,
      "loss": 6.404,
      "mean_token_accuracy": 0.11662679985165596,
      "num_tokens": 1811990.0,
      "step": 880
    },
    {
      "entropy": 6.688130950927734,
      "epoch": 0.8041799182189914,
      "grad_norm": 1.171875,
      "learning_rate": 0.000442,
      "loss": 6.4161,
      "mean_token_accuracy": 0.12564157620072364,
      "num_tokens": 1821725.0,
      "step": 885
    },
    {
      "entropy": 6.628420114517212,
      "epoch": 0.8087233075874602,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004445,
      "loss": 6.3812,
      "mean_token_accuracy": 0.11830286234617234,
      "num_tokens": 1831794.0,
      "step": 890
    },
    {
      "entropy": 6.630347061157226,
      "epoch": 0.8132666969559291,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000447,
      "loss": 6.4406,
      "mean_token_accuracy": 0.11875352710485458,
      "num_tokens": 1842002.0,
      "step": 895
    },
    {
      "entropy": 6.713060092926026,
      "epoch": 0.817810086324398,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044950000000000003,
      "loss": 6.428,
      "mean_token_accuracy": 0.1267362855374813,
      "num_tokens": 1852282.0,
      "step": 900
    },
    {
      "entropy": 6.650756549835205,
      "epoch": 0.8223534756928669,
      "grad_norm": 1.28125,
      "learning_rate": 0.00045200000000000004,
      "loss": 6.4456,
      "mean_token_accuracy": 0.11507766097784042,
      "num_tokens": 1862830.0,
      "step": 905
    },
    {
      "entropy": 6.5850566864013675,
      "epoch": 0.8268968650613358,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045450000000000004,
      "loss": 6.3817,
      "mean_token_accuracy": 0.12454709187150001,
      "num_tokens": 1872746.0,
      "step": 910
    },
    {
      "entropy": 6.640098190307617,
      "epoch": 0.8314402544298046,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00045700000000000005,
      "loss": 6.4152,
      "mean_token_accuracy": 0.1162450760602951,
      "num_tokens": 1884353.0,
      "step": 915
    },
    {
      "entropy": 6.677035713195801,
      "epoch": 0.8359836437982735,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045950000000000006,
      "loss": 6.4494,
      "mean_token_accuracy": 0.1194159023463726,
      "num_tokens": 1895719.0,
      "step": 920
    },
    {
      "entropy": 6.643181371688843,
      "epoch": 0.8405270331667424,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000462,
      "loss": 6.3932,
      "mean_token_accuracy": 0.1279523029923439,
      "num_tokens": 1905905.0,
      "step": 925
    },
    {
      "entropy": 6.710006141662598,
      "epoch": 0.8450704225352113,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004645,
      "loss": 6.4895,
      "mean_token_accuracy": 0.1186818890273571,
      "num_tokens": 1916010.0,
      "step": 930
    },
    {
      "entropy": 6.604852104187012,
      "epoch": 0.8496138119036801,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000467,
      "loss": 6.3789,
      "mean_token_accuracy": 0.12054645344614982,
      "num_tokens": 1926429.0,
      "step": 935
    },
    {
      "entropy": 6.486386346817016,
      "epoch": 0.854157201272149,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004695,
      "loss": 6.3112,
      "mean_token_accuracy": 0.12320371642708779,
      "num_tokens": 1936322.0,
      "step": 940
    },
    {
      "entropy": 6.614945411682129,
      "epoch": 0.8587005906406179,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000472,
      "loss": 6.4767,
      "mean_token_accuracy": 0.12188357040286064,
      "num_tokens": 1947885.0,
      "step": 945
    },
    {
      "entropy": 6.704386138916016,
      "epoch": 0.8632439800090868,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004745,
      "loss": 6.4325,
      "mean_token_accuracy": 0.11737854704260826,
      "num_tokens": 1957876.0,
      "step": 950
    },
    {
      "entropy": 6.615909004211426,
      "epoch": 0.8677873693775556,
      "grad_norm": 1.078125,
      "learning_rate": 0.000477,
      "loss": 6.4402,
      "mean_token_accuracy": 0.12112006917595863,
      "num_tokens": 1969531.0,
      "step": 955
    },
    {
      "entropy": 6.572828912734986,
      "epoch": 0.8723307587460245,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004795,
      "loss": 6.2983,
      "mean_token_accuracy": 0.12579266652464866,
      "num_tokens": 1980787.0,
      "step": 960
    },
    {
      "entropy": 6.5395854949951175,
      "epoch": 0.8768741481144934,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000482,
      "loss": 6.3532,
      "mean_token_accuracy": 0.12586135566234588,
      "num_tokens": 1991645.0,
      "step": 965
    },
    {
      "entropy": 6.570622396469116,
      "epoch": 0.8814175374829623,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004845,
      "loss": 6.4259,
      "mean_token_accuracy": 0.12444038391113281,
      "num_tokens": 2002544.0,
      "step": 970
    },
    {
      "entropy": 6.674864959716797,
      "epoch": 0.8859609268514311,
      "grad_norm": 0.90625,
      "learning_rate": 0.000487,
      "loss": 6.4778,
      "mean_token_accuracy": 0.12012744024395942,
      "num_tokens": 2012813.0,
      "step": 975
    },
    {
      "entropy": 6.624161052703857,
      "epoch": 0.8905043162199,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004895,
      "loss": 6.3342,
      "mean_token_accuracy": 0.1272656500339508,
      "num_tokens": 2023756.0,
      "step": 980
    },
    {
      "entropy": 6.531716966629029,
      "epoch": 0.8950477055883689,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000492,
      "loss": 6.415,
      "mean_token_accuracy": 0.12654544711112975,
      "num_tokens": 2034245.0,
      "step": 985
    },
    {
      "entropy": 6.6605147361755375,
      "epoch": 0.8995910949568378,
      "grad_norm": 1.125,
      "learning_rate": 0.0004945,
      "loss": 6.4622,
      "mean_token_accuracy": 0.12089950144290924,
      "num_tokens": 2045635.0,
      "step": 990
    },
    {
      "entropy": 6.577081489562988,
      "epoch": 0.9041344843253066,
      "grad_norm": 0.8984375,
      "learning_rate": 0.000497,
      "loss": 6.4137,
      "mean_token_accuracy": 0.12277993112802506,
      "num_tokens": 2056501.0,
      "step": 995
    },
    {
      "entropy": 6.601724433898926,
      "epoch": 0.9086778736937755,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004995,
      "loss": 6.4161,
      "mean_token_accuracy": 0.1239933967590332,
      "num_tokens": 2066540.0,
      "step": 1000
    },
    {
      "epoch": 0.9086778736937755,
      "eval_entropy": 6.457386301839074,
      "eval_loss": 6.457565784454346,
      "eval_mean_token_accuracy": 0.12266246589810349,
      "eval_num_tokens": 2066540.0,
      "eval_runtime": 2.0261,
      "eval_samples_per_second": 1695.353,
      "eval_steps_per_second": 212.228,
      "step": 1000
    },
    {
      "entropy": 6.642101049423218,
      "epoch": 0.9132212630622444,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999998223471442,
      "loss": 6.3046,
      "mean_token_accuracy": 0.12406818643212318,
      "num_tokens": 2077008.0,
      "step": 1005
    },
    {
      "entropy": 6.447614336013794,
      "epoch": 0.9177646524307133,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999991006328981,
      "loss": 6.2211,
      "mean_token_accuracy": 0.12996889874339104,
      "num_tokens": 2087314.0,
      "step": 1010
    },
    {
      "entropy": 6.564070749282837,
      "epoch": 0.9223080417991822,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999978237557377,
      "loss": 6.3952,
      "mean_token_accuracy": 0.12220894247293472,
      "num_tokens": 2097928.0,
      "step": 1015
    },
    {
      "entropy": 6.498920154571533,
      "epoch": 0.9268514311676511,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999959917188137,
      "loss": 6.3161,
      "mean_token_accuracy": 0.12189262807369232,
      "num_tokens": 2108662.0,
      "step": 1020
    },
    {
      "entropy": 6.502299547195435,
      "epoch": 0.93139482053612,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004999936045266462,
      "loss": 6.3559,
      "mean_token_accuracy": 0.12635889351367952,
      "num_tokens": 2118546.0,
      "step": 1025
    },
    {
      "entropy": 6.61690583229065,
      "epoch": 0.9359382099045889,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999906621851255,
      "loss": 6.3872,
      "mean_token_accuracy": 0.1270056627690792,
      "num_tokens": 2128012.0,
      "step": 1030
    },
    {
      "entropy": 6.446464252471924,
      "epoch": 0.9404815992730577,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999871647015115,
      "loss": 6.2277,
      "mean_token_accuracy": 0.12683860957622528,
      "num_tokens": 2138241.0,
      "step": 1035
    },
    {
      "entropy": 6.643786096572876,
      "epoch": 0.9450249886415266,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004999831120844339,
      "loss": 6.3889,
      "mean_token_accuracy": 0.12322892174124718,
      "num_tokens": 2150780.0,
      "step": 1040
    },
    {
      "entropy": 6.422167158126831,
      "epoch": 0.9495683780099955,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999785043438922,
      "loss": 6.2808,
      "mean_token_accuracy": 0.12567662224173545,
      "num_tokens": 2161918.0,
      "step": 1045
    },
    {
      "entropy": 6.502702140808106,
      "epoch": 0.9541117673784644,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999733414912555,
      "loss": 6.3161,
      "mean_token_accuracy": 0.12407866790890694,
      "num_tokens": 2172742.0,
      "step": 1050
    },
    {
      "entropy": 6.470433187484741,
      "epoch": 0.9586551567469332,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999676235392626,
      "loss": 6.2963,
      "mean_token_accuracy": 0.13002401813864709,
      "num_tokens": 2182122.0,
      "step": 1055
    },
    {
      "entropy": 6.475074577331543,
      "epoch": 0.9631985461154021,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004999613505020219,
      "loss": 6.3398,
      "mean_token_accuracy": 0.12201474159955979,
      "num_tokens": 2193703.0,
      "step": 1060
    },
    {
      "entropy": 6.485757064819336,
      "epoch": 0.967741935483871,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999545223950116,
      "loss": 6.2524,
      "mean_token_accuracy": 0.1314729295670986,
      "num_tokens": 2204041.0,
      "step": 1065
    },
    {
      "entropy": 6.5328117370605465,
      "epoch": 0.9722853248523399,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999471392350795,
      "loss": 6.2924,
      "mean_token_accuracy": 0.12847484424710273,
      "num_tokens": 2214276.0,
      "step": 1070
    },
    {
      "entropy": 6.443771505355835,
      "epoch": 0.9768287142208087,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999392010404426,
      "loss": 6.2555,
      "mean_token_accuracy": 0.12922132089734079,
      "num_tokens": 2225390.0,
      "step": 1075
    },
    {
      "entropy": 6.469732618331909,
      "epoch": 0.9813721035892776,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999307078306876,
      "loss": 6.236,
      "mean_token_accuracy": 0.12894286066293717,
      "num_tokens": 2235708.0,
      "step": 1080
    },
    {
      "entropy": 6.518327808380127,
      "epoch": 0.9859154929577465,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999216596267708,
      "loss": 6.25,
      "mean_token_accuracy": 0.13040516525506973,
      "num_tokens": 2246284.0,
      "step": 1085
    },
    {
      "entropy": 6.445136451721192,
      "epoch": 0.9904588823262154,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004999120564510177,
      "loss": 6.2634,
      "mean_token_accuracy": 0.1280519634485245,
      "num_tokens": 2257219.0,
      "step": 1090
    },
    {
      "entropy": 6.365058374404907,
      "epoch": 0.9950022716946842,
      "grad_norm": 0.953125,
      "learning_rate": 0.000499901898327123,
      "loss": 6.2679,
      "mean_token_accuracy": 0.12958364114165305,
      "num_tokens": 2268019.0,
      "step": 1095
    },
    {
      "entropy": 6.465979433059692,
      "epoch": 0.9995456610631531,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004998911852801512,
      "loss": 6.2853,
      "mean_token_accuracy": 0.12592302560806273,
      "num_tokens": 2278382.0,
      "step": 1100
    },
    {
      "entropy": 6.542138364579943,
      "epoch": 1.003634711494775,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004998799173365355,
      "loss": 6.1386,
      "mean_token_accuracy": 0.12785813957452774,
      "num_tokens": 2288515.0,
      "step": 1105
    },
    {
      "entropy": 6.407947540283203,
      "epoch": 1.008178100863244,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998680945240783,
      "loss": 5.9955,
      "mean_token_accuracy": 0.13868918940424918,
      "num_tokens": 2298396.0,
      "step": 1110
    },
    {
      "entropy": 6.337148284912109,
      "epoch": 1.0127214902317128,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998557168719513,
      "loss": 6.1121,
      "mean_token_accuracy": 0.13123534470796586,
      "num_tokens": 2308707.0,
      "step": 1115
    },
    {
      "entropy": 6.413614797592163,
      "epoch": 1.0172648796001817,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004998427844106954,
      "loss": 6.0338,
      "mean_token_accuracy": 0.13907074108719825,
      "num_tokens": 2318981.0,
      "step": 1120
    },
    {
      "entropy": 6.406648731231689,
      "epoch": 1.0218082689686505,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004998292971722197,
      "loss": 6.132,
      "mean_token_accuracy": 0.13878810107707978,
      "num_tokens": 2329647.0,
      "step": 1125
    },
    {
      "entropy": 6.378375291824341,
      "epoch": 1.0263516583371195,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000499815255189803,
      "loss": 6.0628,
      "mean_token_accuracy": 0.13353429287672042,
      "num_tokens": 2339449.0,
      "step": 1130
    },
    {
      "entropy": 6.402873182296753,
      "epoch": 1.0308950477055883,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004998006584980922,
      "loss": 6.1569,
      "mean_token_accuracy": 0.13958344906568526,
      "num_tokens": 2349904.0,
      "step": 1135
    },
    {
      "entropy": 6.362394571304321,
      "epoch": 1.0354384370740572,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004997855071331034,
      "loss": 6.0681,
      "mean_token_accuracy": 0.1426156133413315,
      "num_tokens": 2359833.0,
      "step": 1140
    },
    {
      "entropy": 6.438542270660401,
      "epoch": 1.039981826442526,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004997698011322211,
      "loss": 6.0931,
      "mean_token_accuracy": 0.13478946760296823,
      "num_tokens": 2369857.0,
      "step": 1145
    },
    {
      "entropy": 6.419106435775757,
      "epoch": 1.044525215810995,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000499753540534198,
      "loss": 6.1102,
      "mean_token_accuracy": 0.13969969153404235,
      "num_tokens": 2380372.0,
      "step": 1150
    },
    {
      "entropy": 6.359906387329102,
      "epoch": 1.049068605179464,
      "grad_norm": 1.1875,
      "learning_rate": 0.000499736725379156,
      "loss": 6.0892,
      "mean_token_accuracy": 0.13606216683983802,
      "num_tokens": 2391239.0,
      "step": 1155
    },
    {
      "entropy": 6.311654329299927,
      "epoch": 1.0536119945479328,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004997193557085844,
      "loss": 6.0798,
      "mean_token_accuracy": 0.14339477866888045,
      "num_tokens": 2401178.0,
      "step": 1160
    },
    {
      "entropy": 6.429500484466553,
      "epoch": 1.0581553839164017,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004997014315653413,
      "loss": 6.1488,
      "mean_token_accuracy": 0.13889346718788148,
      "num_tokens": 2412496.0,
      "step": 1165
    },
    {
      "entropy": 6.420183420181274,
      "epoch": 1.0626987732848705,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004996829529936527,
      "loss": 6.1371,
      "mean_token_accuracy": 0.13705583438277244,
      "num_tokens": 2423281.0,
      "step": 1170
    },
    {
      "entropy": 6.307903480529785,
      "epoch": 1.0672421626533395,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004996639200391126,
      "loss": 6.0821,
      "mean_token_accuracy": 0.13973685950040818,
      "num_tokens": 2434078.0,
      "step": 1175
    },
    {
      "entropy": 6.311843490600586,
      "epoch": 1.0717855520218083,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004996443327486831,
      "loss": 6.0074,
      "mean_token_accuracy": 0.14173415526747704,
      "num_tokens": 2443808.0,
      "step": 1180
    },
    {
      "entropy": 6.209487628936768,
      "epoch": 1.0763289413902772,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004996241911706937,
      "loss": 6.0393,
      "mean_token_accuracy": 0.14506323635578156,
      "num_tokens": 2453397.0,
      "step": 1185
    },
    {
      "entropy": 6.438489770889282,
      "epoch": 1.080872330758746,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004996034953548418,
      "loss": 6.1409,
      "mean_token_accuracy": 0.135777448117733,
      "num_tokens": 2464295.0,
      "step": 1190
    },
    {
      "entropy": 6.37608585357666,
      "epoch": 1.085415720127215,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004995822453521924,
      "loss": 6.1874,
      "mean_token_accuracy": 0.13546083718538285,
      "num_tokens": 2475311.0,
      "step": 1195
    },
    {
      "entropy": 6.473006963729858,
      "epoch": 1.0899591094956838,
      "grad_norm": 1.25,
      "learning_rate": 0.0004995604412151776,
      "loss": 6.1026,
      "mean_token_accuracy": 0.13419248089194297,
      "num_tokens": 2485968.0,
      "step": 1200
    },
    {
      "entropy": 6.257586097717285,
      "epoch": 1.0945024988641527,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000499538082997597,
      "loss": 6.1261,
      "mean_token_accuracy": 0.13627575635910033,
      "num_tokens": 2496183.0,
      "step": 1205
    },
    {
      "entropy": 6.336587190628052,
      "epoch": 1.0990458882326215,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004995151707546172,
      "loss": 6.0836,
      "mean_token_accuracy": 0.13671565651893616,
      "num_tokens": 2506242.0,
      "step": 1210
    },
    {
      "entropy": 6.3969768524169925,
      "epoch": 1.1035892776010905,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004994917045427721,
      "loss": 6.0852,
      "mean_token_accuracy": 0.1399049386382103,
      "num_tokens": 2516443.0,
      "step": 1215
    },
    {
      "entropy": 6.255747842788696,
      "epoch": 1.1081326669695593,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004994676844199621,
      "loss": 6.0569,
      "mean_token_accuracy": 0.1428102508187294,
      "num_tokens": 2525795.0,
      "step": 1220
    },
    {
      "entropy": 6.344145202636719,
      "epoch": 1.1126760563380282,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004994431104454543,
      "loss": 6.0782,
      "mean_token_accuracy": 0.13670306727290155,
      "num_tokens": 2535046.0,
      "step": 1225
    },
    {
      "entropy": 6.36448302268982,
      "epoch": 1.117219445706497,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004994179826798828,
      "loss": 6.0756,
      "mean_token_accuracy": 0.1401789128780365,
      "num_tokens": 2546489.0,
      "step": 1230
    },
    {
      "entropy": 6.078287601470947,
      "epoch": 1.121762835074966,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004993923011852477,
      "loss": 6.0231,
      "mean_token_accuracy": 0.1417386919260025,
      "num_tokens": 2556335.0,
      "step": 1235
    },
    {
      "entropy": 6.46276741027832,
      "epoch": 1.1263062244434348,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004993660660249158,
      "loss": 6.0399,
      "mean_token_accuracy": 0.1342715211212635,
      "num_tokens": 2567647.0,
      "step": 1240
    },
    {
      "entropy": 6.412200593948365,
      "epoch": 1.1308496138119037,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004993392772636194,
      "loss": 6.2511,
      "mean_token_accuracy": 0.12803458347916602,
      "num_tokens": 2578618.0,
      "step": 1245
    },
    {
      "entropy": 6.206548547744751,
      "epoch": 1.1353930031803725,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004993119349674573,
      "loss": 6.0558,
      "mean_token_accuracy": 0.14603335335850715,
      "num_tokens": 2587797.0,
      "step": 1250
    },
    {
      "entropy": 6.319081544876099,
      "epoch": 1.1399363925488415,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004992840392038938,
      "loss": 6.0984,
      "mean_token_accuracy": 0.133018708974123,
      "num_tokens": 2598403.0,
      "step": 1255
    },
    {
      "entropy": 6.366528797149658,
      "epoch": 1.1444797819173103,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000499255590041759,
      "loss": 6.0943,
      "mean_token_accuracy": 0.13550746738910674,
      "num_tokens": 2608296.0,
      "step": 1260
    },
    {
      "entropy": 6.3718188285827635,
      "epoch": 1.1490231712857792,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004992265875512485,
      "loss": 6.1071,
      "mean_token_accuracy": 0.1415480449795723,
      "num_tokens": 2617906.0,
      "step": 1265
    },
    {
      "entropy": 6.257988023757934,
      "epoch": 1.153566560654248,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004991970318039228,
      "loss": 6.0511,
      "mean_token_accuracy": 0.13821252733469008,
      "num_tokens": 2627699.0,
      "step": 1270
    },
    {
      "entropy": 6.330170059204102,
      "epoch": 1.158109950022717,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004991669228727078,
      "loss": 6.027,
      "mean_token_accuracy": 0.1415513798594475,
      "num_tokens": 2638384.0,
      "step": 1275
    },
    {
      "entropy": 6.220354127883911,
      "epoch": 1.1626533393911858,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004991362608318947,
      "loss": 5.9593,
      "mean_token_accuracy": 0.14534688517451286,
      "num_tokens": 2648279.0,
      "step": 1280
    },
    {
      "entropy": 6.248266410827637,
      "epoch": 1.1671967287596547,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004991050457571384,
      "loss": 6.0408,
      "mean_token_accuracy": 0.12930047884583473,
      "num_tokens": 2657000.0,
      "step": 1285
    },
    {
      "entropy": 6.254126119613647,
      "epoch": 1.1717401181281235,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004990732777254594,
      "loss": 6.0696,
      "mean_token_accuracy": 0.14027479737997056,
      "num_tokens": 2667760.0,
      "step": 1290
    },
    {
      "entropy": 6.35225100517273,
      "epoch": 1.1762835074965925,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004990409568152422,
      "loss": 6.0009,
      "mean_token_accuracy": 0.14144083932042123,
      "num_tokens": 2678450.0,
      "step": 1295
    },
    {
      "entropy": 6.288917922973633,
      "epoch": 1.1808268968650613,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004990080831062352,
      "loss": 6.1168,
      "mean_token_accuracy": 0.13410752713680268,
      "num_tokens": 2689695.0,
      "step": 1300
    },
    {
      "entropy": 6.342508125305176,
      "epoch": 1.1853702862335302,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004989746566795512,
      "loss": 6.1282,
      "mean_token_accuracy": 0.14176513105630875,
      "num_tokens": 2700414.0,
      "step": 1305
    },
    {
      "entropy": 6.242630481719971,
      "epoch": 1.189913675601999,
      "grad_norm": 1.125,
      "learning_rate": 0.0004989406776176665,
      "loss": 5.9502,
      "mean_token_accuracy": 0.15300626754760743,
      "num_tokens": 2709985.0,
      "step": 1310
    },
    {
      "entropy": 6.248297691345215,
      "epoch": 1.194457064970468,
      "grad_norm": 1.140625,
      "learning_rate": 0.000498906146004421,
      "loss": 6.0924,
      "mean_token_accuracy": 0.14182935208082198,
      "num_tokens": 2720990.0,
      "step": 1315
    },
    {
      "entropy": 6.269750690460205,
      "epoch": 1.1990004543389368,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004988710619250181,
      "loss": 5.9948,
      "mean_token_accuracy": 0.14390970543026924,
      "num_tokens": 2731285.0,
      "step": 1320
    },
    {
      "entropy": 6.23701663017273,
      "epoch": 1.2035438437074057,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004988354254660244,
      "loss": 6.0861,
      "mean_token_accuracy": 0.14006828367710114,
      "num_tokens": 2741024.0,
      "step": 1325
    },
    {
      "entropy": 6.250706720352173,
      "epoch": 1.2080872330758745,
      "grad_norm": 1.0,
      "learning_rate": 0.0004987992367153691,
      "loss": 6.0089,
      "mean_token_accuracy": 0.14118139818310738,
      "num_tokens": 2751447.0,
      "step": 1330
    },
    {
      "entropy": 6.288669204711914,
      "epoch": 1.2126306224443435,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004987624957623443,
      "loss": 6.0333,
      "mean_token_accuracy": 0.14222527295351028,
      "num_tokens": 2761283.0,
      "step": 1335
    },
    {
      "entropy": 6.160416793823242,
      "epoch": 1.2171740118128123,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004987252026976049,
      "loss": 5.9472,
      "mean_token_accuracy": 0.14869927018880844,
      "num_tokens": 2771354.0,
      "step": 1340
    },
    {
      "entropy": 6.327591037750244,
      "epoch": 1.2217174011812812,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004986873576131676,
      "loss": 6.1407,
      "mean_token_accuracy": 0.13928610384464263,
      "num_tokens": 2781063.0,
      "step": 1345
    },
    {
      "entropy": 6.322598695755005,
      "epoch": 1.22626079054975,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004986489606024116,
      "loss": 6.0377,
      "mean_token_accuracy": 0.14065276831388474,
      "num_tokens": 2791103.0,
      "step": 1350
    },
    {
      "entropy": 6.268867301940918,
      "epoch": 1.230804179918219,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004986100117600776,
      "loss": 6.1553,
      "mean_token_accuracy": 0.1384136810898781,
      "num_tokens": 2802382.0,
      "step": 1355
    },
    {
      "entropy": 6.20656189918518,
      "epoch": 1.235347569286688,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000498570511182268,
      "loss": 6.0044,
      "mean_token_accuracy": 0.1394697204232216,
      "num_tokens": 2813740.0,
      "step": 1360
    },
    {
      "entropy": 6.308270502090454,
      "epoch": 1.2398909586551567,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004985304589664465,
      "loss": 6.0168,
      "mean_token_accuracy": 0.13902125507593155,
      "num_tokens": 2824121.0,
      "step": 1365
    },
    {
      "entropy": 6.242033958435059,
      "epoch": 1.2444343480236255,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004984898552114381,
      "loss": 6.0452,
      "mean_token_accuracy": 0.14119891151785852,
      "num_tokens": 2833356.0,
      "step": 1370
    },
    {
      "entropy": 6.288197469711304,
      "epoch": 1.2489777373920945,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004984487000174283,
      "loss": 6.0387,
      "mean_token_accuracy": 0.13836880177259445,
      "num_tokens": 2843839.0,
      "step": 1375
    },
    {
      "entropy": 6.218136358261108,
      "epoch": 1.2535211267605635,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004984069934859637,
      "loss": 6.0001,
      "mean_token_accuracy": 0.13987476527690887,
      "num_tokens": 2853523.0,
      "step": 1380
    },
    {
      "entropy": 6.194594192504883,
      "epoch": 1.2580645161290323,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004983647357199509,
      "loss": 6.0492,
      "mean_token_accuracy": 0.13922794684767723,
      "num_tokens": 2863226.0,
      "step": 1385
    },
    {
      "entropy": 6.40929217338562,
      "epoch": 1.262607905497501,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004983219268236567,
      "loss": 6.0971,
      "mean_token_accuracy": 0.14521313607692718,
      "num_tokens": 2873822.0,
      "step": 1390
    },
    {
      "entropy": 6.130873823165894,
      "epoch": 1.26715129486597,
      "grad_norm": 1.0,
      "learning_rate": 0.0004982785669027079,
      "loss": 5.9524,
      "mean_token_accuracy": 0.14649356976151467,
      "num_tokens": 2883645.0,
      "step": 1395
    },
    {
      "entropy": 6.290006494522094,
      "epoch": 1.271694684234439,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004982346560640907,
      "loss": 6.071,
      "mean_token_accuracy": 0.13487468957901,
      "num_tokens": 2894397.0,
      "step": 1400
    },
    {
      "entropy": 6.285066986083985,
      "epoch": 1.2762380736029078,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004981901944161507,
      "loss": 6.0467,
      "mean_token_accuracy": 0.14313390702009202,
      "num_tokens": 2904947.0,
      "step": 1405
    },
    {
      "entropy": 6.171815013885498,
      "epoch": 1.2807814629713765,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004981451820685928,
      "loss": 5.9978,
      "mean_token_accuracy": 0.14585176184773446,
      "num_tokens": 2916183.0,
      "step": 1410
    },
    {
      "entropy": 6.206707763671875,
      "epoch": 1.2853248523398455,
      "grad_norm": 1.125,
      "learning_rate": 0.0004980996191324802,
      "loss": 5.9117,
      "mean_token_accuracy": 0.14459551572799684,
      "num_tokens": 2925429.0,
      "step": 1415
    },
    {
      "entropy": 6.226998424530029,
      "epoch": 1.2898682417083145,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004980535057202351,
      "loss": 5.9839,
      "mean_token_accuracy": 0.14731092602014542,
      "num_tokens": 2935142.0,
      "step": 1420
    },
    {
      "entropy": 6.235570096969605,
      "epoch": 1.2944116310767833,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004980068419456378,
      "loss": 6.051,
      "mean_token_accuracy": 0.1416834071278572,
      "num_tokens": 2945151.0,
      "step": 1425
    },
    {
      "entropy": 6.298262405395508,
      "epoch": 1.2989550204452522,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004979596279238265,
      "loss": 6.1376,
      "mean_token_accuracy": 0.13880389630794526,
      "num_tokens": 2956015.0,
      "step": 1430
    },
    {
      "entropy": 6.257140445709228,
      "epoch": 1.303498409813721,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004979118637712969,
      "loss": 5.9659,
      "mean_token_accuracy": 0.14583468139171601,
      "num_tokens": 2965879.0,
      "step": 1435
    },
    {
      "entropy": 6.238783836364746,
      "epoch": 1.30804179918219,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004978635496059025,
      "loss": 6.0532,
      "mean_token_accuracy": 0.13990017473697663,
      "num_tokens": 2977021.0,
      "step": 1440
    },
    {
      "entropy": 6.180285978317261,
      "epoch": 1.3125851885506588,
      "grad_norm": 1.125,
      "learning_rate": 0.0004978146855468537,
      "loss": 5.9404,
      "mean_token_accuracy": 0.14709279239177703,
      "num_tokens": 2987169.0,
      "step": 1445
    },
    {
      "entropy": 6.32270450592041,
      "epoch": 1.3171285779191277,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004977652717147177,
      "loss": 6.0525,
      "mean_token_accuracy": 0.1423275850713253,
      "num_tokens": 2997666.0,
      "step": 1450
    },
    {
      "entropy": 6.153949642181397,
      "epoch": 1.3216719672875965,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004977153082314182,
      "loss": 5.9168,
      "mean_token_accuracy": 0.14165920540690422,
      "num_tokens": 3007021.0,
      "step": 1455
    },
    {
      "entropy": 6.16786413192749,
      "epoch": 1.3262153566560655,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000497664795220235,
      "loss": 5.9954,
      "mean_token_accuracy": 0.14898994714021682,
      "num_tokens": 3016557.0,
      "step": 1460
    },
    {
      "entropy": 6.173955011367798,
      "epoch": 1.3307587460245343,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004976137328058042,
      "loss": 5.9605,
      "mean_token_accuracy": 0.14649975001811982,
      "num_tokens": 3026146.0,
      "step": 1465
    },
    {
      "entropy": 6.278796911239624,
      "epoch": 1.3353021353930032,
      "grad_norm": 1.015625,
      "learning_rate": 0.000497562121114117,
      "loss": 6.0085,
      "mean_token_accuracy": 0.14478534683585167,
      "num_tokens": 3037234.0,
      "step": 1470
    },
    {
      "entropy": 6.171536302566528,
      "epoch": 1.339845524761472,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004975099602725202,
      "loss": 5.948,
      "mean_token_accuracy": 0.14746320694684983,
      "num_tokens": 3047282.0,
      "step": 1475
    },
    {
      "entropy": 6.13257532119751,
      "epoch": 1.344388914129941,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004974572504097154,
      "loss": 5.9735,
      "mean_token_accuracy": 0.14172942489385604,
      "num_tokens": 3058648.0,
      "step": 1480
    },
    {
      "entropy": 6.366324806213379,
      "epoch": 1.3489323034984098,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004974039916557591,
      "loss": 6.1551,
      "mean_token_accuracy": 0.14091512486338614,
      "num_tokens": 3068996.0,
      "step": 1485
    },
    {
      "entropy": 6.178349304199219,
      "epoch": 1.3534756928668787,
      "grad_norm": 1.109375,
      "learning_rate": 0.000497350184142062,
      "loss": 5.9532,
      "mean_token_accuracy": 0.1532169207930565,
      "num_tokens": 3078991.0,
      "step": 1490
    },
    {
      "entropy": 6.131071615219116,
      "epoch": 1.3580190822353475,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004972958280013886,
      "loss": 5.9267,
      "mean_token_accuracy": 0.14385127127170563,
      "num_tokens": 3090200.0,
      "step": 1495
    },
    {
      "entropy": 6.145272159576416,
      "epoch": 1.3625624716038165,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004972409233678575,
      "loss": 5.9836,
      "mean_token_accuracy": 0.14638550356030464,
      "num_tokens": 3101971.0,
      "step": 1500
    },
    {
      "epoch": 1.3625624716038165,
      "eval_entropy": 6.003304544714994,
      "eval_loss": 6.187674522399902,
      "eval_mean_token_accuracy": 0.13976414828279685,
      "eval_num_tokens": 3101971.0,
      "eval_runtime": 2.0245,
      "eval_samples_per_second": 1696.731,
      "eval_steps_per_second": 212.4,
      "step": 1500
    },
    {
      "entropy": 6.193512916564941,
      "epoch": 1.3671058609722853,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004971854703769401,
      "loss": 6.05,
      "mean_token_accuracy": 0.14252812564373016,
      "num_tokens": 3113381.0,
      "step": 1505
    },
    {
      "entropy": 6.290583896636963,
      "epoch": 1.3716492503407542,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004971294691654616,
      "loss": 6.0279,
      "mean_token_accuracy": 0.1442758098244667,
      "num_tokens": 3125099.0,
      "step": 1510
    },
    {
      "entropy": 6.086841201782226,
      "epoch": 1.376192639709223,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000497072919871599,
      "loss": 5.9198,
      "mean_token_accuracy": 0.14686398953199387,
      "num_tokens": 3135007.0,
      "step": 1515
    },
    {
      "entropy": 6.208672046661377,
      "epoch": 1.380736029077692,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004970158226348824,
      "loss": 5.9621,
      "mean_token_accuracy": 0.14713747948408126,
      "num_tokens": 3145031.0,
      "step": 1520
    },
    {
      "entropy": 6.197163724899292,
      "epoch": 1.3852794184461608,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004969581775961932,
      "loss": 6.0023,
      "mean_token_accuracy": 0.14530939906835555,
      "num_tokens": 3156044.0,
      "step": 1525
    },
    {
      "entropy": 6.12416296005249,
      "epoch": 1.3898228078146297,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004968999848977651,
      "loss": 5.9125,
      "mean_token_accuracy": 0.14605395942926408,
      "num_tokens": 3165124.0,
      "step": 1530
    },
    {
      "entropy": 6.086002159118652,
      "epoch": 1.3943661971830985,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004968412446831827,
      "loss": 5.8588,
      "mean_token_accuracy": 0.15170793682336808,
      "num_tokens": 3175246.0,
      "step": 1535
    },
    {
      "entropy": 6.2570037841796875,
      "epoch": 1.3989095865515675,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004967819570973816,
      "loss": 5.9464,
      "mean_token_accuracy": 0.15199698209762574,
      "num_tokens": 3184817.0,
      "step": 1540
    },
    {
      "entropy": 6.093594932556153,
      "epoch": 1.4034529759200365,
      "grad_norm": 1.0,
      "learning_rate": 0.000496722122286648,
      "loss": 5.9871,
      "mean_token_accuracy": 0.13670103028416633,
      "num_tokens": 3195142.0,
      "step": 1545
    },
    {
      "entropy": 6.152914714813233,
      "epoch": 1.4079963652885052,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004966617403986185,
      "loss": 5.9248,
      "mean_token_accuracy": 0.14729512929916383,
      "num_tokens": 3205724.0,
      "step": 1550
    },
    {
      "entropy": 6.198964929580688,
      "epoch": 1.412539754656974,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004966008115822793,
      "loss": 6.0814,
      "mean_token_accuracy": 0.14055647775530816,
      "num_tokens": 3216919.0,
      "step": 1555
    },
    {
      "entropy": 6.316028022766114,
      "epoch": 1.417083144025443,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004965393359879662,
      "loss": 6.0752,
      "mean_token_accuracy": 0.14144429787993432,
      "num_tokens": 3228929.0,
      "step": 1560
    },
    {
      "entropy": 6.152643489837646,
      "epoch": 1.421626533393912,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004964773137673641,
      "loss": 6.0427,
      "mean_token_accuracy": 0.1418978050351143,
      "num_tokens": 3239541.0,
      "step": 1565
    },
    {
      "entropy": 6.203517246246338,
      "epoch": 1.4261699227623807,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004964147450735067,
      "loss": 5.9156,
      "mean_token_accuracy": 0.14633660465478898,
      "num_tokens": 3249527.0,
      "step": 1570
    },
    {
      "entropy": 6.083546543121338,
      "epoch": 1.4307133121308495,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000496351630060776,
      "loss": 5.8145,
      "mean_token_accuracy": 0.14704043418169022,
      "num_tokens": 3260497.0,
      "step": 1575
    },
    {
      "entropy": 6.252065896987915,
      "epoch": 1.4352567014993185,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004962879688849021,
      "loss": 6.0109,
      "mean_token_accuracy": 0.14513912349939345,
      "num_tokens": 3270865.0,
      "step": 1580
    },
    {
      "entropy": 6.0464317321777346,
      "epoch": 1.4398000908677875,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004962237617029625,
      "loss": 5.9389,
      "mean_token_accuracy": 0.14239763393998145,
      "num_tokens": 3281531.0,
      "step": 1585
    },
    {
      "entropy": 6.243140935897827,
      "epoch": 1.4443434802362562,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004961590086733821,
      "loss": 5.9092,
      "mean_token_accuracy": 0.14480504766106606,
      "num_tokens": 3291970.0,
      "step": 1590
    },
    {
      "entropy": 6.072083616256714,
      "epoch": 1.448886869604725,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004960937099559326,
      "loss": 5.9548,
      "mean_token_accuracy": 0.14921991974115373,
      "num_tokens": 3301638.0,
      "step": 1595
    },
    {
      "entropy": 6.242140626907348,
      "epoch": 1.453430258973194,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004960278657117321,
      "loss": 6.0293,
      "mean_token_accuracy": 0.14663881212472915,
      "num_tokens": 3311780.0,
      "step": 1600
    },
    {
      "entropy": 6.103953409194946,
      "epoch": 1.457973648341663,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004959614761032446,
      "loss": 5.8761,
      "mean_token_accuracy": 0.1474029079079628,
      "num_tokens": 3322096.0,
      "step": 1605
    },
    {
      "entropy": 6.278363752365112,
      "epoch": 1.4625170377101318,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00049589454129428,
      "loss": 6.0609,
      "mean_token_accuracy": 0.1401104636490345,
      "num_tokens": 3332324.0,
      "step": 1610
    },
    {
      "entropy": 6.039387226104736,
      "epoch": 1.4670604270786005,
      "grad_norm": 1.125,
      "learning_rate": 0.0004958270614499932,
      "loss": 5.8519,
      "mean_token_accuracy": 0.15410613864660264,
      "num_tokens": 3342037.0,
      "step": 1615
    },
    {
      "entropy": 6.2045708179473875,
      "epoch": 1.4716038164470695,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004957590367368841,
      "loss": 5.9407,
      "mean_token_accuracy": 0.14948561191558837,
      "num_tokens": 3353047.0,
      "step": 1620
    },
    {
      "entropy": 6.0782692432403564,
      "epoch": 1.4761472058155385,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004956904673227968,
      "loss": 5.849,
      "mean_token_accuracy": 0.14826880693435668,
      "num_tokens": 3363841.0,
      "step": 1625
    },
    {
      "entropy": 6.134917449951172,
      "epoch": 1.4806905951840073,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004956213533769195,
      "loss": 5.9461,
      "mean_token_accuracy": 0.1493829995393753,
      "num_tokens": 3374086.0,
      "step": 1630
    },
    {
      "entropy": 6.147575044631958,
      "epoch": 1.485233984552476,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004955516950697843,
      "loss": 5.9277,
      "mean_token_accuracy": 0.1457385838031769,
      "num_tokens": 3384303.0,
      "step": 1635
    },
    {
      "entropy": 6.106037664413452,
      "epoch": 1.489777373920945,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004954814925732657,
      "loss": 5.9065,
      "mean_token_accuracy": 0.14968472644686698,
      "num_tokens": 3394858.0,
      "step": 1640
    },
    {
      "entropy": 6.126568031311035,
      "epoch": 1.494320763289414,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004954107460605816,
      "loss": 5.921,
      "mean_token_accuracy": 0.1508701652288437,
      "num_tokens": 3405252.0,
      "step": 1645
    },
    {
      "entropy": 6.235639429092407,
      "epoch": 1.4988641526578828,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004953394557062922,
      "loss": 5.9737,
      "mean_token_accuracy": 0.1500774383544922,
      "num_tokens": 3416007.0,
      "step": 1650
    },
    {
      "entropy": 6.198228073120117,
      "epoch": 1.5034075420263515,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004952676216862989,
      "loss": 6.0023,
      "mean_token_accuracy": 0.145437902957201,
      "num_tokens": 3426932.0,
      "step": 1655
    },
    {
      "entropy": 6.046504831314087,
      "epoch": 1.5079509313948205,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004951952441778453,
      "loss": 5.8834,
      "mean_token_accuracy": 0.15331984907388688,
      "num_tokens": 3437262.0,
      "step": 1660
    },
    {
      "entropy": 6.109807920455933,
      "epoch": 1.5124943207632895,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004951223233595158,
      "loss": 5.8692,
      "mean_token_accuracy": 0.14938399195671082,
      "num_tokens": 3447525.0,
      "step": 1665
    },
    {
      "entropy": 6.1399143695831295,
      "epoch": 1.5170377101317583,
      "grad_norm": 0.984375,
      "learning_rate": 0.000495048859411235,
      "loss": 5.9769,
      "mean_token_accuracy": 0.1492663398385048,
      "num_tokens": 3458198.0,
      "step": 1670
    },
    {
      "entropy": 6.119841623306274,
      "epoch": 1.521581099500227,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004949748525142681,
      "loss": 5.8413,
      "mean_token_accuracy": 0.15245268642902374,
      "num_tokens": 3468149.0,
      "step": 1675
    },
    {
      "entropy": 6.060994291305542,
      "epoch": 1.526124488868696,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004949003028512195,
      "loss": 5.831,
      "mean_token_accuracy": 0.14690444469451905,
      "num_tokens": 3478355.0,
      "step": 1680
    },
    {
      "entropy": 6.075498294830322,
      "epoch": 1.530667878237165,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004948252106060334,
      "loss": 5.8679,
      "mean_token_accuracy": 0.1500585861504078,
      "num_tokens": 3488726.0,
      "step": 1685
    },
    {
      "entropy": 6.078313875198364,
      "epoch": 1.5352112676056338,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004947495759639923,
      "loss": 5.8442,
      "mean_token_accuracy": 0.15003334283828734,
      "num_tokens": 3498346.0,
      "step": 1690
    },
    {
      "entropy": 6.047679615020752,
      "epoch": 1.5397546569741025,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004946733991117171,
      "loss": 5.8935,
      "mean_token_accuracy": 0.15071313083171844,
      "num_tokens": 3508213.0,
      "step": 1695
    },
    {
      "entropy": 6.230852746963501,
      "epoch": 1.5442980463425715,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004945966802371668,
      "loss": 6.0305,
      "mean_token_accuracy": 0.14461363554000856,
      "num_tokens": 3518978.0,
      "step": 1700
    },
    {
      "entropy": 6.110667943954468,
      "epoch": 1.5488414357110405,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004945194195296374,
      "loss": 5.9616,
      "mean_token_accuracy": 0.14984720945358276,
      "num_tokens": 3529328.0,
      "step": 1705
    },
    {
      "entropy": 6.103526782989502,
      "epoch": 1.5533848250795095,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000494441617179762,
      "loss": 5.885,
      "mean_token_accuracy": 0.14589395821094514,
      "num_tokens": 3539347.0,
      "step": 1710
    },
    {
      "entropy": 6.139566421508789,
      "epoch": 1.5579282144479782,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004943632733795104,
      "loss": 5.9158,
      "mean_token_accuracy": 0.13954660817980766,
      "num_tokens": 3549798.0,
      "step": 1715
    },
    {
      "entropy": 6.10515341758728,
      "epoch": 1.562471603816447,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004942843883221879,
      "loss": 6.0097,
      "mean_token_accuracy": 0.13724498748779296,
      "num_tokens": 3560363.0,
      "step": 1720
    },
    {
      "entropy": 6.176522636413575,
      "epoch": 1.567014993184916,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004942049622024357,
      "loss": 5.9223,
      "mean_token_accuracy": 0.14831044971942903,
      "num_tokens": 3570080.0,
      "step": 1725
    },
    {
      "entropy": 6.032936000823975,
      "epoch": 1.571558382553385,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004941249952162298,
      "loss": 5.9293,
      "mean_token_accuracy": 0.14602912664413453,
      "num_tokens": 3580053.0,
      "step": 1730
    },
    {
      "entropy": 6.149723958969116,
      "epoch": 1.5761017719218537,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004940444875608807,
      "loss": 6.0176,
      "mean_token_accuracy": 0.14310100600123404,
      "num_tokens": 3590119.0,
      "step": 1735
    },
    {
      "entropy": 6.179568910598755,
      "epoch": 1.5806451612903225,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004939634394350332,
      "loss": 5.9898,
      "mean_token_accuracy": 0.13816896080970764,
      "num_tokens": 3601809.0,
      "step": 1740
    },
    {
      "entropy": 6.254966831207275,
      "epoch": 1.5851885506587915,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004938818510386654,
      "loss": 5.8641,
      "mean_token_accuracy": 0.1515677347779274,
      "num_tokens": 3611752.0,
      "step": 1745
    },
    {
      "entropy": 6.020392608642578,
      "epoch": 1.5897319400272605,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004937997225730886,
      "loss": 5.9126,
      "mean_token_accuracy": 0.14693209081888198,
      "num_tokens": 3621137.0,
      "step": 1750
    },
    {
      "entropy": 6.10863356590271,
      "epoch": 1.5942753293957292,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004937170542409464,
      "loss": 5.8777,
      "mean_token_accuracy": 0.14968368485569955,
      "num_tokens": 3631567.0,
      "step": 1755
    },
    {
      "entropy": 6.066102790832519,
      "epoch": 1.598818718764198,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000493633846246215,
      "loss": 5.845,
      "mean_token_accuracy": 0.14791938215494155,
      "num_tokens": 3641145.0,
      "step": 1760
    },
    {
      "entropy": 6.12293586730957,
      "epoch": 1.603362108132667,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004935500987942018,
      "loss": 5.8942,
      "mean_token_accuracy": 0.1458176776766777,
      "num_tokens": 3651716.0,
      "step": 1765
    },
    {
      "entropy": 6.099333572387695,
      "epoch": 1.607905497501136,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004934658120915452,
      "loss": 5.8407,
      "mean_token_accuracy": 0.15360514372587203,
      "num_tokens": 3662577.0,
      "step": 1770
    },
    {
      "entropy": 6.039719963073731,
      "epoch": 1.6124488868696047,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004933809863462142,
      "loss": 5.8231,
      "mean_token_accuracy": 0.1538111835718155,
      "num_tokens": 3672984.0,
      "step": 1775
    },
    {
      "entropy": 6.089582061767578,
      "epoch": 1.6169922762380735,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004932956217675082,
      "loss": 5.9108,
      "mean_token_accuracy": 0.15286104679107665,
      "num_tokens": 3682634.0,
      "step": 1780
    },
    {
      "entropy": 6.070204401016236,
      "epoch": 1.6215356656065425,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004932097185660555,
      "loss": 5.8804,
      "mean_token_accuracy": 0.15201435387134551,
      "num_tokens": 3693022.0,
      "step": 1785
    },
    {
      "entropy": 5.984148263931274,
      "epoch": 1.6260790549750115,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004931232769538138,
      "loss": 5.757,
      "mean_token_accuracy": 0.15402248799800872,
      "num_tokens": 3702234.0,
      "step": 1790
    },
    {
      "entropy": 6.066392850875855,
      "epoch": 1.6306224443434802,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004930362971440692,
      "loss": 5.9303,
      "mean_token_accuracy": 0.14461359679698943,
      "num_tokens": 3712235.0,
      "step": 1795
    },
    {
      "entropy": 6.167467546463013,
      "epoch": 1.635165833711949,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004929487793514358,
      "loss": 5.8695,
      "mean_token_accuracy": 0.15563700944185258,
      "num_tokens": 3722745.0,
      "step": 1800
    },
    {
      "entropy": 5.976653289794922,
      "epoch": 1.639709223080418,
      "grad_norm": 0.953125,
      "learning_rate": 0.000492860723791855,
      "loss": 5.9161,
      "mean_token_accuracy": 0.14776645004749298,
      "num_tokens": 3732942.0,
      "step": 1805
    },
    {
      "entropy": 6.151669120788574,
      "epoch": 1.644252612448887,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004927721306825952,
      "loss": 5.9,
      "mean_token_accuracy": 0.14856889694929123,
      "num_tokens": 3743806.0,
      "step": 1810
    },
    {
      "entropy": 6.032671546936035,
      "epoch": 1.6487960018173557,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004926830002422509,
      "loss": 5.9028,
      "mean_token_accuracy": 0.15380265712738037,
      "num_tokens": 3755115.0,
      "step": 1815
    },
    {
      "entropy": 6.0663093566894535,
      "epoch": 1.6533393911858245,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000492593332690743,
      "loss": 5.8501,
      "mean_token_accuracy": 0.15821865797042847,
      "num_tokens": 3765867.0,
      "step": 1820
    },
    {
      "entropy": 5.987074756622315,
      "epoch": 1.6578827805542935,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004925031282493167,
      "loss": 5.8326,
      "mean_token_accuracy": 0.14925233870744706,
      "num_tokens": 3777071.0,
      "step": 1825
    },
    {
      "entropy": 6.090896701812744,
      "epoch": 1.6624261699227625,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004924123871405428,
      "loss": 5.8382,
      "mean_token_accuracy": 0.15105062425136567,
      "num_tokens": 3786554.0,
      "step": 1830
    },
    {
      "entropy": 6.056898021697998,
      "epoch": 1.6669695592912313,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004923211095883161,
      "loss": 5.8785,
      "mean_token_accuracy": 0.15210662931203842,
      "num_tokens": 3797627.0,
      "step": 1835
    },
    {
      "entropy": 6.024951410293579,
      "epoch": 1.6715129486597,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004922292958178546,
      "loss": 5.8648,
      "mean_token_accuracy": 0.14908481538295745,
      "num_tokens": 3808841.0,
      "step": 1840
    },
    {
      "entropy": 6.096540880203247,
      "epoch": 1.676056338028169,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004921369460557,
      "loss": 5.8912,
      "mean_token_accuracy": 0.14719861075282098,
      "num_tokens": 3818995.0,
      "step": 1845
    },
    {
      "entropy": 6.090984201431274,
      "epoch": 1.680599727396638,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004920440605297159,
      "loss": 5.9448,
      "mean_token_accuracy": 0.14551446288824083,
      "num_tokens": 3828762.0,
      "step": 1850
    },
    {
      "entropy": 6.118815183639526,
      "epoch": 1.6851431167651068,
      "grad_norm": 1.015625,
      "learning_rate": 0.000491950639469088,
      "loss": 5.7884,
      "mean_token_accuracy": 0.14934536665678025,
      "num_tokens": 3838752.0,
      "step": 1855
    },
    {
      "entropy": 5.968673419952393,
      "epoch": 1.6896865061335755,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004918566831043238,
      "loss": 5.8347,
      "mean_token_accuracy": 0.1585403174161911,
      "num_tokens": 3849664.0,
      "step": 1860
    },
    {
      "entropy": 6.095662546157837,
      "epoch": 1.6942298955020445,
      "grad_norm": 1.03125,
      "learning_rate": 0.000491762191667251,
      "loss": 5.8264,
      "mean_token_accuracy": 0.15558928102254868,
      "num_tokens": 3859053.0,
      "step": 1865
    },
    {
      "entropy": 5.985531139373779,
      "epoch": 1.6987732848705135,
      "grad_norm": 1.046875,
      "learning_rate": 0.000491667165391018,
      "loss": 5.8751,
      "mean_token_accuracy": 0.14425294250249862,
      "num_tokens": 3869654.0,
      "step": 1870
    },
    {
      "entropy": 6.092062377929688,
      "epoch": 1.7033166742389823,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004915716045100927,
      "loss": 5.8641,
      "mean_token_accuracy": 0.14817842692136765,
      "num_tokens": 3879678.0,
      "step": 1875
    },
    {
      "entropy": 5.999444675445557,
      "epoch": 1.707860063607451,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004914755092602621,
      "loss": 5.7796,
      "mean_token_accuracy": 0.15495456159114837,
      "num_tokens": 3889950.0,
      "step": 1880
    },
    {
      "entropy": 6.019551515579224,
      "epoch": 1.71240345297592,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004913788798786314,
      "loss": 5.8649,
      "mean_token_accuracy": 0.15243524610996245,
      "num_tokens": 3900095.0,
      "step": 1885
    },
    {
      "entropy": 6.1658735275268555,
      "epoch": 1.716946842344389,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004912817166036243,
      "loss": 5.9303,
      "mean_token_accuracy": 0.1504468709230423,
      "num_tokens": 3911034.0,
      "step": 1890
    },
    {
      "entropy": 5.986667251586914,
      "epoch": 1.7214902317128578,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004911840196749813,
      "loss": 5.8907,
      "mean_token_accuracy": 0.14699459969997405,
      "num_tokens": 3922219.0,
      "step": 1895
    },
    {
      "entropy": 6.05580701828003,
      "epoch": 1.7260336210813265,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004910857893337602,
      "loss": 5.862,
      "mean_token_accuracy": 0.15054864585399627,
      "num_tokens": 3932235.0,
      "step": 1900
    },
    {
      "entropy": 5.985044717788696,
      "epoch": 1.7305770104497955,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004909870258223342,
      "loss": 5.8085,
      "mean_token_accuracy": 0.15058288127183914,
      "num_tokens": 3941845.0,
      "step": 1905
    },
    {
      "entropy": 6.070745754241943,
      "epoch": 1.7351203998182645,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004908877293843926,
      "loss": 5.8807,
      "mean_token_accuracy": 0.1437092550098896,
      "num_tokens": 3953392.0,
      "step": 1910
    },
    {
      "entropy": 6.0344319343566895,
      "epoch": 1.7396637891867333,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004907879002649395,
      "loss": 5.745,
      "mean_token_accuracy": 0.1542268216609955,
      "num_tokens": 3963294.0,
      "step": 1915
    },
    {
      "entropy": 5.939865493774414,
      "epoch": 1.744207178555202,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004906875387102935,
      "loss": 5.8135,
      "mean_token_accuracy": 0.14705657958984375,
      "num_tokens": 3973660.0,
      "step": 1920
    },
    {
      "entropy": 6.044660520553589,
      "epoch": 1.748750567923671,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004905866449680864,
      "loss": 5.8148,
      "mean_token_accuracy": 0.15370839834213257,
      "num_tokens": 3983587.0,
      "step": 1925
    },
    {
      "entropy": 6.041128301620484,
      "epoch": 1.75329395729214,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004904852192872638,
      "loss": 5.8705,
      "mean_token_accuracy": 0.1499823346734047,
      "num_tokens": 3994346.0,
      "step": 1930
    },
    {
      "entropy": 6.030775451660157,
      "epoch": 1.757837346660609,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004903832619180834,
      "loss": 5.8531,
      "mean_token_accuracy": 0.15244520008563994,
      "num_tokens": 4005234.0,
      "step": 1935
    },
    {
      "entropy": 5.948519134521485,
      "epoch": 1.7623807360290777,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004902807731121147,
      "loss": 5.7664,
      "mean_token_accuracy": 0.15548717826604844,
      "num_tokens": 4015542.0,
      "step": 1940
    },
    {
      "entropy": 6.020337200164795,
      "epoch": 1.7669241253975465,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004901777531222389,
      "loss": 5.7732,
      "mean_token_accuracy": 0.15857137143611907,
      "num_tokens": 4024874.0,
      "step": 1945
    },
    {
      "entropy": 6.0677588939666744,
      "epoch": 1.7714675147660155,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004900742022026474,
      "loss": 5.9227,
      "mean_token_accuracy": 0.15153265595436097,
      "num_tokens": 4035880.0,
      "step": 1950
    },
    {
      "entropy": 6.039951133728027,
      "epoch": 1.7760109041344845,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004899701206088419,
      "loss": 5.8129,
      "mean_token_accuracy": 0.15864262878894805,
      "num_tokens": 4046484.0,
      "step": 1955
    },
    {
      "entropy": 6.101760339736939,
      "epoch": 1.7805542935029532,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004898655085976333,
      "loss": 5.8685,
      "mean_token_accuracy": 0.15120580941438674,
      "num_tokens": 4057037.0,
      "step": 1960
    },
    {
      "entropy": 6.0240397453308105,
      "epoch": 1.785097682871422,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004897603664271413,
      "loss": 5.9492,
      "mean_token_accuracy": 0.14803495183587073,
      "num_tokens": 4066169.0,
      "step": 1965
    },
    {
      "entropy": 6.03475775718689,
      "epoch": 1.789641072239891,
      "grad_norm": 1.015625,
      "learning_rate": 0.000489654694356794,
      "loss": 5.8525,
      "mean_token_accuracy": 0.14776365533471109,
      "num_tokens": 4076697.0,
      "step": 1970
    },
    {
      "entropy": 5.9979242324829105,
      "epoch": 1.79418446160836,
      "grad_norm": 1.0,
      "learning_rate": 0.0004895484926473265,
      "loss": 5.7369,
      "mean_token_accuracy": 0.15955950319766998,
      "num_tokens": 4087372.0,
      "step": 1975
    },
    {
      "entropy": 6.044879007339477,
      "epoch": 1.7987278509768287,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000489441761560781,
      "loss": 5.8513,
      "mean_token_accuracy": 0.14841177314519882,
      "num_tokens": 4097653.0,
      "step": 1980
    },
    {
      "entropy": 6.096712970733643,
      "epoch": 1.8032712403452975,
      "grad_norm": 0.9296875,
      "learning_rate": 0.000489334501360506,
      "loss": 5.789,
      "mean_token_accuracy": 0.15217158794403077,
      "num_tokens": 4107381.0,
      "step": 1985
    },
    {
      "entropy": 6.004293537139892,
      "epoch": 1.8078146297137665,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004892267123111551,
      "loss": 5.879,
      "mean_token_accuracy": 0.14941695034503938,
      "num_tokens": 4118762.0,
      "step": 1990
    },
    {
      "entropy": 5.994526815414429,
      "epoch": 1.8123580190822355,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004891183946786873,
      "loss": 5.7885,
      "mean_token_accuracy": 0.1529120072722435,
      "num_tokens": 4129011.0,
      "step": 1995
    },
    {
      "entropy": 6.1039690494537355,
      "epoch": 1.8169014084507042,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004890095487303655,
      "loss": 5.8606,
      "mean_token_accuracy": 0.14621007740497588,
      "num_tokens": 4138975.0,
      "step": 2000
    },
    {
      "epoch": 1.8169014084507042,
      "eval_entropy": 5.77301181748856,
      "eval_loss": 5.996906757354736,
      "eval_mean_token_accuracy": 0.15077170481862023,
      "eval_num_tokens": 4138975.0,
      "eval_runtime": 2.0213,
      "eval_samples_per_second": 1699.416,
      "eval_steps_per_second": 212.736,
      "step": 2000
    },
    {
      "entropy": 6.0030182838439945,
      "epoch": 1.821444797819173,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004889001747347562,
      "loss": 5.8817,
      "mean_token_accuracy": 0.1444647118449211,
      "num_tokens": 4150164.0,
      "step": 2005
    },
    {
      "entropy": 5.977216529846191,
      "epoch": 1.825988187187642,
      "grad_norm": 1.109375,
      "learning_rate": 0.000488790272961729,
      "loss": 5.7721,
      "mean_token_accuracy": 0.15824113339185714,
      "num_tokens": 4159872.0,
      "step": 2010
    },
    {
      "entropy": 5.98731369972229,
      "epoch": 1.830531576556111,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004886798436824556,
      "loss": 5.8545,
      "mean_token_accuracy": 0.14761832132935523,
      "num_tokens": 4170665.0,
      "step": 2015
    },
    {
      "entropy": 6.060095405578613,
      "epoch": 1.8350749659245797,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004885688871694091,
      "loss": 5.7386,
      "mean_token_accuracy": 0.16259402483701707,
      "num_tokens": 4180476.0,
      "step": 2020
    },
    {
      "entropy": 6.02204909324646,
      "epoch": 1.8396183552930485,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004884574036963639,
      "loss": 5.8705,
      "mean_token_accuracy": 0.15071415081620215,
      "num_tokens": 4190740.0,
      "step": 2025
    },
    {
      "entropy": 5.967234563827515,
      "epoch": 1.8441617446615175,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004883453935383941,
      "loss": 5.7552,
      "mean_token_accuracy": 0.15657350271940232,
      "num_tokens": 4201504.0,
      "step": 2030
    },
    {
      "entropy": 5.858735990524292,
      "epoch": 1.8487051340299865,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004882328569718738,
      "loss": 5.7043,
      "mean_token_accuracy": 0.1618433803319931,
      "num_tokens": 4211614.0,
      "step": 2035
    },
    {
      "entropy": 6.028314018249512,
      "epoch": 1.8532485233984552,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004881197942744758,
      "loss": 5.8137,
      "mean_token_accuracy": 0.15237652361392975,
      "num_tokens": 4221141.0,
      "step": 2040
    },
    {
      "entropy": 5.973033046722412,
      "epoch": 1.857791912766924,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004880062057251711,
      "loss": 5.8692,
      "mean_token_accuracy": 0.1551216259598732,
      "num_tokens": 4231877.0,
      "step": 2045
    },
    {
      "entropy": 5.900773811340332,
      "epoch": 1.862335302135393,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000487892091604228,
      "loss": 5.6775,
      "mean_token_accuracy": 0.1587761864066124,
      "num_tokens": 4242331.0,
      "step": 2050
    },
    {
      "entropy": 5.9782391548156735,
      "epoch": 1.866878691503862,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048777745219321186,
      "loss": 5.6977,
      "mean_token_accuracy": 0.154694864153862,
      "num_tokens": 4252406.0,
      "step": 2055
    },
    {
      "entropy": 5.941512489318848,
      "epoch": 1.8714220808723308,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00048766228777498405,
      "loss": 5.7876,
      "mean_token_accuracy": 0.15253498554229736,
      "num_tokens": 4263036.0,
      "step": 2060
    },
    {
      "entropy": 5.984394598007202,
      "epoch": 1.8759654702407995,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048754659863370125,
      "loss": 5.7579,
      "mean_token_accuracy": 0.15677025616168977,
      "num_tokens": 4273032.0,
      "step": 2065
    },
    {
      "entropy": 5.966323184967041,
      "epoch": 1.8805088596092685,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048743038505481503,
      "loss": 5.8364,
      "mean_token_accuracy": 0.15176032930612565,
      "num_tokens": 4283360.0,
      "step": 2070
    },
    {
      "entropy": 5.9785511016845705,
      "epoch": 1.8850522489777375,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004873136473250708,
      "loss": 5.7799,
      "mean_token_accuracy": 0.15718763768672944,
      "num_tokens": 4292623.0,
      "step": 2075
    },
    {
      "entropy": 5.977948713302612,
      "epoch": 1.8895956383462063,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00048719638573250715,
      "loss": 5.7919,
      "mean_token_accuracy": 0.158698408305645,
      "num_tokens": 4302817.0,
      "step": 2080
    },
    {
      "entropy": 6.035198163986206,
      "epoch": 1.894139027714675,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004870786005664555,
      "loss": 5.7915,
      "mean_token_accuracy": 0.1563709110021591,
      "num_tokens": 4313049.0,
      "step": 2085
    },
    {
      "entropy": 5.93516058921814,
      "epoch": 1.898682417083144,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004869602921175392,
      "loss": 5.6693,
      "mean_token_accuracy": 0.16123159378767013,
      "num_tokens": 4322873.0,
      "step": 2090
    },
    {
      "entropy": 5.913444948196411,
      "epoch": 1.903225806451613,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004868414606776724,
      "loss": 5.8258,
      "mean_token_accuracy": 0.14879332184791566,
      "num_tokens": 4333339.0,
      "step": 2095
    },
    {
      "entropy": 6.007962846755982,
      "epoch": 1.9077691958200818,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00048672210654005996,
      "loss": 5.7947,
      "mean_token_accuracy": 0.15515013188123702,
      "num_tokens": 4344917.0,
      "step": 2100
    },
    {
      "entropy": 5.8405719757080075,
      "epoch": 1.9123125851885505,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048660222999919616,
      "loss": 5.65,
      "mean_token_accuracy": 0.16141374558210372,
      "num_tokens": 4354140.0,
      "step": 2105
    },
    {
      "entropy": 5.947803544998169,
      "epoch": 1.9168559745570195,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004864818313508647,
      "loss": 5.8904,
      "mean_token_accuracy": 0.14953781366348268,
      "num_tokens": 4364976.0,
      "step": 2110
    },
    {
      "entropy": 5.978219223022461,
      "epoch": 1.9213993639254885,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004863609108921372,
      "loss": 5.6633,
      "mean_token_accuracy": 0.1581929862499237,
      "num_tokens": 4376423.0,
      "step": 2115
    },
    {
      "entropy": 5.968920946121216,
      "epoch": 1.9259427532939573,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004862394689213728,
      "loss": 5.7977,
      "mean_token_accuracy": 0.15216519683599472,
      "num_tokens": 4387865.0,
      "step": 2120
    },
    {
      "entropy": 5.940340948104859,
      "epoch": 1.930486142662426,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004861175057382175,
      "loss": 5.7982,
      "mean_token_accuracy": 0.15568334609270096,
      "num_tokens": 4398275.0,
      "step": 2125
    },
    {
      "entropy": 6.0297112464904785,
      "epoch": 1.935029532030895,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004859950216436034,
      "loss": 5.7144,
      "mean_token_accuracy": 0.15438816994428634,
      "num_tokens": 4407396.0,
      "step": 2130
    },
    {
      "entropy": 6.145289564132691,
      "epoch": 1.939572921399364,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004858720169397478,
      "loss": 5.9707,
      "mean_token_accuracy": 0.14780526086688042,
      "num_tokens": 4419682.0,
      "step": 2135
    },
    {
      "entropy": 5.941552352905274,
      "epoch": 1.9441163107678328,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004857484919301525,
      "loss": 5.8372,
      "mean_token_accuracy": 0.15454522371292115,
      "num_tokens": 4430590.0,
      "step": 2140
    },
    {
      "entropy": 5.9843121528625485,
      "epoch": 1.9486597001363015,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004856244469196034,
      "loss": 5.7445,
      "mean_token_accuracy": 0.1609554573893547,
      "num_tokens": 4441979.0,
      "step": 2145
    },
    {
      "entropy": 5.954929447174072,
      "epoch": 1.9532030895047705,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000485499882214169,
      "loss": 5.7625,
      "mean_token_accuracy": 0.15590231418609618,
      "num_tokens": 4451975.0,
      "step": 2150
    },
    {
      "entropy": 5.888317966461182,
      "epoch": 1.9577464788732395,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004853747981212003,
      "loss": 5.7169,
      "mean_token_accuracy": 0.15523761063814162,
      "num_tokens": 4462778.0,
      "step": 2155
    },
    {
      "entropy": 6.031228065490723,
      "epoch": 1.9622898682417085,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004852491949493301,
      "loss": 5.7566,
      "mean_token_accuracy": 0.16247936636209487,
      "num_tokens": 4472461.0,
      "step": 2160
    },
    {
      "entropy": 5.948527526855469,
      "epoch": 1.9668332576101772,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004851230730084716,
      "loss": 5.7118,
      "mean_token_accuracy": 0.1612512245774269,
      "num_tokens": 4482507.0,
      "step": 2165
    },
    {
      "entropy": 5.965464496612549,
      "epoch": 1.971376646978646,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00048499643260981816,
      "loss": 5.7382,
      "mean_token_accuracy": 0.16152099072933196,
      "num_tokens": 4492601.0,
      "step": 2170
    },
    {
      "entropy": 5.95245327949524,
      "epoch": 1.975920036347115,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004848692740658424,
      "loss": 5.8439,
      "mean_token_accuracy": 0.15757981091737747,
      "num_tokens": 4503696.0,
      "step": 2175
    },
    {
      "entropy": 5.935918807983398,
      "epoch": 1.980463425715584,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004847415976902954,
      "loss": 5.7342,
      "mean_token_accuracy": 0.15598483383655548,
      "num_tokens": 4513649.0,
      "step": 2180
    },
    {
      "entropy": 5.9791747570037845,
      "epoch": 1.9850068150840527,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048461340379820587,
      "loss": 5.7571,
      "mean_token_accuracy": 0.15890643149614334,
      "num_tokens": 4524054.0,
      "step": 2185
    },
    {
      "entropy": 6.00588173866272,
      "epoch": 1.9895502044525215,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048448469270587945,
      "loss": 5.7411,
      "mean_token_accuracy": 0.14926089346408844,
      "num_tokens": 4535549.0,
      "step": 2190
    },
    {
      "entropy": 5.926804304122925,
      "epoch": 1.9940935938209905,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00048435546473089814,
      "loss": 5.7689,
      "mean_token_accuracy": 0.15297144651412964,
      "num_tokens": 4545654.0,
      "step": 2195
    },
    {
      "entropy": 5.972972822189331,
      "epoch": 1.9986369831894595,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048422572019211896,
      "loss": 5.7879,
      "mean_token_accuracy": 0.15464622229337693,
      "num_tokens": 4556230.0,
      "step": 2200
    },
    {
      "entropy": 5.915124787224664,
      "epoch": 2.0027260336210815,
      "grad_norm": 0.8828125,
      "learning_rate": 0.00048409545940967375,
      "loss": 5.5392,
      "mean_token_accuracy": 0.16189659469657475,
      "num_tokens": 4564486.0,
      "step": 2205
    },
    {
      "entropy": 5.868302679061889,
      "epoch": 2.00726942298955,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000483964682704968,
      "loss": 5.4421,
      "mean_token_accuracy": 0.16875875890254974,
      "num_tokens": 4575026.0,
      "step": 2210
    },
    {
      "entropy": 5.845345067977905,
      "epoch": 2.011812812358019,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048383339040068024,
      "loss": 5.4781,
      "mean_token_accuracy": 0.1636953830718994,
      "num_tokens": 4585902.0,
      "step": 2215
    },
    {
      "entropy": 6.082960891723633,
      "epoch": 2.016356201726488,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004837015828207612,
      "loss": 5.5773,
      "mean_token_accuracy": 0.16727212071418762,
      "num_tokens": 4597609.0,
      "step": 2220
    },
    {
      "entropy": 5.8468077182769775,
      "epoch": 2.020899591094957,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004835692602904329,
      "loss": 5.4492,
      "mean_token_accuracy": 0.17395626306533812,
      "num_tokens": 4607531.0,
      "step": 2225
    },
    {
      "entropy": 5.864391136169433,
      "epoch": 2.0254429804634255,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004834364231361882,
      "loss": 5.4745,
      "mean_token_accuracy": 0.16630882173776626,
      "num_tokens": 4618826.0,
      "step": 2230
    },
    {
      "entropy": 5.952245712280273,
      "epoch": 2.0299863698318945,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004833030716857894,
      "loss": 5.4981,
      "mean_token_accuracy": 0.17101365178823472,
      "num_tokens": 4629526.0,
      "step": 2235
    },
    {
      "entropy": 5.907699346542358,
      "epoch": 2.0345297592003635,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048316920626826807,
      "loss": 5.52,
      "mean_token_accuracy": 0.1672904148697853,
      "num_tokens": 4639930.0,
      "step": 2240
    },
    {
      "entropy": 5.886648368835449,
      "epoch": 2.0390731485688325,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048303482721392363,
      "loss": 5.4154,
      "mean_token_accuracy": 0.17574702948331833,
      "num_tokens": 4649661.0,
      "step": 2245
    },
    {
      "entropy": 5.764517688751221,
      "epoch": 2.043616537937301,
      "grad_norm": 0.86328125,
      "learning_rate": 0.00048289993485432325,
      "loss": 5.4134,
      "mean_token_accuracy": 0.1753751739859581,
      "num_tokens": 4660178.0,
      "step": 2250
    },
    {
      "entropy": 5.907838964462281,
      "epoch": 2.04815992730577,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048276452952230036,
      "loss": 5.471,
      "mean_token_accuracy": 0.16770458370447158,
      "num_tokens": 4670417.0,
      "step": 2255
    },
    {
      "entropy": 5.821921634674072,
      "epoch": 2.052703316674239,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048262861155195407,
      "loss": 5.424,
      "mean_token_accuracy": 0.1805216059088707,
      "num_tokens": 4680221.0,
      "step": 2260
    },
    {
      "entropy": 5.863324069976807,
      "epoch": 2.057246706042708,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004824921812786486,
      "loss": 5.473,
      "mean_token_accuracy": 0.16433386206626893,
      "num_tokens": 4691136.0,
      "step": 2265
    },
    {
      "entropy": 5.720443773269653,
      "epoch": 2.0617900954111765,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000482355239039012,
      "loss": 5.3165,
      "mean_token_accuracy": 0.18293814212083817,
      "num_tokens": 4700879.0,
      "step": 2270
    },
    {
      "entropy": 5.850162220001221,
      "epoch": 2.0663334847796455,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004822177851709356,
      "loss": 5.4953,
      "mean_token_accuracy": 0.16432842761278152,
      "num_tokens": 4711477.0,
      "step": 2275
    },
    {
      "entropy": 5.857503032684326,
      "epoch": 2.0708768741481145,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048207982001357327,
      "loss": 5.4756,
      "mean_token_accuracy": 0.17294807881116867,
      "num_tokens": 4721973.0,
      "step": 2280
    },
    {
      "entropy": 5.782174730300904,
      "epoch": 2.0754202635165835,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004819413439073402,
      "loss": 5.459,
      "mean_token_accuracy": 0.1705305427312851,
      "num_tokens": 4732425.0,
      "step": 2285
    },
    {
      "entropy": 5.843368148803711,
      "epoch": 2.079963652885052,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004818023571939125,
      "loss": 5.4453,
      "mean_token_accuracy": 0.17461275309324265,
      "num_tokens": 4743377.0,
      "step": 2290
    },
    {
      "entropy": 5.879863929748535,
      "epoch": 2.084507042253521,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000481662860216226,
      "loss": 5.5012,
      "mean_token_accuracy": 0.16959404051303864,
      "num_tokens": 4753534.0,
      "step": 2295
    },
    {
      "entropy": 5.800211668014526,
      "epoch": 2.08905043162199,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048152285331847574,
      "loss": 5.4677,
      "mean_token_accuracy": 0.16852423399686814,
      "num_tokens": 4763159.0,
      "step": 2300
    },
    {
      "entropy": 5.798802137374878,
      "epoch": 2.093593820990459,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004813823368461148,
      "loss": 5.4238,
      "mean_token_accuracy": 0.17590796947479248,
      "num_tokens": 4773497.0,
      "step": 2305
    },
    {
      "entropy": 5.817171812057495,
      "epoch": 2.098137210358928,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004812413111458534,
      "loss": 5.5236,
      "mean_token_accuracy": 0.17162022292613982,
      "num_tokens": 4784291.0,
      "step": 2310
    },
    {
      "entropy": 5.813042688369751,
      "epoch": 2.1026805997273965,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004810997765656587,
      "loss": 5.4691,
      "mean_token_accuracy": 0.17058789283037185,
      "num_tokens": 4794762.0,
      "step": 2315
    },
    {
      "entropy": 5.961149787902832,
      "epoch": 2.1072239890958655,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048095773345475294,
      "loss": 5.5994,
      "mean_token_accuracy": 0.1690542832016945,
      "num_tokens": 4804449.0,
      "step": 2320
    },
    {
      "entropy": 5.824402856826782,
      "epoch": 2.1117673784643345,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048081518216361367,
      "loss": 5.4483,
      "mean_token_accuracy": 0.17529074400663375,
      "num_tokens": 4814172.0,
      "step": 2325
    },
    {
      "entropy": 5.839179754257202,
      "epoch": 2.1163107678328035,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004806721230439719,
      "loss": 5.5368,
      "mean_token_accuracy": 0.16765836775302886,
      "num_tokens": 4824909.0,
      "step": 2330
    },
    {
      "entropy": 5.90317006111145,
      "epoch": 2.120854157201272,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004805285564488117,
      "loss": 5.581,
      "mean_token_accuracy": 0.17135870456695557,
      "num_tokens": 4834546.0,
      "step": 2335
    },
    {
      "entropy": 5.81043848991394,
      "epoch": 2.125397546569741,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004803844827323696,
      "loss": 5.4821,
      "mean_token_accuracy": 0.17135401517152787,
      "num_tokens": 4844809.0,
      "step": 2340
    },
    {
      "entropy": 5.8183534145355225,
      "epoch": 2.12994093593821,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004802399022501329,
      "loss": 5.4689,
      "mean_token_accuracy": 0.1713588982820511,
      "num_tokens": 4855883.0,
      "step": 2345
    },
    {
      "entropy": 5.844845867156982,
      "epoch": 2.134484325306679,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004800948153588397,
      "loss": 5.5044,
      "mean_token_accuracy": 0.173297718167305,
      "num_tokens": 4865936.0,
      "step": 2350
    },
    {
      "entropy": 5.786552810668946,
      "epoch": 2.1390277146751475,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00047994922241647746,
      "loss": 5.4245,
      "mean_token_accuracy": 0.17181466668844222,
      "num_tokens": 4876621.0,
      "step": 2355
    },
    {
      "entropy": 5.680826997756958,
      "epoch": 2.1435711040436165,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047980312378228244,
      "loss": 5.4063,
      "mean_token_accuracy": 0.16907930374145508,
      "num_tokens": 4886315.0,
      "step": 2360
    },
    {
      "entropy": 5.817247343063355,
      "epoch": 2.1481144934120855,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047965651981673823,
      "loss": 5.4871,
      "mean_token_accuracy": 0.1724187269806862,
      "num_tokens": 4897214.0,
      "step": 2365
    },
    {
      "entropy": 5.777086544036865,
      "epoch": 2.1526578827805545,
      "grad_norm": 0.96875,
      "learning_rate": 0.00047950941088157576,
      "loss": 5.4859,
      "mean_token_accuracy": 0.17181222438812255,
      "num_tokens": 4907360.0,
      "step": 2370
    },
    {
      "entropy": 5.832090520858765,
      "epoch": 2.157201272149023,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004793617973397715,
      "loss": 5.4438,
      "mean_token_accuracy": 0.17732187807559968,
      "num_tokens": 4917591.0,
      "step": 2375
    },
    {
      "entropy": 5.823357343673706,
      "epoch": 2.161744661517492,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047921367955554745,
      "loss": 5.5262,
      "mean_token_accuracy": 0.1624151811003685,
      "num_tokens": 4928075.0,
      "step": 2380
    },
    {
      "entropy": 5.799165534973144,
      "epoch": 2.166288050885961,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004790650578943693,
      "loss": 5.4252,
      "mean_token_accuracy": 0.17624630630016327,
      "num_tokens": 4937898.0,
      "step": 2385
    },
    {
      "entropy": 5.806081628799438,
      "epoch": 2.17083144025443,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00047891593272294636,
      "loss": 5.4877,
      "mean_token_accuracy": 0.16705317199230194,
      "num_tokens": 4948152.0,
      "step": 2390
    },
    {
      "entropy": 5.839706134796143,
      "epoch": 2.1753748296228985,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047876630440923,
      "loss": 5.4654,
      "mean_token_accuracy": 0.17072801142930985,
      "num_tokens": 4957973.0,
      "step": 2395
    },
    {
      "entropy": 5.655703830718994,
      "epoch": 2.1799182189913675,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004786161733224134,
      "loss": 5.4288,
      "mean_token_accuracy": 0.1745622530579567,
      "num_tokens": 4967375.0,
      "step": 2400
    },
    {
      "entropy": 5.7824067115783695,
      "epoch": 2.1844616083598365,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047846553983293006,
      "loss": 5.433,
      "mean_token_accuracy": 0.175946144759655,
      "num_tokens": 4977286.0,
      "step": 2405
    },
    {
      "entropy": 5.73899655342102,
      "epoch": 2.1890049977283055,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047831440431245307,
      "loss": 5.4962,
      "mean_token_accuracy": 0.1643753618001938,
      "num_tokens": 4987930.0,
      "step": 2410
    },
    {
      "entropy": 5.910732936859131,
      "epoch": 2.193548387096774,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047816276713389424,
      "loss": 5.5566,
      "mean_token_accuracy": 0.1674664482474327,
      "num_tokens": 4998707.0,
      "step": 2415
    },
    {
      "entropy": 5.692483139038086,
      "epoch": 2.198091776465243,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004780106286714034,
      "loss": 5.4128,
      "mean_token_accuracy": 0.1799857273697853,
      "num_tokens": 5008540.0,
      "step": 2420
    },
    {
      "entropy": 5.769286012649536,
      "epoch": 2.202635165833712,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000477857989300367,
      "loss": 5.4721,
      "mean_token_accuracy": 0.17131308019161223,
      "num_tokens": 5018199.0,
      "step": 2425
    },
    {
      "entropy": 5.860927391052246,
      "epoch": 2.207178555202181,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047770484939740757,
      "loss": 5.479,
      "mean_token_accuracy": 0.1737558886408806,
      "num_tokens": 5028251.0,
      "step": 2430
    },
    {
      "entropy": 5.736874771118164,
      "epoch": 2.2117219445706495,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004775512093403824,
      "loss": 5.4791,
      "mean_token_accuracy": 0.1656083032488823,
      "num_tokens": 5038290.0,
      "step": 2435
    },
    {
      "entropy": 5.770257186889649,
      "epoch": 2.2162653339391185,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047739706950838327,
      "loss": 5.4409,
      "mean_token_accuracy": 0.1719879612326622,
      "num_tokens": 5048337.0,
      "step": 2440
    },
    {
      "entropy": 5.858885717391968,
      "epoch": 2.2208087233075875,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00047724243028173484,
      "loss": 5.4713,
      "mean_token_accuracy": 0.16860313415527345,
      "num_tokens": 5058533.0,
      "step": 2445
    },
    {
      "entropy": 5.788866710662842,
      "epoch": 2.2253521126760565,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047708729204199406,
      "loss": 5.4809,
      "mean_token_accuracy": 0.17258588522672652,
      "num_tokens": 5068556.0,
      "step": 2450
    },
    {
      "entropy": 5.765106821060181,
      "epoch": 2.229895502044525,
      "grad_norm": 0.96875,
      "learning_rate": 0.000476931655171949,
      "loss": 5.5676,
      "mean_token_accuracy": 0.16873901337385178,
      "num_tokens": 5079317.0,
      "step": 2455
    },
    {
      "entropy": 5.8879783153533936,
      "epoch": 2.234438891412994,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004767755200556183,
      "loss": 5.542,
      "mean_token_accuracy": 0.16907548457384108,
      "num_tokens": 5089544.0,
      "step": 2460
    },
    {
      "entropy": 5.746640348434449,
      "epoch": 2.238982280781463,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004766188870782497,
      "loss": 5.5254,
      "mean_token_accuracy": 0.16796262562274933,
      "num_tokens": 5099491.0,
      "step": 2465
    },
    {
      "entropy": 5.87215256690979,
      "epoch": 2.243525670149932,
      "grad_norm": 0.984375,
      "learning_rate": 0.00047646175662631967,
      "loss": 5.5314,
      "mean_token_accuracy": 0.16680036932229997,
      "num_tokens": 5109523.0,
      "step": 2470
    },
    {
      "entropy": 5.846458387374878,
      "epoch": 2.2480690595184005,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047630412908753183,
      "loss": 5.4624,
      "mean_token_accuracy": 0.17200492322444916,
      "num_tokens": 5119940.0,
      "step": 2475
    },
    {
      "entropy": 5.731270360946655,
      "epoch": 2.2526124488868695,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047614600485081657,
      "loss": 5.4543,
      "mean_token_accuracy": 0.17587835788726808,
      "num_tokens": 5129789.0,
      "step": 2480
    },
    {
      "entropy": 5.773299503326416,
      "epoch": 2.2571558382553385,
      "grad_norm": 0.953125,
      "learning_rate": 0.00047598738430632963,
      "loss": 5.4736,
      "mean_token_accuracy": 0.16850591599941253,
      "num_tokens": 5139570.0,
      "step": 2485
    },
    {
      "entropy": 5.750662326812744,
      "epoch": 2.2616992276238075,
      "grad_norm": 0.90625,
      "learning_rate": 0.00047582826784545146,
      "loss": 5.4433,
      "mean_token_accuracy": 0.16742100864648818,
      "num_tokens": 5150526.0,
      "step": 2490
    },
    {
      "entropy": 5.732232856750488,
      "epoch": 2.266242616992276,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00047566865586078615,
      "loss": 5.3944,
      "mean_token_accuracy": 0.16691581904888153,
      "num_tokens": 5160895.0,
      "step": 2495
    },
    {
      "entropy": 5.711820888519287,
      "epoch": 2.270786006360745,
      "grad_norm": 0.953125,
      "learning_rate": 0.00047550854874616047,
      "loss": 5.3854,
      "mean_token_accuracy": 0.17315618991851806,
      "num_tokens": 5171054.0,
      "step": 2500
    },
    {
      "epoch": 2.270786006360745,
      "eval_entropy": 5.636157930729001,
      "eval_loss": 5.90526819229126,
      "eval_mean_token_accuracy": 0.15578120942552423,
      "eval_num_tokens": 5171054.0,
      "eval_runtime": 2.024,
      "eval_samples_per_second": 1697.137,
      "eval_steps_per_second": 212.451,
      "step": 2500
    },
    {
      "entropy": 5.849636268615723,
      "epoch": 2.275329395729214,
      "grad_norm": 0.984375,
      "learning_rate": 0.00047534794689662257,
      "loss": 5.5413,
      "mean_token_accuracy": 0.15833096504211425,
      "num_tokens": 5180870.0,
      "step": 2505
    },
    {
      "entropy": 5.7907532215118405,
      "epoch": 2.279872785097683,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004751868507084418,
      "loss": 5.4966,
      "mean_token_accuracy": 0.1661762699484825,
      "num_tokens": 5190388.0,
      "step": 2510
    },
    {
      "entropy": 5.8149584293365475,
      "epoch": 2.284416174466152,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004750252605791068,
      "loss": 5.4615,
      "mean_token_accuracy": 0.16411382257938384,
      "num_tokens": 5200582.0,
      "step": 2515
    },
    {
      "entropy": 5.760423040390014,
      "epoch": 2.2889595638346205,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004748631769073254,
      "loss": 5.4342,
      "mean_token_accuracy": 0.16865904480218888,
      "num_tokens": 5211079.0,
      "step": 2520
    },
    {
      "entropy": 5.801602220535278,
      "epoch": 2.2935029532030895,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004747006000930227,
      "loss": 5.5714,
      "mean_token_accuracy": 0.16202916502952575,
      "num_tokens": 5221481.0,
      "step": 2525
    },
    {
      "entropy": 5.815705060958862,
      "epoch": 2.2980463425715585,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004745375305373409,
      "loss": 5.447,
      "mean_token_accuracy": 0.1744874358177185,
      "num_tokens": 5231387.0,
      "step": 2530
    },
    {
      "entropy": 5.666153478622436,
      "epoch": 2.302589731940027,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047437396864263804,
      "loss": 5.3901,
      "mean_token_accuracy": 0.17715386897325516,
      "num_tokens": 5240758.0,
      "step": 2535
    },
    {
      "entropy": 5.777317094802856,
      "epoch": 2.307133121308496,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004742099148124866,
      "loss": 5.4235,
      "mean_token_accuracy": 0.17053455710411072,
      "num_tokens": 5250839.0,
      "step": 2540
    },
    {
      "entropy": 5.750581836700439,
      "epoch": 2.311676510676965,
      "grad_norm": 1.0,
      "learning_rate": 0.0004740453694516734,
      "loss": 5.5259,
      "mean_token_accuracy": 0.16632520109415055,
      "num_tokens": 5261554.0,
      "step": 2545
    },
    {
      "entropy": 5.765299463272095,
      "epoch": 2.316219900045434,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00047388033296619746,
      "loss": 5.4849,
      "mean_token_accuracy": 0.17075166702270508,
      "num_tokens": 5272422.0,
      "step": 2550
    },
    {
      "entropy": 5.768901777267456,
      "epoch": 2.320763289413903,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004737148057632701,
      "loss": 5.4227,
      "mean_token_accuracy": 0.1710539698600769,
      "num_tokens": 5282522.0,
      "step": 2555
    },
    {
      "entropy": 5.795893335342408,
      "epoch": 2.3253066787823715,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004735487882513131,
      "loss": 5.4757,
      "mean_token_accuracy": 0.17355609238147734,
      "num_tokens": 5293100.0,
      "step": 2560
    },
    {
      "entropy": 5.766564846038818,
      "epoch": 2.3298500681508405,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004733822808399583,
      "loss": 5.5089,
      "mean_token_accuracy": 0.16715673357248306,
      "num_tokens": 5303103.0,
      "step": 2565
    },
    {
      "entropy": 5.784523344039917,
      "epoch": 2.3343934575193095,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00047321528394004615,
      "loss": 5.4945,
      "mean_token_accuracy": 0.17657329887151718,
      "num_tokens": 5313765.0,
      "step": 2570
    },
    {
      "entropy": 5.85287389755249,
      "epoch": 2.3389368468877785,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047304779796362485,
      "loss": 5.5624,
      "mean_token_accuracy": 0.16382403075695037,
      "num_tokens": 5324816.0,
      "step": 2575
    },
    {
      "entropy": 5.752615022659302,
      "epoch": 2.343480236256247,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00047287982332394943,
      "loss": 5.4588,
      "mean_token_accuracy": 0.1738062635064125,
      "num_tokens": 5334882.0,
      "step": 2580
    },
    {
      "entropy": 5.724595737457276,
      "epoch": 2.348023625624716,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047271136043548057,
      "loss": 5.4589,
      "mean_token_accuracy": 0.16862281113862992,
      "num_tokens": 5344623.0,
      "step": 2585
    },
    {
      "entropy": 5.8034827709198,
      "epoch": 2.352567014993185,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004725424097138838,
      "loss": 5.4573,
      "mean_token_accuracy": 0.17020931392908095,
      "num_tokens": 5355786.0,
      "step": 2590
    },
    {
      "entropy": 5.6630857467651365,
      "epoch": 2.357110404361654,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047237297157602816,
      "loss": 5.4676,
      "mean_token_accuracy": 0.16971983313560485,
      "num_tokens": 5366139.0,
      "step": 2595
    },
    {
      "entropy": 5.8910150051116945,
      "epoch": 2.3616537937301225,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004722030464399854,
      "loss": 5.5624,
      "mean_token_accuracy": 0.1644546926021576,
      "num_tokens": 5376890.0,
      "step": 2600
    },
    {
      "entropy": 5.726341629028321,
      "epoch": 2.3661971830985915,
      "grad_norm": 1.046875,
      "learning_rate": 0.000472032634725029,
      "loss": 5.4635,
      "mean_token_accuracy": 0.1742193356156349,
      "num_tokens": 5386602.0,
      "step": 2605
    },
    {
      "entropy": 5.805834245681763,
      "epoch": 2.3707405724670605,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047186173685163273,
      "loss": 5.5376,
      "mean_token_accuracy": 0.16890167444944382,
      "num_tokens": 5397749.0,
      "step": 2610
    },
    {
      "entropy": 5.771027374267578,
      "epoch": 2.3752839618355295,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047169035324147023,
      "loss": 5.5064,
      "mean_token_accuracy": 0.16528881788253785,
      "num_tokens": 5407666.0,
      "step": 2615
    },
    {
      "entropy": 5.756225061416626,
      "epoch": 2.379827351203998,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00047151848431741363,
      "loss": 5.5793,
      "mean_token_accuracy": 0.16117600351572037,
      "num_tokens": 5418668.0,
      "step": 2620
    },
    {
      "entropy": 5.756460332870484,
      "epoch": 2.384370740572467,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004713461305035322,
      "loss": 5.4456,
      "mean_token_accuracy": 0.1783042386174202,
      "num_tokens": 5429098.0,
      "step": 2625
    },
    {
      "entropy": 5.706565999984742,
      "epoch": 2.388914129940936,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000471173292225092,
      "loss": 5.3989,
      "mean_token_accuracy": 0.17706974297761918,
      "num_tokens": 5438997.0,
      "step": 2630
    },
    {
      "entropy": 5.765311002731323,
      "epoch": 2.393457519309405,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047099996990855424,
      "loss": 5.4473,
      "mean_token_accuracy": 0.17235431224107742,
      "num_tokens": 5449315.0,
      "step": 2635
    },
    {
      "entropy": 5.753643417358399,
      "epoch": 2.3980009086778735,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00047082616398157456,
      "loss": 5.4932,
      "mean_token_accuracy": 0.16773738861083984,
      "num_tokens": 5460523.0,
      "step": 2640
    },
    {
      "entropy": 5.830755090713501,
      "epoch": 2.4025442980463425,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004706518748730017,
      "loss": 5.5191,
      "mean_token_accuracy": 0.17407918721437454,
      "num_tokens": 5470012.0,
      "step": 2645
    },
    {
      "entropy": 5.681747961044311,
      "epoch": 2.4070876874148115,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004704771030128768,
      "loss": 5.4507,
      "mean_token_accuracy": 0.17485994547605516,
      "num_tokens": 5480364.0,
      "step": 2650
    },
    {
      "entropy": 5.875375699996948,
      "epoch": 2.4116310767832805,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047030184883243203,
      "loss": 5.5932,
      "mean_token_accuracy": 0.1619901478290558,
      "num_tokens": 5492138.0,
      "step": 2655
    },
    {
      "entropy": 5.729516458511353,
      "epoch": 2.416174466151749,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047012611276408964,
      "loss": 5.3698,
      "mean_token_accuracy": 0.1795819506049156,
      "num_tokens": 5501935.0,
      "step": 2660
    },
    {
      "entropy": 5.675403738021851,
      "epoch": 2.420717855520218,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004699498952414609,
      "loss": 5.5615,
      "mean_token_accuracy": 0.16278308480978013,
      "num_tokens": 5513993.0,
      "step": 2665
    },
    {
      "entropy": 5.840928411483764,
      "epoch": 2.425261244888687,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00046977319669934506,
      "loss": 5.4896,
      "mean_token_accuracy": 0.172481207549572,
      "num_tokens": 5524822.0,
      "step": 2670
    },
    {
      "entropy": 5.776667928695678,
      "epoch": 2.429804634257156,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00046959601757372814,
      "loss": 5.4699,
      "mean_token_accuracy": 0.1702266275882721,
      "num_tokens": 5535583.0,
      "step": 2675
    },
    {
      "entropy": 5.763628625869751,
      "epoch": 2.4343480236256245,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046941835830178216,
      "loss": 5.4528,
      "mean_token_accuracy": 0.17762083411216736,
      "num_tokens": 5546199.0,
      "step": 2680
    },
    {
      "entropy": 5.731304073333741,
      "epoch": 2.4388914129940935,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046924021932186357,
      "loss": 5.4217,
      "mean_token_accuracy": 0.17636795341968536,
      "num_tokens": 5556409.0,
      "step": 2685
    },
    {
      "entropy": 5.746342039108276,
      "epoch": 2.4434348023625625,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046906160107351263,
      "loss": 5.4949,
      "mean_token_accuracy": 0.17396488785743713,
      "num_tokens": 5565854.0,
      "step": 2690
    },
    {
      "entropy": 5.7632523536682125,
      "epoch": 2.4479781917310315,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00046888250399745217,
      "loss": 5.5313,
      "mean_token_accuracy": 0.16098303645849227,
      "num_tokens": 5576194.0,
      "step": 2695
    },
    {
      "entropy": 5.767217111587525,
      "epoch": 2.4525215810995,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00046870292853558646,
      "loss": 5.435,
      "mean_token_accuracy": 0.17658399194478988,
      "num_tokens": 5586679.0,
      "step": 2700
    },
    {
      "entropy": 5.706618356704712,
      "epoch": 2.457064970467969,
      "grad_norm": 0.984375,
      "learning_rate": 0.00046852287513099997,
      "loss": 5.4678,
      "mean_token_accuracy": 0.16140853315591813,
      "num_tokens": 5597268.0,
      "step": 2705
    },
    {
      "entropy": 5.781574392318726,
      "epoch": 2.461608359836438,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004683423442279566,
      "loss": 5.4453,
      "mean_token_accuracy": 0.1683684453368187,
      "num_tokens": 5608046.0,
      "step": 2710
    },
    {
      "entropy": 5.765046501159668,
      "epoch": 2.466151749204907,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004681613362718985,
      "loss": 5.3535,
      "mean_token_accuracy": 0.17913431525230408,
      "num_tokens": 5617541.0,
      "step": 2715
    },
    {
      "entropy": 5.7153278350830075,
      "epoch": 2.470695138573376,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004679798517094447,
      "loss": 5.42,
      "mean_token_accuracy": 0.16958775669336318,
      "num_tokens": 5628428.0,
      "step": 2720
    },
    {
      "entropy": 5.795324802398682,
      "epoch": 2.4752385279418445,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00046779789098839043,
      "loss": 5.611,
      "mean_token_accuracy": 0.1603081412613392,
      "num_tokens": 5639557.0,
      "step": 2725
    },
    {
      "entropy": 5.821906805038452,
      "epoch": 2.4797819173103135,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046761545455770555,
      "loss": 5.5793,
      "mean_token_accuracy": 0.17082345634698867,
      "num_tokens": 5650586.0,
      "step": 2730
    },
    {
      "entropy": 5.813580751419067,
      "epoch": 2.4843253066787825,
      "grad_norm": 1.125,
      "learning_rate": 0.0004674325428675339,
      "loss": 5.4886,
      "mean_token_accuracy": 0.17099237740039824,
      "num_tokens": 5661525.0,
      "step": 2735
    },
    {
      "entropy": 5.7512657165527346,
      "epoch": 2.488868696047251,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004672491563691919,
      "loss": 5.5151,
      "mean_token_accuracy": 0.17055589258670806,
      "num_tokens": 5671272.0,
      "step": 2740
    },
    {
      "entropy": 5.757832193374634,
      "epoch": 2.49341208541572,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046706529551516753,
      "loss": 5.4371,
      "mean_token_accuracy": 0.17337453812360765,
      "num_tokens": 5680384.0,
      "step": 2745
    },
    {
      "entropy": 5.744565677642822,
      "epoch": 2.497955474784189,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046688096075911905,
      "loss": 5.4569,
      "mean_token_accuracy": 0.17362446933984757,
      "num_tokens": 5690983.0,
      "step": 2750
    },
    {
      "entropy": 5.731819486618042,
      "epoch": 2.502498864152658,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004666961525558743,
      "loss": 5.4706,
      "mean_token_accuracy": 0.16703911125659943,
      "num_tokens": 5702144.0,
      "step": 2755
    },
    {
      "entropy": 5.734309244155884,
      "epoch": 2.507042253521127,
      "grad_norm": 1.09375,
      "learning_rate": 0.000466510871361429,
      "loss": 5.5364,
      "mean_token_accuracy": 0.16374242752790452,
      "num_tokens": 5713283.0,
      "step": 2760
    },
    {
      "entropy": 5.783908367156982,
      "epoch": 2.5115856428895955,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00046632511763294615,
      "loss": 5.4904,
      "mean_token_accuracy": 0.16583993136882783,
      "num_tokens": 5724533.0,
      "step": 2765
    },
    {
      "entropy": 5.721942949295044,
      "epoch": 2.5161290322580645,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004661388918287546,
      "loss": 5.4478,
      "mean_token_accuracy": 0.1701280266046524,
      "num_tokens": 5735186.0,
      "step": 2770
    },
    {
      "entropy": 5.756659650802613,
      "epoch": 2.5206724216265335,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004659521944083481,
      "loss": 5.5114,
      "mean_token_accuracy": 0.16702331006526946,
      "num_tokens": 5746537.0,
      "step": 2775
    },
    {
      "entropy": 5.775514125823975,
      "epoch": 2.525215810995002,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004657650258323838,
      "loss": 5.4821,
      "mean_token_accuracy": 0.1666402906179428,
      "num_tokens": 5756666.0,
      "step": 2780
    },
    {
      "entropy": 5.770417213439941,
      "epoch": 2.529759200363471,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046557738656268166,
      "loss": 5.489,
      "mean_token_accuracy": 0.16813197433948518,
      "num_tokens": 5766870.0,
      "step": 2785
    },
    {
      "entropy": 5.729695224761963,
      "epoch": 2.53430258973194,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000465389277062223,
      "loss": 5.4768,
      "mean_token_accuracy": 0.1742140159010887,
      "num_tokens": 5776775.0,
      "step": 2790
    },
    {
      "entropy": 5.826140737533569,
      "epoch": 2.538845979100409,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046520069779514917,
      "loss": 5.4894,
      "mean_token_accuracy": 0.1699837863445282,
      "num_tokens": 5786993.0,
      "step": 2795
    },
    {
      "entropy": 5.704866170883179,
      "epoch": 2.543389368468878,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004650116492267609,
      "loss": 5.4869,
      "mean_token_accuracy": 0.17456516176462172,
      "num_tokens": 5796230.0,
      "step": 2800
    },
    {
      "entropy": 5.8377176284790036,
      "epoch": 2.5479327578373465,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004648221318235165,
      "loss": 5.4793,
      "mean_token_accuracy": 0.16887702643871308,
      "num_tokens": 5807086.0,
      "step": 2805
    },
    {
      "entropy": 5.723038625717163,
      "epoch": 2.5524761472058155,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004646321460530316,
      "loss": 5.5184,
      "mean_token_accuracy": 0.16765621900558472,
      "num_tokens": 5817897.0,
      "step": 2810
    },
    {
      "entropy": 5.768254089355469,
      "epoch": 2.5570195365742845,
      "grad_norm": 1.015625,
      "learning_rate": 0.000464441692384077,
      "loss": 5.4054,
      "mean_token_accuracy": 0.17909743785858154,
      "num_tokens": 5827794.0,
      "step": 2815
    },
    {
      "entropy": 5.641049146652222,
      "epoch": 2.561562925942753,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004642507712865784,
      "loss": 5.4935,
      "mean_token_accuracy": 0.16935296952724457,
      "num_tokens": 5837603.0,
      "step": 2820
    },
    {
      "entropy": 5.736937093734741,
      "epoch": 2.566106315311222,
      "grad_norm": 1.0,
      "learning_rate": 0.00046405938323161453,
      "loss": 5.4024,
      "mean_token_accuracy": 0.17678112238645555,
      "num_tokens": 5847013.0,
      "step": 2825
    },
    {
      "entropy": 5.739947891235351,
      "epoch": 2.570649704679691,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046386752869141624,
      "loss": 5.3962,
      "mean_token_accuracy": 0.1710570216178894,
      "num_tokens": 5856783.0,
      "step": 2830
    },
    {
      "entropy": 5.7007232189178465,
      "epoch": 2.57519309404816,
      "grad_norm": 1.125,
      "learning_rate": 0.0004636752081393658,
      "loss": 5.4047,
      "mean_token_accuracy": 0.17469320744276046,
      "num_tokens": 5866241.0,
      "step": 2835
    },
    {
      "entropy": 5.66242561340332,
      "epoch": 2.579736483416629,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046348242204999487,
      "loss": 5.4609,
      "mean_token_accuracy": 0.17164679020643234,
      "num_tokens": 5876135.0,
      "step": 2840
    },
    {
      "entropy": 5.7505950927734375,
      "epoch": 2.5842798727850975,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046328917089898396,
      "loss": 5.5089,
      "mean_token_accuracy": 0.17529316395521163,
      "num_tokens": 5886841.0,
      "step": 2845
    },
    {
      "entropy": 5.752353000640869,
      "epoch": 2.5888232621535665,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004630954551631612,
      "loss": 5.4435,
      "mean_token_accuracy": 0.17276009321212768,
      "num_tokens": 5896884.0,
      "step": 2850
    },
    {
      "entropy": 5.6858922958374025,
      "epoch": 2.5933666515220355,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046290127532050087,
      "loss": 5.4951,
      "mean_token_accuracy": 0.17261727750301362,
      "num_tokens": 5906194.0,
      "step": 2855
    },
    {
      "entropy": 5.732201480865479,
      "epoch": 2.5979100408905045,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004627066318501223,
      "loss": 5.4011,
      "mean_token_accuracy": 0.17907560914754866,
      "num_tokens": 5917217.0,
      "step": 2860
    },
    {
      "entropy": 5.697934770584107,
      "epoch": 2.602453430258973,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046251152523228906,
      "loss": 5.3475,
      "mean_token_accuracy": 0.17773282378911973,
      "num_tokens": 5926307.0,
      "step": 2865
    },
    {
      "entropy": 5.704933404922485,
      "epoch": 2.606996819627442,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004623159559484074,
      "loss": 5.4538,
      "mean_token_accuracy": 0.17124793231487273,
      "num_tokens": 5936718.0,
      "step": 2870
    },
    {
      "entropy": 5.638150882720947,
      "epoch": 2.611540208995911,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004621199244810248,
      "loss": 5.3988,
      "mean_token_accuracy": 0.1745395377278328,
      "num_tokens": 5947870.0,
      "step": 2875
    },
    {
      "entropy": 5.737524318695068,
      "epoch": 2.61608359836438,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00046192343131382964,
      "loss": 5.5731,
      "mean_token_accuracy": 0.16838356107473373,
      "num_tokens": 5959090.0,
      "step": 2880
    },
    {
      "entropy": 5.822002315521241,
      "epoch": 2.620626987732849,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00046172647693164926,
      "loss": 5.5495,
      "mean_token_accuracy": 0.16430439352989196,
      "num_tokens": 5970435.0,
      "step": 2885
    },
    {
      "entropy": 5.6967254161834715,
      "epoch": 2.6251703771013175,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004615290618204489,
      "loss": 5.3781,
      "mean_token_accuracy": 0.18003890067338943,
      "num_tokens": 5979732.0,
      "step": 2890
    },
    {
      "entropy": 5.699376964569092,
      "epoch": 2.6297137664697865,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004613311864673309,
      "loss": 5.5337,
      "mean_token_accuracy": 0.17351010888814927,
      "num_tokens": 5989965.0,
      "step": 2895
    },
    {
      "entropy": 5.8251628398895265,
      "epoch": 2.6342571558382555,
      "grad_norm": 1.125,
      "learning_rate": 0.00046113285136053286,
      "loss": 5.5009,
      "mean_token_accuracy": 0.17705196887254715,
      "num_tokens": 6000752.0,
      "step": 2900
    },
    {
      "entropy": 5.717715692520142,
      "epoch": 2.638800545206724,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000460934056989427,
      "loss": 5.4695,
      "mean_token_accuracy": 0.16725421994924544,
      "num_tokens": 6011318.0,
      "step": 2905
    },
    {
      "entropy": 5.7150286674499515,
      "epoch": 2.643343934575193,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004607348038445187,
      "loss": 5.4717,
      "mean_token_accuracy": 0.17503757178783416,
      "num_tokens": 6022024.0,
      "step": 2910
    },
    {
      "entropy": 5.652255630493164,
      "epoch": 2.647887323943662,
      "grad_norm": 0.96875,
      "learning_rate": 0.00046053509241744526,
      "loss": 5.4324,
      "mean_token_accuracy": 0.17356153279542924,
      "num_tokens": 6032118.0,
      "step": 2915
    },
    {
      "entropy": 5.7666980743408205,
      "epoch": 2.652430713312131,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046033492320097483,
      "loss": 5.4456,
      "mean_token_accuracy": 0.17727430313825607,
      "num_tokens": 6042607.0,
      "step": 2920
    },
    {
      "entropy": 5.664867258071899,
      "epoch": 2.6569741026806,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004601342966890049,
      "loss": 5.457,
      "mean_token_accuracy": 0.1739836573600769,
      "num_tokens": 6053625.0,
      "step": 2925
    },
    {
      "entropy": 5.716910886764526,
      "epoch": 2.6615174920490685,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045993321337656163,
      "loss": 5.4276,
      "mean_token_accuracy": 0.17799544036388398,
      "num_tokens": 6063823.0,
      "step": 2930
    },
    {
      "entropy": 5.716484594345093,
      "epoch": 2.6660608814175375,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000459731673759798,
      "loss": 5.4083,
      "mean_token_accuracy": 0.17423638105392455,
      "num_tokens": 6073914.0,
      "step": 2935
    },
    {
      "entropy": 5.692704772949218,
      "epoch": 2.6706042707860065,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00045952967833599303,
      "loss": 5.3949,
      "mean_token_accuracy": 0.17471350133419036,
      "num_tokens": 6085121.0,
      "step": 2940
    },
    {
      "entropy": 5.614338064193726,
      "epoch": 2.675147660154475,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004593272276035502,
      "loss": 5.3177,
      "mean_token_accuracy": 0.1846148192882538,
      "num_tokens": 6095324.0,
      "step": 2945
    },
    {
      "entropy": 5.68001651763916,
      "epoch": 2.679691049522944,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004591243220619969,
      "loss": 5.4584,
      "mean_token_accuracy": 0.18262093365192414,
      "num_tokens": 6104704.0,
      "step": 2950
    },
    {
      "entropy": 5.763574266433716,
      "epoch": 2.684234438891413,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000458920962211982,
      "loss": 5.5411,
      "mean_token_accuracy": 0.16811863780021669,
      "num_tokens": 6116569.0,
      "step": 2955
    },
    {
      "entropy": 5.7841808795928955,
      "epoch": 2.688777828259882,
      "grad_norm": 1.0,
      "learning_rate": 0.0004587171485552759,
      "loss": 5.5251,
      "mean_token_accuracy": 0.17157238423824311,
      "num_tokens": 6126166.0,
      "step": 2960
    },
    {
      "entropy": 5.709067296981812,
      "epoch": 2.693321217628351,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00045851288159476854,
      "loss": 5.3965,
      "mean_token_accuracy": 0.17820557951927185,
      "num_tokens": 6136169.0,
      "step": 2965
    },
    {
      "entropy": 5.717296457290649,
      "epoch": 2.6978646069968195,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004583081618344684,
      "loss": 5.4616,
      "mean_token_accuracy": 0.177069228887558,
      "num_tokens": 6146249.0,
      "step": 2970
    },
    {
      "entropy": 5.830132627487183,
      "epoch": 2.7024079963652885,
      "grad_norm": 0.921875,
      "learning_rate": 0.000458102989779501,
      "loss": 5.5397,
      "mean_token_accuracy": 0.1752156972885132,
      "num_tokens": 6156892.0,
      "step": 2975
    },
    {
      "entropy": 5.726925420761108,
      "epoch": 2.7069513857337575,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004578973659361081,
      "loss": 5.4199,
      "mean_token_accuracy": 0.18360553979873656,
      "num_tokens": 6166915.0,
      "step": 2980
    },
    {
      "entropy": 5.675100183486938,
      "epoch": 2.711494775102226,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045769129081164596,
      "loss": 5.3677,
      "mean_token_accuracy": 0.17813336104154587,
      "num_tokens": 6176837.0,
      "step": 2985
    },
    {
      "entropy": 5.628390121459961,
      "epoch": 2.716038164470695,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004574847649145845,
      "loss": 5.4544,
      "mean_token_accuracy": 0.17339005023241044,
      "num_tokens": 6188038.0,
      "step": 2990
    },
    {
      "entropy": 5.720661163330078,
      "epoch": 2.720581553839164,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000457277788754506,
      "loss": 5.4817,
      "mean_token_accuracy": 0.17173578292131425,
      "num_tokens": 6199835.0,
      "step": 2995
    },
    {
      "entropy": 5.7931312084197994,
      "epoch": 2.725124943207633,
      "grad_norm": 1.0,
      "learning_rate": 0.00045707036284210344,
      "loss": 5.3881,
      "mean_token_accuracy": 0.1742375299334526,
      "num_tokens": 6209122.0,
      "step": 3000
    },
    {
      "epoch": 2.725124943207633,
      "eval_entropy": 5.51928047801173,
      "eval_loss": 5.816741943359375,
      "eval_mean_token_accuracy": 0.16149290649350298,
      "eval_num_tokens": 6209122.0,
      "eval_runtime": 2.0336,
      "eval_samples_per_second": 1689.087,
      "eval_steps_per_second": 211.443,
      "step": 3000
    },
    {
      "entropy": 5.630403614044189,
      "epoch": 2.729668332576102,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004568624876891796,
      "loss": 5.4582,
      "mean_token_accuracy": 0.1750628024339676,
      "num_tokens": 6219317.0,
      "step": 3005
    },
    {
      "entropy": 5.744443941116333,
      "epoch": 2.7342117219445705,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00045665416380864587,
      "loss": 5.4982,
      "mean_token_accuracy": 0.1748853400349617,
      "num_tokens": 6229561.0,
      "step": 3010
    },
    {
      "entropy": 5.750857496261597,
      "epoch": 2.7387551113130395,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004564453917145206,
      "loss": 5.47,
      "mean_token_accuracy": 0.17052508145570755,
      "num_tokens": 6240150.0,
      "step": 3015
    },
    {
      "entropy": 5.644581031799317,
      "epoch": 2.7432985006815085,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004562361719219283,
      "loss": 5.3541,
      "mean_token_accuracy": 0.17477230727672577,
      "num_tokens": 6250641.0,
      "step": 3020
    },
    {
      "entropy": 5.684276437759399,
      "epoch": 2.747841890049977,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045602650494709786,
      "loss": 5.4413,
      "mean_token_accuracy": 0.17121798396110535,
      "num_tokens": 6261239.0,
      "step": 3025
    },
    {
      "entropy": 5.697177171707153,
      "epoch": 2.752385279418446,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00045581639130736173,
      "loss": 5.4512,
      "mean_token_accuracy": 0.17007984220981598,
      "num_tokens": 6270919.0,
      "step": 3030
    },
    {
      "entropy": 5.7353699684143065,
      "epoch": 2.756928668786915,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004556058315211546,
      "loss": 5.5087,
      "mean_token_accuracy": 0.1678515136241913,
      "num_tokens": 6283007.0,
      "step": 3035
    },
    {
      "entropy": 5.600628280639649,
      "epoch": 2.761472058155384,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00045539482610801156,
      "loss": 5.3734,
      "mean_token_accuracy": 0.18075916171073914,
      "num_tokens": 6293596.0,
      "step": 3040
    },
    {
      "entropy": 5.701785564422607,
      "epoch": 2.766015447523853,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004551833755885677,
      "loss": 5.4211,
      "mean_token_accuracy": 0.1742391288280487,
      "num_tokens": 6304568.0,
      "step": 3045
    },
    {
      "entropy": 5.713755416870117,
      "epoch": 2.7705588368923215,
      "grad_norm": 1.0,
      "learning_rate": 0.0004549714804845561,
      "loss": 5.3921,
      "mean_token_accuracy": 0.17213882356882096,
      "num_tokens": 6315275.0,
      "step": 3050
    },
    {
      "entropy": 5.740856266021728,
      "epoch": 2.7751022262607905,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004547591413188068,
      "loss": 5.5018,
      "mean_token_accuracy": 0.17504518181085588,
      "num_tokens": 6324986.0,
      "step": 3055
    },
    {
      "entropy": 5.67020320892334,
      "epoch": 2.7796456156292595,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004545463586152456,
      "loss": 5.5076,
      "mean_token_accuracy": 0.16599090695381163,
      "num_tokens": 6336318.0,
      "step": 3060
    },
    {
      "entropy": 5.666765832901001,
      "epoch": 2.784189004997728,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004543331328988927,
      "loss": 5.3917,
      "mean_token_accuracy": 0.1782316729426384,
      "num_tokens": 6345986.0,
      "step": 3065
    },
    {
      "entropy": 5.7263648986816404,
      "epoch": 2.788732394366197,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004541194646958614,
      "loss": 5.4765,
      "mean_token_accuracy": 0.17485484927892686,
      "num_tokens": 6357025.0,
      "step": 3070
    },
    {
      "entropy": 5.699453401565552,
      "epoch": 2.793275783734666,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00045390535453335666,
      "loss": 5.4684,
      "mean_token_accuracy": 0.17214902639389038,
      "num_tokens": 6367885.0,
      "step": 3075
    },
    {
      "entropy": 5.580998706817627,
      "epoch": 2.797819173103135,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045369080293967414,
      "loss": 5.3865,
      "mean_token_accuracy": 0.1732836037874222,
      "num_tokens": 6377729.0,
      "step": 3080
    },
    {
      "entropy": 5.644871044158935,
      "epoch": 2.802362562471604,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00045347581044419846,
      "loss": 5.3845,
      "mean_token_accuracy": 0.17982030361890794,
      "num_tokens": 6387368.0,
      "step": 3085
    },
    {
      "entropy": 5.755515241622925,
      "epoch": 2.806905951840073,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00045326037757740216,
      "loss": 5.4903,
      "mean_token_accuracy": 0.16899367719888686,
      "num_tokens": 6397771.0,
      "step": 3090
    },
    {
      "entropy": 5.728139019012451,
      "epoch": 2.8114493412085415,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004530445048708447,
      "loss": 5.3671,
      "mean_token_accuracy": 0.18123927861452102,
      "num_tokens": 6408238.0,
      "step": 3095
    },
    {
      "entropy": 5.689693593978882,
      "epoch": 2.8159927305770105,
      "grad_norm": 1.0,
      "learning_rate": 0.00045282819285717024,
      "loss": 5.5077,
      "mean_token_accuracy": 0.17035059332847596,
      "num_tokens": 6419196.0,
      "step": 3100
    },
    {
      "entropy": 5.6974506855010985,
      "epoch": 2.8205361199454795,
      "grad_norm": 0.9375,
      "learning_rate": 0.00045261144207010736,
      "loss": 5.4643,
      "mean_token_accuracy": 0.17225333750247956,
      "num_tokens": 6430342.0,
      "step": 3105
    },
    {
      "entropy": 5.664730215072632,
      "epoch": 2.825079509313948,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00045239425304446684,
      "loss": 5.3624,
      "mean_token_accuracy": 0.18250575959682463,
      "num_tokens": 6440218.0,
      "step": 3110
    },
    {
      "entropy": 5.656542062759399,
      "epoch": 2.829622898682417,
      "grad_norm": 1.0,
      "learning_rate": 0.0004521766263161413,
      "loss": 5.3688,
      "mean_token_accuracy": 0.1843707248568535,
      "num_tokens": 6450128.0,
      "step": 3115
    },
    {
      "entropy": 5.653999519348145,
      "epoch": 2.834166288050886,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000451958562422103,
      "loss": 5.3778,
      "mean_token_accuracy": 0.18543238192796707,
      "num_tokens": 6459916.0,
      "step": 3120
    },
    {
      "entropy": 5.707833099365234,
      "epoch": 2.838709677419355,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00045174006190040267,
      "loss": 5.4168,
      "mean_token_accuracy": 0.17574355900287628,
      "num_tokens": 6470250.0,
      "step": 3125
    },
    {
      "entropy": 5.690685749053955,
      "epoch": 2.843253066787824,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000451521125290169,
      "loss": 5.5111,
      "mean_token_accuracy": 0.1676422730088234,
      "num_tokens": 6481623.0,
      "step": 3130
    },
    {
      "entropy": 5.72694411277771,
      "epoch": 2.8477964561562925,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000451301753131606,
      "loss": 5.4067,
      "mean_token_accuracy": 0.17806993573904037,
      "num_tokens": 6492143.0,
      "step": 3135
    },
    {
      "entropy": 5.749693536758423,
      "epoch": 2.8523398455247615,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045108194596599286,
      "loss": 5.4264,
      "mean_token_accuracy": 0.17895904928445816,
      "num_tokens": 6501643.0,
      "step": 3140
    },
    {
      "entropy": 5.61626091003418,
      "epoch": 2.8568832348932305,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004508617043356818,
      "loss": 5.3662,
      "mean_token_accuracy": 0.18171129375696182,
      "num_tokens": 6511442.0,
      "step": 3145
    },
    {
      "entropy": 5.646936321258545,
      "epoch": 2.861426624261699,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004506410287840972,
      "loss": 5.2834,
      "mean_token_accuracy": 0.18167806565761566,
      "num_tokens": 6521303.0,
      "step": 3150
    },
    {
      "entropy": 5.628382539749145,
      "epoch": 2.865970013630168,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000450419919855734,
      "loss": 5.4212,
      "mean_token_accuracy": 0.17254047691822053,
      "num_tokens": 6531143.0,
      "step": 3155
    },
    {
      "entropy": 5.744618225097656,
      "epoch": 2.870513402998637,
      "grad_norm": 1.0,
      "learning_rate": 0.0004501983780961566,
      "loss": 5.4823,
      "mean_token_accuracy": 0.17173656523227693,
      "num_tokens": 6540899.0,
      "step": 3160
    },
    {
      "entropy": 5.8308022022247314,
      "epoch": 2.875056792367106,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004499764040519972,
      "loss": 5.531,
      "mean_token_accuracy": 0.1689829409122467,
      "num_tokens": 6552131.0,
      "step": 3165
    },
    {
      "entropy": 5.6553051471710205,
      "epoch": 2.879600181735575,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00044975399827095467,
      "loss": 5.4437,
      "mean_token_accuracy": 0.16968075484037398,
      "num_tokens": 6563275.0,
      "step": 3170
    },
    {
      "entropy": 5.698030757904053,
      "epoch": 2.8841435711040435,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00044953116130179313,
      "loss": 5.3142,
      "mean_token_accuracy": 0.18318966925144195,
      "num_tokens": 6573109.0,
      "step": 3175
    },
    {
      "entropy": 5.569756889343262,
      "epoch": 2.8886869604725125,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004493078936943408,
      "loss": 5.4244,
      "mean_token_accuracy": 0.16793778985738755,
      "num_tokens": 6584978.0,
      "step": 3180
    },
    {
      "entropy": 5.652692461013794,
      "epoch": 2.8932303498409815,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004490841959994882,
      "loss": 5.4301,
      "mean_token_accuracy": 0.1768659383058548,
      "num_tokens": 6595608.0,
      "step": 3185
    },
    {
      "entropy": 5.64333143234253,
      "epoch": 2.89777373920945,
      "grad_norm": 1.03125,
      "learning_rate": 0.00044886006876918716,
      "loss": 5.3221,
      "mean_token_accuracy": 0.1744397297501564,
      "num_tokens": 6605983.0,
      "step": 3190
    },
    {
      "entropy": 5.658440685272216,
      "epoch": 2.902317128577919,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004486355125564493,
      "loss": 5.4439,
      "mean_token_accuracy": 0.168673013150692,
      "num_tokens": 6616682.0,
      "step": 3195
    },
    {
      "entropy": 5.8313366889953615,
      "epoch": 2.906860517946388,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004484105279153449,
      "loss": 5.466,
      "mean_token_accuracy": 0.17580256462097169,
      "num_tokens": 6628715.0,
      "step": 3200
    },
    {
      "entropy": 5.690737056732178,
      "epoch": 2.911403907314857,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000448185115401001,
      "loss": 5.4531,
      "mean_token_accuracy": 0.171098530292511,
      "num_tokens": 6639219.0,
      "step": 3205
    },
    {
      "entropy": 5.6329793453216555,
      "epoch": 2.915947296683326,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004479592755696008,
      "loss": 5.3503,
      "mean_token_accuracy": 0.1818231984972954,
      "num_tokens": 6649516.0,
      "step": 3210
    },
    {
      "entropy": 5.746676778793335,
      "epoch": 2.9204906860517945,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044773300897838137,
      "loss": 5.4603,
      "mean_token_accuracy": 0.17531079798936844,
      "num_tokens": 6659470.0,
      "step": 3215
    },
    {
      "entropy": 5.630096864700318,
      "epoch": 2.9250340754202635,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004475063161856332,
      "loss": 5.4167,
      "mean_token_accuracy": 0.17711722254753112,
      "num_tokens": 6669637.0,
      "step": 3220
    },
    {
      "entropy": 5.599526596069336,
      "epoch": 2.9295774647887325,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004472791977506983,
      "loss": 5.3779,
      "mean_token_accuracy": 0.1744503155350685,
      "num_tokens": 6680450.0,
      "step": 3225
    },
    {
      "entropy": 5.707645654678345,
      "epoch": 2.934120854157201,
      "grad_norm": 0.984375,
      "learning_rate": 0.00044705165423396855,
      "loss": 5.4683,
      "mean_token_accuracy": 0.17224285751581192,
      "num_tokens": 6691526.0,
      "step": 3230
    },
    {
      "entropy": 5.701753950119018,
      "epoch": 2.93866424352567,
      "grad_norm": 0.9375,
      "learning_rate": 0.00044682368619688523,
      "loss": 5.403,
      "mean_token_accuracy": 0.17708637714385986,
      "num_tokens": 6702543.0,
      "step": 3235
    },
    {
      "entropy": 5.6747335433959964,
      "epoch": 2.943207632894139,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004465952942019367,
      "loss": 5.4879,
      "mean_token_accuracy": 0.1681745857000351,
      "num_tokens": 6712374.0,
      "step": 3240
    },
    {
      "entropy": 5.70989351272583,
      "epoch": 2.947751022262608,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00044636647881265745,
      "loss": 5.3799,
      "mean_token_accuracy": 0.17257385700941086,
      "num_tokens": 6722091.0,
      "step": 3245
    },
    {
      "entropy": 5.631851482391357,
      "epoch": 2.952294411631077,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004461372405936269,
      "loss": 5.3637,
      "mean_token_accuracy": 0.18677607774734498,
      "num_tokens": 6732024.0,
      "step": 3250
    },
    {
      "entropy": 5.725279092788696,
      "epoch": 2.9568378009995455,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004459075801104673,
      "loss": 5.4762,
      "mean_token_accuracy": 0.17148397564888002,
      "num_tokens": 6741480.0,
      "step": 3255
    },
    {
      "entropy": 5.711555862426758,
      "epoch": 2.9613811903680145,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00044567749792984325,
      "loss": 5.3616,
      "mean_token_accuracy": 0.17898045629262924,
      "num_tokens": 6752504.0,
      "step": 3260
    },
    {
      "entropy": 5.685667896270752,
      "epoch": 2.9659245797364835,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044544699461945966,
      "loss": 5.4192,
      "mean_token_accuracy": 0.1693477913737297,
      "num_tokens": 6762945.0,
      "step": 3265
    },
    {
      "entropy": 5.635536813735962,
      "epoch": 2.970467969104952,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00044521607074806037,
      "loss": 5.3931,
      "mean_token_accuracy": 0.17766736894845964,
      "num_tokens": 6772862.0,
      "step": 3270
    },
    {
      "entropy": 5.729037952423096,
      "epoch": 2.975011358473421,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004449847268854272,
      "loss": 5.4192,
      "mean_token_accuracy": 0.1795768156647682,
      "num_tokens": 6782912.0,
      "step": 3275
    },
    {
      "entropy": 5.732499122619629,
      "epoch": 2.97955474784189,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004447529636023782,
      "loss": 5.4429,
      "mean_token_accuracy": 0.17771411538124085,
      "num_tokens": 6793249.0,
      "step": 3280
    },
    {
      "entropy": 5.687345933914185,
      "epoch": 2.984098137210359,
      "grad_norm": 0.953125,
      "learning_rate": 0.00044452078147076614,
      "loss": 5.4553,
      "mean_token_accuracy": 0.1770368903875351,
      "num_tokens": 6803721.0,
      "step": 3285
    },
    {
      "entropy": 5.74183497428894,
      "epoch": 2.988641526578828,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004442881810634774,
      "loss": 5.4182,
      "mean_token_accuracy": 0.1799023449420929,
      "num_tokens": 6813637.0,
      "step": 3290
    },
    {
      "entropy": 5.662044906616211,
      "epoch": 2.9931849159472965,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004440551629544303,
      "loss": 5.4269,
      "mean_token_accuracy": 0.17125516533851623,
      "num_tokens": 6824171.0,
      "step": 3295
    },
    {
      "entropy": 5.615179538726807,
      "epoch": 2.9977283053157655,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00044382172771857383,
      "loss": 5.2838,
      "mean_token_accuracy": 0.18038785606622695,
      "num_tokens": 6833668.0,
      "step": 3300
    },
    {
      "entropy": 5.607766734229194,
      "epoch": 3.0018173557473875,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004435878759318864,
      "loss": 5.2214,
      "mean_token_accuracy": 0.18619143135017818,
      "num_tokens": 6842297.0,
      "step": 3305
    },
    {
      "entropy": 5.708400058746338,
      "epoch": 3.0063607451158565,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004433536081713738,
      "loss": 5.1965,
      "mean_token_accuracy": 0.18496198803186417,
      "num_tokens": 6853124.0,
      "step": 3310
    },
    {
      "entropy": 5.683142566680909,
      "epoch": 3.0109041344843255,
      "grad_norm": 1.0,
      "learning_rate": 0.0004431189250150686,
      "loss": 5.0761,
      "mean_token_accuracy": 0.19732627719640733,
      "num_tokens": 6861990.0,
      "step": 3315
    },
    {
      "entropy": 5.643543529510498,
      "epoch": 3.015447523852794,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00044288382704202816,
      "loss": 5.0725,
      "mean_token_accuracy": 0.1970564529299736,
      "num_tokens": 6872024.0,
      "step": 3320
    },
    {
      "entropy": 5.6079028129577635,
      "epoch": 3.019990913221263,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044264831483233335,
      "loss": 5.1409,
      "mean_token_accuracy": 0.19019072204828263,
      "num_tokens": 6881205.0,
      "step": 3325
    },
    {
      "entropy": 5.531577253341675,
      "epoch": 3.024534302589732,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00044241238896708716,
      "loss": 5.0945,
      "mean_token_accuracy": 0.19641977101564406,
      "num_tokens": 6891741.0,
      "step": 3330
    },
    {
      "entropy": 5.542751836776733,
      "epoch": 3.029077691958201,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004421760500284131,
      "loss": 5.1675,
      "mean_token_accuracy": 0.18492870330810546,
      "num_tokens": 6902122.0,
      "step": 3335
    },
    {
      "entropy": 5.643613529205322,
      "epoch": 3.0336210813266695,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004419392985994541,
      "loss": 5.1294,
      "mean_token_accuracy": 0.18925008177757263,
      "num_tokens": 6912129.0,
      "step": 3340
    },
    {
      "entropy": 5.559886693954468,
      "epoch": 3.0381644706951385,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044170213526437076,
      "loss": 5.0153,
      "mean_token_accuracy": 0.19870110154151915,
      "num_tokens": 6921409.0,
      "step": 3345
    },
    {
      "entropy": 5.573070955276489,
      "epoch": 3.0427078600636075,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00044146456060833994,
      "loss": 5.1324,
      "mean_token_accuracy": 0.19372012466192245,
      "num_tokens": 6932346.0,
      "step": 3350
    },
    {
      "entropy": 5.650528621673584,
      "epoch": 3.0472512494320765,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044122657521755373,
      "loss": 5.144,
      "mean_token_accuracy": 0.1928184673190117,
      "num_tokens": 6942139.0,
      "step": 3355
    },
    {
      "entropy": 5.623360443115234,
      "epoch": 3.051794638800545,
      "grad_norm": 0.96875,
      "learning_rate": 0.00044098817967921716,
      "loss": 5.1476,
      "mean_token_accuracy": 0.19087578505277633,
      "num_tokens": 6952344.0,
      "step": 3360
    },
    {
      "entropy": 5.651640224456787,
      "epoch": 3.056338028169014,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00044074937458154763,
      "loss": 5.1319,
      "mean_token_accuracy": 0.19229988306760787,
      "num_tokens": 6962430.0,
      "step": 3365
    },
    {
      "entropy": 5.654034042358399,
      "epoch": 3.060881417537483,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00044051016051377303,
      "loss": 5.1799,
      "mean_token_accuracy": 0.18911581188440324,
      "num_tokens": 6973213.0,
      "step": 3370
    },
    {
      "entropy": 5.538857269287109,
      "epoch": 3.065424806905952,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004402705380661302,
      "loss": 5.13,
      "mean_token_accuracy": 0.18752226531505584,
      "num_tokens": 6983896.0,
      "step": 3375
    },
    {
      "entropy": 5.611310291290283,
      "epoch": 3.0699681962744205,
      "grad_norm": 1.015625,
      "learning_rate": 0.00044003050782986377,
      "loss": 5.1092,
      "mean_token_accuracy": 0.19607766568660737,
      "num_tokens": 6993549.0,
      "step": 3380
    },
    {
      "entropy": 5.579682397842407,
      "epoch": 3.0745115856428895,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004397900703972245,
      "loss": 5.1446,
      "mean_token_accuracy": 0.19014745205640793,
      "num_tokens": 7003893.0,
      "step": 3385
    },
    {
      "entropy": 5.648293733596802,
      "epoch": 3.0790549750113585,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004395492263614677,
      "loss": 5.2424,
      "mean_token_accuracy": 0.18216302543878554,
      "num_tokens": 7014826.0,
      "step": 3390
    },
    {
      "entropy": 5.678756046295166,
      "epoch": 3.0835983643798275,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00043930797631685234,
      "loss": 5.1681,
      "mean_token_accuracy": 0.19137286394834518,
      "num_tokens": 7024758.0,
      "step": 3395
    },
    {
      "entropy": 5.5866437435150145,
      "epoch": 3.088141753748296,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043906632085863876,
      "loss": 5.136,
      "mean_token_accuracy": 0.19628849029541015,
      "num_tokens": 7034063.0,
      "step": 3400
    },
    {
      "entropy": 5.568763542175293,
      "epoch": 3.092685143116765,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043882426058308775,
      "loss": 5.1336,
      "mean_token_accuracy": 0.18538877815008165,
      "num_tokens": 7045137.0,
      "step": 3405
    },
    {
      "entropy": 5.520676231384277,
      "epoch": 3.097228532485234,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00043858179608745915,
      "loss": 5.0652,
      "mean_token_accuracy": 0.19561208188533782,
      "num_tokens": 7056381.0,
      "step": 3410
    },
    {
      "entropy": 5.698847913742066,
      "epoch": 3.101771921853703,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004383389279700098,
      "loss": 5.2015,
      "mean_token_accuracy": 0.18250014185905455,
      "num_tokens": 7066462.0,
      "step": 3415
    },
    {
      "entropy": 5.671674394607544,
      "epoch": 3.1063153112221715,
      "grad_norm": 0.96875,
      "learning_rate": 0.00043809565682999274,
      "loss": 5.2231,
      "mean_token_accuracy": 0.18543293178081513,
      "num_tokens": 7078438.0,
      "step": 3420
    },
    {
      "entropy": 5.60929307937622,
      "epoch": 3.1108587005906405,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00043785198326765535,
      "loss": 5.1001,
      "mean_token_accuracy": 0.192493799328804,
      "num_tokens": 7088981.0,
      "step": 3425
    },
    {
      "entropy": 5.565306806564331,
      "epoch": 3.1154020899591095,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00043760790788423785,
      "loss": 5.1574,
      "mean_token_accuracy": 0.19066230803728104,
      "num_tokens": 7099407.0,
      "step": 3430
    },
    {
      "entropy": 5.564446830749512,
      "epoch": 3.1199454793275785,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043736343128197207,
      "loss": 5.0399,
      "mean_token_accuracy": 0.2005670338869095,
      "num_tokens": 7109454.0,
      "step": 3435
    },
    {
      "entropy": 5.5073576927185055,
      "epoch": 3.124488868696047,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004371185540640796,
      "loss": 5.1398,
      "mean_token_accuracy": 0.18948666602373124,
      "num_tokens": 7119639.0,
      "step": 3440
    },
    {
      "entropy": 5.62153434753418,
      "epoch": 3.129032258064516,
      "grad_norm": 1.03125,
      "learning_rate": 0.00043687327683477077,
      "loss": 5.1077,
      "mean_token_accuracy": 0.19161372035741805,
      "num_tokens": 7129872.0,
      "step": 3445
    },
    {
      "entropy": 5.553510665893555,
      "epoch": 3.133575647432985,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004366276001992427,
      "loss": 5.0572,
      "mean_token_accuracy": 0.1962740823626518,
      "num_tokens": 7140456.0,
      "step": 3450
    },
    {
      "entropy": 5.492085123062134,
      "epoch": 3.138119036801454,
      "grad_norm": 1.140625,
      "learning_rate": 0.000436381524763678,
      "loss": 5.0866,
      "mean_token_accuracy": 0.19258544892072677,
      "num_tokens": 7149787.0,
      "step": 3455
    },
    {
      "entropy": 5.56104006767273,
      "epoch": 3.142662426169923,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004361350511352435,
      "loss": 5.0703,
      "mean_token_accuracy": 0.19688417315483092,
      "num_tokens": 7160482.0,
      "step": 3460
    },
    {
      "entropy": 5.643281745910644,
      "epoch": 3.1472058155383915,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004358881799220882,
      "loss": 5.1104,
      "mean_token_accuracy": 0.1937023714184761,
      "num_tokens": 7170818.0,
      "step": 3465
    },
    {
      "entropy": 5.535250854492188,
      "epoch": 3.1517492049068605,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043564091173334236,
      "loss": 5.1345,
      "mean_token_accuracy": 0.1935590460896492,
      "num_tokens": 7180770.0,
      "step": 3470
    },
    {
      "entropy": 5.527993488311767,
      "epoch": 3.1562925942753295,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043539324717911573,
      "loss": 5.0564,
      "mean_token_accuracy": 0.1996495321393013,
      "num_tokens": 7190751.0,
      "step": 3475
    },
    {
      "entropy": 5.549295473098755,
      "epoch": 3.1608359836437985,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043514518687049587,
      "loss": 5.0693,
      "mean_token_accuracy": 0.20097184926271439,
      "num_tokens": 7200093.0,
      "step": 3480
    },
    {
      "entropy": 5.624794006347656,
      "epoch": 3.165379373012267,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000434896731419547,
      "loss": 5.1693,
      "mean_token_accuracy": 0.19156304150819778,
      "num_tokens": 7210526.0,
      "step": 3485
    },
    {
      "entropy": 5.512590074539185,
      "epoch": 3.169922762380736,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004346478814393082,
      "loss": 5.1222,
      "mean_token_accuracy": 0.19338342547416687,
      "num_tokens": 7222127.0,
      "step": 3490
    },
    {
      "entropy": 5.549171590805054,
      "epoch": 3.174466151749205,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004343986375437921,
      "loss": 5.0886,
      "mean_token_accuracy": 0.19761230945587158,
      "num_tokens": 7233418.0,
      "step": 3495
    },
    {
      "entropy": 5.546758079528809,
      "epoch": 3.179009541117674,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043414900034798314,
      "loss": 5.1424,
      "mean_token_accuracy": 0.19149672836065293,
      "num_tokens": 7243743.0,
      "step": 3500
    },
    {
      "epoch": 3.179009541117674,
      "eval_entropy": 5.3101175984670945,
      "eval_loss": 5.796391487121582,
      "eval_mean_token_accuracy": 0.16366152772030165,
      "eval_num_tokens": 7243743.0,
      "eval_runtime": 2.0265,
      "eval_samples_per_second": 1695.046,
      "eval_steps_per_second": 212.189,
      "step": 3500
    },
    {
      "entropy": 5.5297328472137455,
      "epoch": 3.1835529304861425,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004338989704678364,
      "loss": 5.1766,
      "mean_token_accuracy": 0.18635060936212539,
      "num_tokens": 7254906.0,
      "step": 3505
    },
    {
      "entropy": 5.581528568267823,
      "epoch": 3.1880963198546115,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043364854852027566,
      "loss": 5.1295,
      "mean_token_accuracy": 0.1920384258031845,
      "num_tokens": 7264193.0,
      "step": 3510
    },
    {
      "entropy": 5.64374623298645,
      "epoch": 3.1926397092230805,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004333977351231924,
      "loss": 5.1828,
      "mean_token_accuracy": 0.1904108554124832,
      "num_tokens": 7273723.0,
      "step": 3515
    },
    {
      "entropy": 5.548360109329224,
      "epoch": 3.1971830985915495,
      "grad_norm": 1.0,
      "learning_rate": 0.00043314653089544336,
      "loss": 5.1979,
      "mean_token_accuracy": 0.18939443975687026,
      "num_tokens": 7284142.0,
      "step": 3520
    },
    {
      "entropy": 5.621256923675537,
      "epoch": 3.201726487960018,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004328949364568504,
      "loss": 5.0954,
      "mean_token_accuracy": 0.19565156549215318,
      "num_tokens": 7294228.0,
      "step": 3525
    },
    {
      "entropy": 5.573927927017212,
      "epoch": 3.206269877328487,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004326429524281975,
      "loss": 5.1195,
      "mean_token_accuracy": 0.191754150390625,
      "num_tokens": 7304555.0,
      "step": 3530
    },
    {
      "entropy": 5.5959819793701175,
      "epoch": 3.210813266696956,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004323905794312303,
      "loss": 5.2104,
      "mean_token_accuracy": 0.1781022861599922,
      "num_tokens": 7315955.0,
      "step": 3535
    },
    {
      "entropy": 5.5397686004638675,
      "epoch": 3.215356656065425,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043213781808865414,
      "loss": 5.1518,
      "mean_token_accuracy": 0.1881447434425354,
      "num_tokens": 7326757.0,
      "step": 3540
    },
    {
      "entropy": 5.499069261550903,
      "epoch": 3.2199000454338935,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004318846690241324,
      "loss": 5.13,
      "mean_token_accuracy": 0.19944306910037995,
      "num_tokens": 7335969.0,
      "step": 3545
    },
    {
      "entropy": 5.553613758087158,
      "epoch": 3.2244434348023625,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043163113286228533,
      "loss": 5.1173,
      "mean_token_accuracy": 0.19021477550268173,
      "num_tokens": 7345824.0,
      "step": 3550
    },
    {
      "entropy": 5.624339914321899,
      "epoch": 3.2289868241708315,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004313772102286882,
      "loss": 5.21,
      "mean_token_accuracy": 0.18719030320644378,
      "num_tokens": 7356292.0,
      "step": 3555
    },
    {
      "entropy": 5.562908792495728,
      "epoch": 3.2335302135393005,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004311229017498698,
      "loss": 5.1838,
      "mean_token_accuracy": 0.1888285219669342,
      "num_tokens": 7366809.0,
      "step": 3560
    },
    {
      "entropy": 5.5770941257476805,
      "epoch": 3.238073602907769,
      "grad_norm": 1.15625,
      "learning_rate": 0.000430868208053311,
      "loss": 5.1184,
      "mean_token_accuracy": 0.19796868562698364,
      "num_tokens": 7376076.0,
      "step": 3565
    },
    {
      "entropy": 5.539325284957886,
      "epoch": 3.242616992276238,
      "grad_norm": 1.015625,
      "learning_rate": 0.00043061312976744326,
      "loss": 5.1363,
      "mean_token_accuracy": 0.18888795524835586,
      "num_tokens": 7386338.0,
      "step": 3570
    },
    {
      "entropy": 5.5191186428070065,
      "epoch": 3.247160381644707,
      "grad_norm": 1.125,
      "learning_rate": 0.0004303576675216469,
      "loss": 5.1058,
      "mean_token_accuracy": 0.1901090145111084,
      "num_tokens": 7395792.0,
      "step": 3575
    },
    {
      "entropy": 5.534206247329712,
      "epoch": 3.251703771013176,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00043010182194624957,
      "loss": 5.1388,
      "mean_token_accuracy": 0.1897549256682396,
      "num_tokens": 7405765.0,
      "step": 3580
    },
    {
      "entropy": 5.518572235107422,
      "epoch": 3.2562471603816445,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004298455936725248,
      "loss": 5.1837,
      "mean_token_accuracy": 0.18658362030982972,
      "num_tokens": 7415213.0,
      "step": 3585
    },
    {
      "entropy": 5.646478748321533,
      "epoch": 3.2607905497501135,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004295889833326903,
      "loss": 5.1625,
      "mean_token_accuracy": 0.18602795898914337,
      "num_tokens": 7425722.0,
      "step": 3590
    },
    {
      "entropy": 5.628579378128052,
      "epoch": 3.2653339391185825,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00042933199155990675,
      "loss": 5.2156,
      "mean_token_accuracy": 0.19165341556072235,
      "num_tokens": 7436240.0,
      "step": 3595
    },
    {
      "entropy": 5.516749715805053,
      "epoch": 3.2698773284870515,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004290746189882757,
      "loss": 5.1655,
      "mean_token_accuracy": 0.18851610124111176,
      "num_tokens": 7447878.0,
      "step": 3600
    },
    {
      "entropy": 5.58643102645874,
      "epoch": 3.27442071785552,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042881686625283844,
      "loss": 5.16,
      "mean_token_accuracy": 0.19029032588005065,
      "num_tokens": 7458635.0,
      "step": 3605
    },
    {
      "entropy": 5.596550846099854,
      "epoch": 3.278964107223989,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004285587339895742,
      "loss": 5.1862,
      "mean_token_accuracy": 0.1820325121283531,
      "num_tokens": 7468740.0,
      "step": 3610
    },
    {
      "entropy": 5.663819265365601,
      "epoch": 3.283507496592458,
      "grad_norm": 1.03125,
      "learning_rate": 0.00042830022283539863,
      "loss": 5.2418,
      "mean_token_accuracy": 0.18801323175430298,
      "num_tokens": 7480977.0,
      "step": 3615
    },
    {
      "entropy": 5.554131126403808,
      "epoch": 3.288050885960927,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004280413334281623,
      "loss": 5.2345,
      "mean_token_accuracy": 0.18782038539648055,
      "num_tokens": 7491758.0,
      "step": 3620
    },
    {
      "entropy": 5.631022787094116,
      "epoch": 3.292594275329396,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042778206640664915,
      "loss": 5.1859,
      "mean_token_accuracy": 0.1864585369825363,
      "num_tokens": 7500486.0,
      "step": 3625
    },
    {
      "entropy": 5.494282627105713,
      "epoch": 3.2971376646978645,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004275224224105748,
      "loss": 5.1563,
      "mean_token_accuracy": 0.19796231836080552,
      "num_tokens": 7510371.0,
      "step": 3630
    },
    {
      "entropy": 5.585437297821045,
      "epoch": 3.3016810540663335,
      "grad_norm": 1.03125,
      "learning_rate": 0.00042726240208058484,
      "loss": 5.2018,
      "mean_token_accuracy": 0.19522295743227006,
      "num_tokens": 7520697.0,
      "step": 3635
    },
    {
      "entropy": 5.531649971008301,
      "epoch": 3.3062244434348025,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004270020060582538,
      "loss": 5.1262,
      "mean_token_accuracy": 0.18982232958078385,
      "num_tokens": 7530757.0,
      "step": 3640
    },
    {
      "entropy": 5.5757252216339115,
      "epoch": 3.310767832803271,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00042674123498608285,
      "loss": 5.1564,
      "mean_token_accuracy": 0.18989616334438325,
      "num_tokens": 7541810.0,
      "step": 3645
    },
    {
      "entropy": 5.610417318344116,
      "epoch": 3.31531122217174,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00042648008950749877,
      "loss": 5.344,
      "mean_token_accuracy": 0.17718006819486617,
      "num_tokens": 7552754.0,
      "step": 3650
    },
    {
      "entropy": 5.612136316299439,
      "epoch": 3.319854611540209,
      "grad_norm": 1.140625,
      "learning_rate": 0.000426218570266852,
      "loss": 5.1311,
      "mean_token_accuracy": 0.1953836351633072,
      "num_tokens": 7562906.0,
      "step": 3655
    },
    {
      "entropy": 5.476711845397949,
      "epoch": 3.324398000908678,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00042595667790941536,
      "loss": 5.1415,
      "mean_token_accuracy": 0.19171063899993895,
      "num_tokens": 7574149.0,
      "step": 3660
    },
    {
      "entropy": 5.549911260604858,
      "epoch": 3.328941390277147,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00042569441308138204,
      "loss": 5.1318,
      "mean_token_accuracy": 0.1959947481751442,
      "num_tokens": 7584610.0,
      "step": 3665
    },
    {
      "entropy": 5.531237411499023,
      "epoch": 3.3334847796456155,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004254317764298645,
      "loss": 5.1007,
      "mean_token_accuracy": 0.1887877479195595,
      "num_tokens": 7594954.0,
      "step": 3670
    },
    {
      "entropy": 5.601019716262817,
      "epoch": 3.3380281690140845,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004251687686028927,
      "loss": 5.2096,
      "mean_token_accuracy": 0.18012717813253404,
      "num_tokens": 7606069.0,
      "step": 3675
    },
    {
      "entropy": 5.489976835250855,
      "epoch": 3.3425715583825535,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042490539024941216,
      "loss": 5.0582,
      "mean_token_accuracy": 0.19151556342840195,
      "num_tokens": 7615671.0,
      "step": 3680
    },
    {
      "entropy": 5.5457679271698,
      "epoch": 3.347114947751022,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042464164201928286,
      "loss": 5.2538,
      "mean_token_accuracy": 0.18164459019899368,
      "num_tokens": 7626740.0,
      "step": 3685
    },
    {
      "entropy": 5.580724477767944,
      "epoch": 3.351658337119491,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042437752456327734,
      "loss": 5.1501,
      "mean_token_accuracy": 0.1935126468539238,
      "num_tokens": 7636459.0,
      "step": 3690
    },
    {
      "entropy": 5.619305562973023,
      "epoch": 3.35620172648796,
      "grad_norm": 1.015625,
      "learning_rate": 0.00042411303853307907,
      "loss": 5.1354,
      "mean_token_accuracy": 0.1886074110865593,
      "num_tokens": 7646974.0,
      "step": 3695
    },
    {
      "entropy": 5.4681860446929935,
      "epoch": 3.360745115856429,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004238481845812811,
      "loss": 5.0903,
      "mean_token_accuracy": 0.19271185696125032,
      "num_tokens": 7657372.0,
      "step": 3700
    },
    {
      "entropy": 5.501412963867187,
      "epoch": 3.365288505224898,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00042358296336138436,
      "loss": 5.1348,
      "mean_token_accuracy": 0.189212603867054,
      "num_tokens": 7668972.0,
      "step": 3705
    },
    {
      "entropy": 5.584800386428833,
      "epoch": 3.3698318945933665,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004233173755277956,
      "loss": 5.1421,
      "mean_token_accuracy": 0.20665653944015502,
      "num_tokens": 7679742.0,
      "step": 3710
    },
    {
      "entropy": 5.559856414794922,
      "epoch": 3.3743752839618355,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00042305142173582657,
      "loss": 5.1867,
      "mean_token_accuracy": 0.19379900693893432,
      "num_tokens": 7690759.0,
      "step": 3715
    },
    {
      "entropy": 5.545025634765625,
      "epoch": 3.3789186733303045,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042278510264169165,
      "loss": 5.1391,
      "mean_token_accuracy": 0.19267158657312394,
      "num_tokens": 7700482.0,
      "step": 3720
    },
    {
      "entropy": 5.5073751449584964,
      "epoch": 3.3834620626987735,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042251841890250696,
      "loss": 5.1209,
      "mean_token_accuracy": 0.19036568254232406,
      "num_tokens": 7710414.0,
      "step": 3725
    },
    {
      "entropy": 5.601136684417725,
      "epoch": 3.388005452067242,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000422251371176288,
      "loss": 5.156,
      "mean_token_accuracy": 0.1919710785150528,
      "num_tokens": 7721214.0,
      "step": 3730
    },
    {
      "entropy": 5.5292555809021,
      "epoch": 3.392548841435711,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00042198396012194854,
      "loss": 5.1168,
      "mean_token_accuracy": 0.19231873750686646,
      "num_tokens": 7732551.0,
      "step": 3735
    },
    {
      "entropy": 5.571150875091552,
      "epoch": 3.39709223080418,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00042171618639929865,
      "loss": 5.2091,
      "mean_token_accuracy": 0.191018545627594,
      "num_tokens": 7742266.0,
      "step": 3740
    },
    {
      "entropy": 5.495872163772583,
      "epoch": 3.401635620172649,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042144805066904345,
      "loss": 5.0588,
      "mean_token_accuracy": 0.18994052112102508,
      "num_tokens": 7753152.0,
      "step": 3745
    },
    {
      "entropy": 5.516603899002075,
      "epoch": 3.4061790095411175,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00042117955359278125,
      "loss": 5.1685,
      "mean_token_accuracy": 0.18809825032949448,
      "num_tokens": 7764420.0,
      "step": 3750
    },
    {
      "entropy": 5.576537799835205,
      "epoch": 3.4107223989095865,
      "grad_norm": 0.9375,
      "learning_rate": 0.00042091069583300177,
      "loss": 5.1298,
      "mean_token_accuracy": 0.19266747385263444,
      "num_tokens": 7776044.0,
      "step": 3755
    },
    {
      "entropy": 5.5177184581756595,
      "epoch": 3.4152657882780555,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004206414780530849,
      "loss": 5.1603,
      "mean_token_accuracy": 0.19181600511074065,
      "num_tokens": 7786737.0,
      "step": 3760
    },
    {
      "entropy": 5.506136226654053,
      "epoch": 3.4198091776465245,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004203719009172987,
      "loss": 5.0854,
      "mean_token_accuracy": 0.19453130811452865,
      "num_tokens": 7796974.0,
      "step": 3765
    },
    {
      "entropy": 5.5970865249633786,
      "epoch": 3.424352567014993,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000420101965090798,
      "loss": 5.2717,
      "mean_token_accuracy": 0.18344012647867203,
      "num_tokens": 7808221.0,
      "step": 3770
    },
    {
      "entropy": 5.500562858581543,
      "epoch": 3.428895956383462,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004198316712396226,
      "loss": 5.0898,
      "mean_token_accuracy": 0.19185245484113694,
      "num_tokens": 7818422.0,
      "step": 3775
    },
    {
      "entropy": 5.540801000595093,
      "epoch": 3.433439345751931,
      "grad_norm": 0.984375,
      "learning_rate": 0.00041956102003069573,
      "loss": 5.1114,
      "mean_token_accuracy": 0.19109012335538864,
      "num_tokens": 7828313.0,
      "step": 3780
    },
    {
      "entropy": 5.5394628047943115,
      "epoch": 3.4379827351204,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004192900121318223,
      "loss": 5.1337,
      "mean_token_accuracy": 0.19941307753324508,
      "num_tokens": 7839493.0,
      "step": 3785
    },
    {
      "entropy": 5.543460845947266,
      "epoch": 3.4425261244888685,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004190186482116874,
      "loss": 5.134,
      "mean_token_accuracy": 0.18883001506328584,
      "num_tokens": 7849796.0,
      "step": 3790
    },
    {
      "entropy": 5.5130970001220705,
      "epoch": 3.4470695138573375,
      "grad_norm": 1.0,
      "learning_rate": 0.0004187469289398545,
      "loss": 5.1957,
      "mean_token_accuracy": 0.18865694403648375,
      "num_tokens": 7860865.0,
      "step": 3795
    },
    {
      "entropy": 5.49743332862854,
      "epoch": 3.4516129032258065,
      "grad_norm": 1.03125,
      "learning_rate": 0.00041847485498676385,
      "loss": 5.1403,
      "mean_token_accuracy": 0.19718256145715712,
      "num_tokens": 7871144.0,
      "step": 3800
    },
    {
      "entropy": 5.522391748428345,
      "epoch": 3.4561562925942755,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041820242702373094,
      "loss": 5.1888,
      "mean_token_accuracy": 0.1942734718322754,
      "num_tokens": 7880916.0,
      "step": 3805
    },
    {
      "entropy": 5.484518671035767,
      "epoch": 3.460699681962744,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004179296457229448,
      "loss": 5.1211,
      "mean_token_accuracy": 0.1911142200231552,
      "num_tokens": 7891813.0,
      "step": 3810
    },
    {
      "entropy": 5.585445737838745,
      "epoch": 3.465243071331213,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004176565117574659,
      "loss": 5.173,
      "mean_token_accuracy": 0.1877262234687805,
      "num_tokens": 7902697.0,
      "step": 3815
    },
    {
      "entropy": 5.537561893463135,
      "epoch": 3.469786460699682,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004173830258012255,
      "loss": 5.1464,
      "mean_token_accuracy": 0.1930139735341072,
      "num_tokens": 7912777.0,
      "step": 3820
    },
    {
      "entropy": 5.553113269805908,
      "epoch": 3.474329850068151,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004171091885290226,
      "loss": 5.1873,
      "mean_token_accuracy": 0.19049507528543472,
      "num_tokens": 7923255.0,
      "step": 3825
    },
    {
      "entropy": 5.537462568283081,
      "epoch": 3.4788732394366195,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004168350006165238,
      "loss": 5.0658,
      "mean_token_accuracy": 0.19168077856302262,
      "num_tokens": 7934268.0,
      "step": 3830
    },
    {
      "entropy": 5.432240676879883,
      "epoch": 3.4834166288050885,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041656046274026034,
      "loss": 5.0676,
      "mean_token_accuracy": 0.19443341344594955,
      "num_tokens": 7944526.0,
      "step": 3835
    },
    {
      "entropy": 5.554276704788208,
      "epoch": 3.4879600181735575,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004162855755776272,
      "loss": 5.1781,
      "mean_token_accuracy": 0.18912502825260163,
      "num_tokens": 7954512.0,
      "step": 3840
    },
    {
      "entropy": 5.529981517791748,
      "epoch": 3.4925034075420265,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00041601033980688106,
      "loss": 5.196,
      "mean_token_accuracy": 0.18088309913873674,
      "num_tokens": 7965703.0,
      "step": 3845
    },
    {
      "entropy": 5.5635253429412845,
      "epoch": 3.497046796910495,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004157347561071389,
      "loss": 5.1868,
      "mean_token_accuracy": 0.18955022394657134,
      "num_tokens": 7977822.0,
      "step": 3850
    },
    {
      "entropy": 5.510582113265992,
      "epoch": 3.501590186278964,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00041545882515837604,
      "loss": 5.0973,
      "mean_token_accuracy": 0.18894753307104112,
      "num_tokens": 7988318.0,
      "step": 3855
    },
    {
      "entropy": 5.550647926330567,
      "epoch": 3.506133575647433,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00041518254764142477,
      "loss": 5.0629,
      "mean_token_accuracy": 0.19622776061296462,
      "num_tokens": 7998076.0,
      "step": 3860
    },
    {
      "entropy": 5.529185485839844,
      "epoch": 3.510676965015902,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00041490592423797226,
      "loss": 5.1715,
      "mean_token_accuracy": 0.19466058611869813,
      "num_tokens": 8007913.0,
      "step": 3865
    },
    {
      "entropy": 5.489879989624024,
      "epoch": 3.515220354384371,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004146289556305594,
      "loss": 5.1338,
      "mean_token_accuracy": 0.18910700529813768,
      "num_tokens": 8018340.0,
      "step": 3870
    },
    {
      "entropy": 5.61398458480835,
      "epoch": 3.5197637437528395,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004143516425025786,
      "loss": 5.2044,
      "mean_token_accuracy": 0.18639347851276397,
      "num_tokens": 8028312.0,
      "step": 3875
    },
    {
      "entropy": 5.59273042678833,
      "epoch": 3.5243071331213085,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004140739855382724,
      "loss": 5.1965,
      "mean_token_accuracy": 0.1921403169631958,
      "num_tokens": 8038580.0,
      "step": 3880
    },
    {
      "entropy": 5.574572420120239,
      "epoch": 3.5288505224897775,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041379598542273197,
      "loss": 5.1486,
      "mean_token_accuracy": 0.18995873779058456,
      "num_tokens": 8048365.0,
      "step": 3885
    },
    {
      "entropy": 5.552021026611328,
      "epoch": 3.533393911858246,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00041351764284189464,
      "loss": 5.2077,
      "mean_token_accuracy": 0.1823919713497162,
      "num_tokens": 8059638.0,
      "step": 3890
    },
    {
      "entropy": 5.579764318466187,
      "epoch": 3.537937301226715,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004132389584825433,
      "loss": 5.2502,
      "mean_token_accuracy": 0.18523784279823302,
      "num_tokens": 8070993.0,
      "step": 3895
    },
    {
      "entropy": 5.5283417224884035,
      "epoch": 3.542480690595184,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041295993303230383,
      "loss": 5.0907,
      "mean_token_accuracy": 0.1978906959295273,
      "num_tokens": 8081203.0,
      "step": 3900
    },
    {
      "entropy": 5.454664659500122,
      "epoch": 3.547024079963653,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004126805671796438,
      "loss": 5.0981,
      "mean_token_accuracy": 0.1887076824903488,
      "num_tokens": 8091564.0,
      "step": 3905
    },
    {
      "entropy": 5.5413196086883545,
      "epoch": 3.551567469332122,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004124008616138707,
      "loss": 5.087,
      "mean_token_accuracy": 0.19312319606542588,
      "num_tokens": 8100675.0,
      "step": 3910
    },
    {
      "entropy": 5.520606803894043,
      "epoch": 3.5561108587005905,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041212081702513015,
      "loss": 5.1482,
      "mean_token_accuracy": 0.1945698454976082,
      "num_tokens": 8110605.0,
      "step": 3915
    },
    {
      "entropy": 5.5288787364959715,
      "epoch": 3.5606542480690595,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004118404341044044,
      "loss": 5.1986,
      "mean_token_accuracy": 0.18797775804996492,
      "num_tokens": 8120910.0,
      "step": 3920
    },
    {
      "entropy": 5.55887804031372,
      "epoch": 3.5651976374375285,
      "grad_norm": 1.03125,
      "learning_rate": 0.00041155971354351034,
      "loss": 5.1492,
      "mean_token_accuracy": 0.19329776912927626,
      "num_tokens": 8131827.0,
      "step": 3925
    },
    {
      "entropy": 5.540145587921143,
      "epoch": 3.569741026805997,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041127865603509815,
      "loss": 5.1235,
      "mean_token_accuracy": 0.1942894384264946,
      "num_tokens": 8141785.0,
      "step": 3930
    },
    {
      "entropy": 5.574447584152222,
      "epoch": 3.574284416174466,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004109972622726493,
      "loss": 5.1515,
      "mean_token_accuracy": 0.1875311553478241,
      "num_tokens": 8151732.0,
      "step": 3935
    },
    {
      "entropy": 5.563373231887818,
      "epoch": 3.578827805542935,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00041071553295047485,
      "loss": 5.2329,
      "mean_token_accuracy": 0.18431088477373123,
      "num_tokens": 8162334.0,
      "step": 3940
    },
    {
      "entropy": 5.535441780090332,
      "epoch": 3.583371194911404,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004104334687637139,
      "loss": 5.1146,
      "mean_token_accuracy": 0.19262330383062362,
      "num_tokens": 8172810.0,
      "step": 3945
    },
    {
      "entropy": 5.4998149394989015,
      "epoch": 3.587914584279873,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00041015107040833187,
      "loss": 5.1454,
      "mean_token_accuracy": 0.1860748738050461,
      "num_tokens": 8183240.0,
      "step": 3950
    },
    {
      "entropy": 5.54986252784729,
      "epoch": 3.5924579736483415,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004098683385811185,
      "loss": 5.103,
      "mean_token_accuracy": 0.19000099450349808,
      "num_tokens": 8193824.0,
      "step": 3955
    },
    {
      "entropy": 5.496000814437866,
      "epoch": 3.5970013630168105,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040958527397968666,
      "loss": 5.0843,
      "mean_token_accuracy": 0.19927169978618622,
      "num_tokens": 8203792.0,
      "step": 3960
    },
    {
      "entropy": 5.54337158203125,
      "epoch": 3.6015447523852795,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004093018773024699,
      "loss": 5.2784,
      "mean_token_accuracy": 0.18759161233901978,
      "num_tokens": 8214436.0,
      "step": 3965
    },
    {
      "entropy": 5.576527786254883,
      "epoch": 3.606088141753748,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004090181492487215,
      "loss": 5.1298,
      "mean_token_accuracy": 0.1908451110124588,
      "num_tokens": 8224518.0,
      "step": 3970
    },
    {
      "entropy": 5.563905239105225,
      "epoch": 3.610631531122217,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040873409051851223,
      "loss": 5.2793,
      "mean_token_accuracy": 0.18164435774087906,
      "num_tokens": 8235234.0,
      "step": 3975
    },
    {
      "entropy": 5.530511856079102,
      "epoch": 3.615174920490686,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004084497018127286,
      "loss": 5.1545,
      "mean_token_accuracy": 0.18730711191892624,
      "num_tokens": 8246612.0,
      "step": 3980
    },
    {
      "entropy": 5.554295444488526,
      "epoch": 3.619718309859155,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004081649838330717,
      "loss": 5.1303,
      "mean_token_accuracy": 0.19477790594100952,
      "num_tokens": 8256720.0,
      "step": 3985
    },
    {
      "entropy": 5.533069324493408,
      "epoch": 3.624261699227624,
      "grad_norm": 0.953125,
      "learning_rate": 0.00040787993728205464,
      "loss": 5.1758,
      "mean_token_accuracy": 0.1920352280139923,
      "num_tokens": 8267666.0,
      "step": 3990
    },
    {
      "entropy": 5.5814110279083256,
      "epoch": 3.6288050885960925,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040759456286300154,
      "loss": 5.1561,
      "mean_token_accuracy": 0.1906946375966072,
      "num_tokens": 8277882.0,
      "step": 3995
    },
    {
      "entropy": 5.49630331993103,
      "epoch": 3.6333484779645615,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040730886128004544,
      "loss": 5.0717,
      "mean_token_accuracy": 0.19409287124872207,
      "num_tokens": 8286968.0,
      "step": 4000
    },
    {
      "epoch": 3.6333484779645615,
      "eval_entropy": 5.366523528653522,
      "eval_loss": 5.737530708312988,
      "eval_mean_token_accuracy": 0.167983505424372,
      "eval_num_tokens": 8286968.0,
      "eval_runtime": 2.0103,
      "eval_samples_per_second": 1708.737,
      "eval_steps_per_second": 213.903,
      "step": 4000
    },
    {
      "entropy": 5.52182788848877,
      "epoch": 3.6378918673330305,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004070228332381266,
      "loss": 5.1491,
      "mean_token_accuracy": 0.1925757884979248,
      "num_tokens": 8297154.0,
      "step": 4005
    },
    {
      "entropy": 5.524565696716309,
      "epoch": 3.6424352567014995,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004067364794429908,
      "loss": 5.1243,
      "mean_token_accuracy": 0.18758785873651504,
      "num_tokens": 8307954.0,
      "step": 4010
    },
    {
      "entropy": 5.456177711486816,
      "epoch": 3.646978646069968,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004064498006011874,
      "loss": 5.119,
      "mean_token_accuracy": 0.18877227902412413,
      "num_tokens": 8317714.0,
      "step": 4015
    },
    {
      "entropy": 5.471064186096191,
      "epoch": 3.651522035438437,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00040616279742006813,
      "loss": 5.1475,
      "mean_token_accuracy": 0.19805022776126863,
      "num_tokens": 8327963.0,
      "step": 4020
    },
    {
      "entropy": 5.52384934425354,
      "epoch": 3.656065424806906,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00040587547060778466,
      "loss": 5.0653,
      "mean_token_accuracy": 0.19838137030601502,
      "num_tokens": 8338190.0,
      "step": 4025
    },
    {
      "entropy": 5.548328828811646,
      "epoch": 3.660608814175375,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004055878208732874,
      "loss": 5.2136,
      "mean_token_accuracy": 0.18524187058210373,
      "num_tokens": 8348564.0,
      "step": 4030
    },
    {
      "entropy": 5.502794885635376,
      "epoch": 3.665152203543844,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00040529984892632363,
      "loss": 5.1363,
      "mean_token_accuracy": 0.19377148598432542,
      "num_tokens": 8359463.0,
      "step": 4035
    },
    {
      "entropy": 5.570143795013427,
      "epoch": 3.6696955929123125,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004050115554774353,
      "loss": 5.1795,
      "mean_token_accuracy": 0.18666360080242156,
      "num_tokens": 8369676.0,
      "step": 4040
    },
    {
      "entropy": 5.533338785171509,
      "epoch": 3.6742389822807815,
      "grad_norm": 1.0,
      "learning_rate": 0.0004047229412379578,
      "loss": 5.0961,
      "mean_token_accuracy": 0.20030711591243744,
      "num_tokens": 8379579.0,
      "step": 4045
    },
    {
      "entropy": 5.492454290390015,
      "epoch": 3.6787823716492505,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004044340069200183,
      "loss": 5.1595,
      "mean_token_accuracy": 0.1914926901459694,
      "num_tokens": 8390164.0,
      "step": 4050
    },
    {
      "entropy": 5.478480815887451,
      "epoch": 3.683325761017719,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040414475323653326,
      "loss": 5.0681,
      "mean_token_accuracy": 0.1989881858229637,
      "num_tokens": 8399220.0,
      "step": 4055
    },
    {
      "entropy": 5.561425495147705,
      "epoch": 3.687869150386188,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004038551809012076,
      "loss": 5.1386,
      "mean_token_accuracy": 0.1941610500216484,
      "num_tokens": 8409488.0,
      "step": 4060
    },
    {
      "entropy": 5.535262680053711,
      "epoch": 3.692412539754657,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004035652906285322,
      "loss": 5.0841,
      "mean_token_accuracy": 0.19989518225193023,
      "num_tokens": 8418817.0,
      "step": 4065
    },
    {
      "entropy": 5.4969947814941404,
      "epoch": 3.696955929123126,
      "grad_norm": 1.015625,
      "learning_rate": 0.00040327508313378244,
      "loss": 5.2075,
      "mean_token_accuracy": 0.18858725428581238,
      "num_tokens": 8428742.0,
      "step": 4070
    },
    {
      "entropy": 5.502942323684692,
      "epoch": 3.701499318491595,
      "grad_norm": 1.125,
      "learning_rate": 0.0004029845591330164,
      "loss": 5.0796,
      "mean_token_accuracy": 0.20013698935508728,
      "num_tokens": 8438016.0,
      "step": 4075
    },
    {
      "entropy": 5.561799764633179,
      "epoch": 3.7060427078600635,
      "grad_norm": 1.0,
      "learning_rate": 0.00040269371934307336,
      "loss": 5.1885,
      "mean_token_accuracy": 0.19165659993886947,
      "num_tokens": 8448428.0,
      "step": 4080
    },
    {
      "entropy": 5.4906106948852536,
      "epoch": 3.7105860972285325,
      "grad_norm": 1.03125,
      "learning_rate": 0.00040240256448157146,
      "loss": 5.0346,
      "mean_token_accuracy": 0.19949788153171538,
      "num_tokens": 8458444.0,
      "step": 4085
    },
    {
      "entropy": 5.4608070850372314,
      "epoch": 3.7151294865970015,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004021110952669065,
      "loss": 5.1778,
      "mean_token_accuracy": 0.19027193188667296,
      "num_tokens": 8467721.0,
      "step": 4090
    },
    {
      "entropy": 5.444411182403565,
      "epoch": 3.71967287596547,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004018193124182497,
      "loss": 5.0971,
      "mean_token_accuracy": 0.19552406966686248,
      "num_tokens": 8477484.0,
      "step": 4095
    },
    {
      "entropy": 5.535136079788208,
      "epoch": 3.724216265333939,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004015272166555462,
      "loss": 5.1081,
      "mean_token_accuracy": 0.19443877935409545,
      "num_tokens": 8486902.0,
      "step": 4100
    },
    {
      "entropy": 5.491853332519531,
      "epoch": 3.728759654702408,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004012348086995132,
      "loss": 5.0687,
      "mean_token_accuracy": 0.19258963763713838,
      "num_tokens": 8497626.0,
      "step": 4105
    },
    {
      "entropy": 5.496549987792969,
      "epoch": 3.733303044070877,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004009420892716385,
      "loss": 5.1802,
      "mean_token_accuracy": 0.18394527435302735,
      "num_tokens": 8507518.0,
      "step": 4110
    },
    {
      "entropy": 5.501980113983154,
      "epoch": 3.737846433439346,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004006490590941781,
      "loss": 5.1408,
      "mean_token_accuracy": 0.1926366701722145,
      "num_tokens": 8517835.0,
      "step": 4115
    },
    {
      "entropy": 5.581077623367309,
      "epoch": 3.7423898228078145,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004003557188901547,
      "loss": 5.1884,
      "mean_token_accuracy": 0.18677228540182114,
      "num_tokens": 8527552.0,
      "step": 4120
    },
    {
      "entropy": 5.554247140884399,
      "epoch": 3.7469332121762835,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004000620693833562,
      "loss": 5.1705,
      "mean_token_accuracy": 0.19123418778181075,
      "num_tokens": 8538082.0,
      "step": 4125
    },
    {
      "entropy": 5.482239723205566,
      "epoch": 3.7514766015447525,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003997681112983335,
      "loss": 5.1108,
      "mean_token_accuracy": 0.19062484800815582,
      "num_tokens": 8548558.0,
      "step": 4130
    },
    {
      "entropy": 5.463817310333252,
      "epoch": 3.756019990913221,
      "grad_norm": 1.015625,
      "learning_rate": 0.00039947384536039905,
      "loss": 5.1191,
      "mean_token_accuracy": 0.19476200342178346,
      "num_tokens": 8559406.0,
      "step": 4135
    },
    {
      "entropy": 5.536861801147461,
      "epoch": 3.76056338028169,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00039917927229562474,
      "loss": 5.1518,
      "mean_token_accuracy": 0.18347567021846772,
      "num_tokens": 8569519.0,
      "step": 4140
    },
    {
      "entropy": 5.470212841033936,
      "epoch": 3.765106769650159,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039888439283084035,
      "loss": 5.1716,
      "mean_token_accuracy": 0.18914507031440736,
      "num_tokens": 8579658.0,
      "step": 4145
    },
    {
      "entropy": 5.47109317779541,
      "epoch": 3.769650159018628,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00039858920769363164,
      "loss": 5.0862,
      "mean_token_accuracy": 0.19758331775665283,
      "num_tokens": 8589681.0,
      "step": 4150
    },
    {
      "entropy": 5.552435827255249,
      "epoch": 3.774193548387097,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003982937176123386,
      "loss": 5.1617,
      "mean_token_accuracy": 0.19099028259515763,
      "num_tokens": 8600310.0,
      "step": 4155
    },
    {
      "entropy": 5.514196252822876,
      "epoch": 3.7787369377555655,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039799792331605364,
      "loss": 5.1631,
      "mean_token_accuracy": 0.19098785817623137,
      "num_tokens": 8610920.0,
      "step": 4160
    },
    {
      "entropy": 5.460143089294434,
      "epoch": 3.7832803271240345,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039770182553461965,
      "loss": 5.0909,
      "mean_token_accuracy": 0.19817202240228654,
      "num_tokens": 8620536.0,
      "step": 4165
    },
    {
      "entropy": 5.512439584732055,
      "epoch": 3.7878237164925035,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039740542499862866,
      "loss": 5.1863,
      "mean_token_accuracy": 0.1910782739520073,
      "num_tokens": 8631647.0,
      "step": 4170
    },
    {
      "entropy": 5.543681955337524,
      "epoch": 3.792367105860972,
      "grad_norm": 0.90625,
      "learning_rate": 0.0003971087224394194,
      "loss": 5.1511,
      "mean_token_accuracy": 0.18959263563156128,
      "num_tokens": 8642659.0,
      "step": 4175
    },
    {
      "entropy": 5.51680908203125,
      "epoch": 3.796910495229441,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003968117185890761,
      "loss": 5.1398,
      "mean_token_accuracy": 0.2002093091607094,
      "num_tokens": 8651775.0,
      "step": 4180
    },
    {
      "entropy": 5.495999479293824,
      "epoch": 3.80145388459791,
      "grad_norm": 1.015625,
      "learning_rate": 0.00039651441418042606,
      "loss": 5.1075,
      "mean_token_accuracy": 0.18484094887971877,
      "num_tokens": 8662520.0,
      "step": 4185
    },
    {
      "entropy": 5.547680759429932,
      "epoch": 3.805997273966379,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003962168099470384,
      "loss": 5.1428,
      "mean_token_accuracy": 0.19475969523191453,
      "num_tokens": 8672630.0,
      "step": 4190
    },
    {
      "entropy": 5.550949287414551,
      "epoch": 3.810540663334848,
      "grad_norm": 1.046875,
      "learning_rate": 0.000395918906623222,
      "loss": 5.2396,
      "mean_token_accuracy": 0.17787621170282364,
      "num_tokens": 8683348.0,
      "step": 4195
    },
    {
      "entropy": 5.47524356842041,
      "epoch": 3.8150840527033165,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003956207049440238,
      "loss": 5.0696,
      "mean_token_accuracy": 0.19471167922019958,
      "num_tokens": 8693423.0,
      "step": 4200
    },
    {
      "entropy": 5.549236488342285,
      "epoch": 3.8196274420717855,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003953222056452265,
      "loss": 5.1367,
      "mean_token_accuracy": 0.19452614784240724,
      "num_tokens": 8703766.0,
      "step": 4205
    },
    {
      "entropy": 5.579219770431519,
      "epoch": 3.8241708314402545,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003950234094633479,
      "loss": 5.2482,
      "mean_token_accuracy": 0.1882735773921013,
      "num_tokens": 8715330.0,
      "step": 4210
    },
    {
      "entropy": 5.516354990005493,
      "epoch": 3.828714220808723,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00039472431713563746,
      "loss": 5.1172,
      "mean_token_accuracy": 0.19752700477838517,
      "num_tokens": 8725732.0,
      "step": 4215
    },
    {
      "entropy": 5.595075464248657,
      "epoch": 3.833257610177192,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000394424929400076,
      "loss": 5.099,
      "mean_token_accuracy": 0.1987996757030487,
      "num_tokens": 8736377.0,
      "step": 4220
    },
    {
      "entropy": 5.521892261505127,
      "epoch": 3.837800999545661,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003941252469953729,
      "loss": 5.1336,
      "mean_token_accuracy": 0.18955939561128615,
      "num_tokens": 8746991.0,
      "step": 4225
    },
    {
      "entropy": 5.473519945144654,
      "epoch": 3.84234438891413,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003938252706609647,
      "loss": 5.1368,
      "mean_token_accuracy": 0.19018320888280868,
      "num_tokens": 8756920.0,
      "step": 4230
    },
    {
      "entropy": 5.518662929534912,
      "epoch": 3.846887778282599,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003935250011370132,
      "loss": 5.2034,
      "mean_token_accuracy": 0.18204569369554519,
      "num_tokens": 8768617.0,
      "step": 4235
    },
    {
      "entropy": 5.552821969985962,
      "epoch": 3.851431167651068,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0003932244391644037,
      "loss": 5.1556,
      "mean_token_accuracy": 0.19870792776346208,
      "num_tokens": 8778427.0,
      "step": 4240
    },
    {
      "entropy": 5.5150621891021725,
      "epoch": 3.8559745570195365,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003929235854847428,
      "loss": 5.1919,
      "mean_token_accuracy": 0.19484769701957702,
      "num_tokens": 8789093.0,
      "step": 4245
    },
    {
      "entropy": 5.491849899291992,
      "epoch": 3.8605179463880055,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00039262244084035724,
      "loss": 5.1368,
      "mean_token_accuracy": 0.18463968336582184,
      "num_tokens": 8799832.0,
      "step": 4250
    },
    {
      "entropy": 5.54138765335083,
      "epoch": 3.8650613357564745,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00039232100597429143,
      "loss": 5.1533,
      "mean_token_accuracy": 0.19056304693222045,
      "num_tokens": 8810149.0,
      "step": 4255
    },
    {
      "entropy": 5.562223672866821,
      "epoch": 3.869604725124943,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00039201928163030584,
      "loss": 5.167,
      "mean_token_accuracy": 0.19634356051683427,
      "num_tokens": 8820087.0,
      "step": 4260
    },
    {
      "entropy": 5.506262493133545,
      "epoch": 3.874148114493412,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00039171726855287544,
      "loss": 5.2038,
      "mean_token_accuracy": 0.18511039316654204,
      "num_tokens": 8830959.0,
      "step": 4265
    },
    {
      "entropy": 5.567284631729126,
      "epoch": 3.878691503861881,
      "grad_norm": 0.984375,
      "learning_rate": 0.00039141496748718735,
      "loss": 5.1472,
      "mean_token_accuracy": 0.18979677706956863,
      "num_tokens": 8842522.0,
      "step": 4270
    },
    {
      "entropy": 5.456068134307861,
      "epoch": 3.88323489323035,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003911123791791395,
      "loss": 5.0568,
      "mean_token_accuracy": 0.19832214415073396,
      "num_tokens": 8852521.0,
      "step": 4275
    },
    {
      "entropy": 5.581479692459107,
      "epoch": 3.887778282598819,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003908095043753384,
      "loss": 5.2005,
      "mean_token_accuracy": 0.1849861741065979,
      "num_tokens": 8862156.0,
      "step": 4280
    },
    {
      "entropy": 5.484117078781128,
      "epoch": 3.8923216719672875,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00039050634382309753,
      "loss": 5.0636,
      "mean_token_accuracy": 0.19180549681186676,
      "num_tokens": 8872772.0,
      "step": 4285
    },
    {
      "entropy": 5.52656741142273,
      "epoch": 3.8968650613357565,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003902028982704355,
      "loss": 5.1534,
      "mean_token_accuracy": 0.19367601871490478,
      "num_tokens": 8883229.0,
      "step": 4290
    },
    {
      "entropy": 5.505592489242554,
      "epoch": 3.9014084507042255,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003898991684660738,
      "loss": 5.1373,
      "mean_token_accuracy": 0.19029328972101212,
      "num_tokens": 8892403.0,
      "step": 4295
    },
    {
      "entropy": 5.450834465026856,
      "epoch": 3.905951840072694,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003895951551594358,
      "loss": 5.1501,
      "mean_token_accuracy": 0.18638637363910676,
      "num_tokens": 8902617.0,
      "step": 4300
    },
    {
      "entropy": 5.591155481338501,
      "epoch": 3.910495229441163,
      "grad_norm": 1.0625,
      "learning_rate": 0.000389290859100644,
      "loss": 5.1793,
      "mean_token_accuracy": 0.19201241731643676,
      "num_tokens": 8913099.0,
      "step": 4305
    },
    {
      "entropy": 5.469971990585327,
      "epoch": 3.915038618809632,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003889862810405187,
      "loss": 5.0832,
      "mean_token_accuracy": 0.19475504755973816,
      "num_tokens": 8922469.0,
      "step": 4310
    },
    {
      "entropy": 5.491753816604614,
      "epoch": 3.919582008178101,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000388681421730576,
      "loss": 5.1678,
      "mean_token_accuracy": 0.19475577622652054,
      "num_tokens": 8933261.0,
      "step": 4315
    },
    {
      "entropy": 5.534124183654785,
      "epoch": 3.92412539754657,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003883762819230259,
      "loss": 5.1622,
      "mean_token_accuracy": 0.19470106661319733,
      "num_tokens": 8943873.0,
      "step": 4320
    },
    {
      "entropy": 5.508684396743774,
      "epoch": 3.9286687869150385,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00038807086237077057,
      "loss": 5.0587,
      "mean_token_accuracy": 0.20081716030836105,
      "num_tokens": 8953802.0,
      "step": 4325
    },
    {
      "entropy": 5.542525959014893,
      "epoch": 3.9332121762835075,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00038776516382740237,
      "loss": 5.2566,
      "mean_token_accuracy": 0.18533951491117479,
      "num_tokens": 8964309.0,
      "step": 4330
    },
    {
      "entropy": 5.540783882141113,
      "epoch": 3.9377555656519765,
      "grad_norm": 1.171875,
      "learning_rate": 0.00038745918704720205,
      "loss": 5.1192,
      "mean_token_accuracy": 0.19729201942682267,
      "num_tokens": 8973611.0,
      "step": 4335
    },
    {
      "entropy": 5.483542442321777,
      "epoch": 3.942298955020445,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00038715293278513693,
      "loss": 5.1504,
      "mean_token_accuracy": 0.18619137406349182,
      "num_tokens": 8985783.0,
      "step": 4340
    },
    {
      "entropy": 5.516713809967041,
      "epoch": 3.946842344388914,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00038684640179685896,
      "loss": 5.2019,
      "mean_token_accuracy": 0.18765413016080856,
      "num_tokens": 8996736.0,
      "step": 4345
    },
    {
      "entropy": 5.541996002197266,
      "epoch": 3.951385733757383,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003865395948387029,
      "loss": 5.2536,
      "mean_token_accuracy": 0.18431503921747208,
      "num_tokens": 9007635.0,
      "step": 4350
    },
    {
      "entropy": 5.550665187835693,
      "epoch": 3.955929123125852,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00038623251266768437,
      "loss": 5.1076,
      "mean_token_accuracy": 0.18914185464382172,
      "num_tokens": 9017892.0,
      "step": 4355
    },
    {
      "entropy": 5.51224308013916,
      "epoch": 3.960472512494321,
      "grad_norm": 1.03125,
      "learning_rate": 0.00038592515604149815,
      "loss": 5.1531,
      "mean_token_accuracy": 0.19024809151887895,
      "num_tokens": 9027594.0,
      "step": 4360
    },
    {
      "entropy": 5.510853958129883,
      "epoch": 3.9650159018627895,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003856175257185162,
      "loss": 5.1818,
      "mean_token_accuracy": 0.1894442170858383,
      "num_tokens": 9037083.0,
      "step": 4365
    },
    {
      "entropy": 5.539366436004639,
      "epoch": 3.9695592912312585,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00038530962245778564,
      "loss": 5.1427,
      "mean_token_accuracy": 0.19987967610359192,
      "num_tokens": 9047176.0,
      "step": 4370
    },
    {
      "entropy": 5.52049036026001,
      "epoch": 3.9741026805997275,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003850014470190272,
      "loss": 5.0994,
      "mean_token_accuracy": 0.19403926134109498,
      "num_tokens": 9057810.0,
      "step": 4375
    },
    {
      "entropy": 5.57022910118103,
      "epoch": 3.978646069968196,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003846930001626332,
      "loss": 5.2113,
      "mean_token_accuracy": 0.19076910316944123,
      "num_tokens": 9067765.0,
      "step": 4380
    },
    {
      "entropy": 5.5833155632019045,
      "epoch": 3.983189459336665,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003843842826496656,
      "loss": 5.1433,
      "mean_token_accuracy": 0.19173349142074586,
      "num_tokens": 9078796.0,
      "step": 4385
    },
    {
      "entropy": 5.526275014877319,
      "epoch": 3.987732848705134,
      "grad_norm": 1.203125,
      "learning_rate": 0.000384075295241854,
      "loss": 5.1003,
      "mean_token_accuracy": 0.19770094603300095,
      "num_tokens": 9088368.0,
      "step": 4390
    },
    {
      "entropy": 5.446280860900879,
      "epoch": 3.992276238073603,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003837660387015944,
      "loss": 5.0586,
      "mean_token_accuracy": 0.1904596656560898,
      "num_tokens": 9099082.0,
      "step": 4395
    },
    {
      "entropy": 5.5042637348175045,
      "epoch": 3.996819627442072,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0003834565137919463,
      "loss": 5.1916,
      "mean_token_accuracy": 0.19258765280246734,
      "num_tokens": 9110543.0,
      "step": 4400
    },
    {
      "entropy": 5.554924858940972,
      "epoch": 4.000908677873694,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003831467212766318,
      "loss": 5.1196,
      "mean_token_accuracy": 0.1920791053109699,
      "num_tokens": 9119347.0,
      "step": 4405
    },
    {
      "entropy": 5.485638809204102,
      "epoch": 4.005452067242163,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003828366619200331,
      "loss": 4.8071,
      "mean_token_accuracy": 0.21238681972026824,
      "num_tokens": 9129154.0,
      "step": 4410
    },
    {
      "entropy": 5.360383224487305,
      "epoch": 4.009995456610632,
      "grad_norm": 1.0,
      "learning_rate": 0.0003825263364871908,
      "loss": 4.8569,
      "mean_token_accuracy": 0.2095795601606369,
      "num_tokens": 9139877.0,
      "step": 4415
    },
    {
      "entropy": 5.501720714569092,
      "epoch": 4.0145388459791,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003822157457438022,
      "loss": 4.907,
      "mean_token_accuracy": 0.20996578335762023,
      "num_tokens": 9150508.0,
      "step": 4420
    },
    {
      "entropy": 5.473970031738281,
      "epoch": 4.019082235347569,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00038190489045621893,
      "loss": 4.8556,
      "mean_token_accuracy": 0.21571922153234482,
      "num_tokens": 9160403.0,
      "step": 4425
    },
    {
      "entropy": 5.534111642837525,
      "epoch": 4.023625624716038,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0003815937713914455,
      "loss": 4.9039,
      "mean_token_accuracy": 0.21270331144332885,
      "num_tokens": 9171026.0,
      "step": 4430
    },
    {
      "entropy": 5.479460382461548,
      "epoch": 4.028169014084507,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00038128238931713743,
      "loss": 4.8739,
      "mean_token_accuracy": 0.2098907485604286,
      "num_tokens": 9181252.0,
      "step": 4435
    },
    {
      "entropy": 5.548450946807861,
      "epoch": 4.032712403452976,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00038097074500159883,
      "loss": 4.9629,
      "mean_token_accuracy": 0.20318549871444702,
      "num_tokens": 9190803.0,
      "step": 4440
    },
    {
      "entropy": 5.508038902282715,
      "epoch": 4.037255792821445,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00038065883921378123,
      "loss": 4.9175,
      "mean_token_accuracy": 0.2061125874519348,
      "num_tokens": 9201077.0,
      "step": 4445
    },
    {
      "entropy": 5.483087968826294,
      "epoch": 4.041799182189914,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003803466727232811,
      "loss": 4.9517,
      "mean_token_accuracy": 0.20602241903543472,
      "num_tokens": 9211239.0,
      "step": 4450
    },
    {
      "entropy": 5.516057109832763,
      "epoch": 4.046342571558383,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003800342463003383,
      "loss": 4.9128,
      "mean_token_accuracy": 0.214850452542305,
      "num_tokens": 9222683.0,
      "step": 4455
    },
    {
      "entropy": 5.467364931106568,
      "epoch": 4.050885960926851,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00037972156071583386,
      "loss": 4.8893,
      "mean_token_accuracy": 0.2070636495947838,
      "num_tokens": 9232099.0,
      "step": 4460
    },
    {
      "entropy": 5.444888544082642,
      "epoch": 4.05542935029532,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003794086167412884,
      "loss": 4.9293,
      "mean_token_accuracy": 0.20776356160640716,
      "num_tokens": 9242660.0,
      "step": 4465
    },
    {
      "entropy": 5.469394302368164,
      "epoch": 4.059972739663789,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003790954151488601,
      "loss": 4.9583,
      "mean_token_accuracy": 0.19615956991910935,
      "num_tokens": 9253981.0,
      "step": 4470
    },
    {
      "entropy": 5.511186790466309,
      "epoch": 4.064516129032258,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003787819567113427,
      "loss": 4.892,
      "mean_token_accuracy": 0.20929009914398194,
      "num_tokens": 9264262.0,
      "step": 4475
    },
    {
      "entropy": 5.446306610107422,
      "epoch": 4.069059518400727,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003784682422021638,
      "loss": 4.9092,
      "mean_token_accuracy": 0.20419019907712938,
      "num_tokens": 9273562.0,
      "step": 4480
    },
    {
      "entropy": 5.45888032913208,
      "epoch": 4.073602907769196,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003781542723953828,
      "loss": 4.9064,
      "mean_token_accuracy": 0.20672675371170043,
      "num_tokens": 9283171.0,
      "step": 4485
    },
    {
      "entropy": 5.489746379852295,
      "epoch": 4.078146297137665,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00037784004806568885,
      "loss": 4.8994,
      "mean_token_accuracy": 0.21013279855251313,
      "num_tokens": 9292624.0,
      "step": 4490
    },
    {
      "entropy": 5.499894905090332,
      "epoch": 4.082689686506134,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00037752556998839934,
      "loss": 4.9632,
      "mean_token_accuracy": 0.203107587993145,
      "num_tokens": 9303159.0,
      "step": 4495
    },
    {
      "entropy": 5.479097747802735,
      "epoch": 4.087233075874602,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003772108389394576,
      "loss": 4.9242,
      "mean_token_accuracy": 0.20648837685585023,
      "num_tokens": 9314340.0,
      "step": 4500
    },
    {
      "epoch": 4.087233075874602,
      "eval_entropy": 5.345510090229123,
      "eval_loss": 5.720780849456787,
      "eval_mean_token_accuracy": 0.16857837187342867,
      "eval_num_tokens": 9314340.0,
      "eval_runtime": 2.0131,
      "eval_samples_per_second": 1706.342,
      "eval_steps_per_second": 213.603,
      "step": 4500
    },
    {
      "entropy": 5.532383441925049,
      "epoch": 4.091776465243071,
      "grad_norm": 1.0,
      "learning_rate": 0.00037689585569543137,
      "loss": 4.9908,
      "mean_token_accuracy": 0.20776456147432326,
      "num_tokens": 9325381.0,
      "step": 4505
    },
    {
      "entropy": 5.446325588226318,
      "epoch": 4.09631985461154,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003765806210335103,
      "loss": 4.966,
      "mean_token_accuracy": 0.2022661805152893,
      "num_tokens": 9336639.0,
      "step": 4510
    },
    {
      "entropy": 5.485771179199219,
      "epoch": 4.100863243980009,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00037626513573150473,
      "loss": 4.8616,
      "mean_token_accuracy": 0.21219029277563095,
      "num_tokens": 9346794.0,
      "step": 4515
    },
    {
      "entropy": 5.434124374389649,
      "epoch": 4.105406633348478,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00037594940056784315,
      "loss": 4.917,
      "mean_token_accuracy": 0.21037623882293702,
      "num_tokens": 9356326.0,
      "step": 4520
    },
    {
      "entropy": 5.511466026306152,
      "epoch": 4.109950022716947,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003756334163215708,
      "loss": 4.9142,
      "mean_token_accuracy": 0.2065751701593399,
      "num_tokens": 9366402.0,
      "step": 4525
    },
    {
      "entropy": 5.453936910629272,
      "epoch": 4.114493412085416,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00037531718377234723,
      "loss": 4.8736,
      "mean_token_accuracy": 0.20992049127817153,
      "num_tokens": 9377067.0,
      "step": 4530
    },
    {
      "entropy": 5.464564037322998,
      "epoch": 4.119036801453885,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003750007037004451,
      "loss": 4.9256,
      "mean_token_accuracy": 0.20936121344566344,
      "num_tokens": 9387206.0,
      "step": 4535
    },
    {
      "entropy": 5.477842140197754,
      "epoch": 4.123580190822353,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003746839768867473,
      "loss": 4.8998,
      "mean_token_accuracy": 0.2088559791445732,
      "num_tokens": 9397195.0,
      "step": 4540
    },
    {
      "entropy": 5.412002801895142,
      "epoch": 4.128123580190822,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003743670041127458,
      "loss": 4.8742,
      "mean_token_accuracy": 0.20805131942033767,
      "num_tokens": 9407611.0,
      "step": 4545
    },
    {
      "entropy": 5.522242069244385,
      "epoch": 4.132666969559291,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003740497861605395,
      "loss": 4.9743,
      "mean_token_accuracy": 0.203430113196373,
      "num_tokens": 9417897.0,
      "step": 4550
    },
    {
      "entropy": 5.461606168746949,
      "epoch": 4.13721035892776,
      "grad_norm": 1.046875,
      "learning_rate": 0.000373732323812832,
      "loss": 4.9118,
      "mean_token_accuracy": 0.20127494335174562,
      "num_tokens": 9428896.0,
      "step": 4555
    },
    {
      "entropy": 5.412542772293091,
      "epoch": 4.141753748296229,
      "grad_norm": 1.09375,
      "learning_rate": 0.00037341461785293025,
      "loss": 4.8494,
      "mean_token_accuracy": 0.21423585265874862,
      "num_tokens": 9438854.0,
      "step": 4560
    },
    {
      "entropy": 5.389967012405395,
      "epoch": 4.146297137664698,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003730966690647421,
      "loss": 4.9051,
      "mean_token_accuracy": 0.207192562520504,
      "num_tokens": 9447798.0,
      "step": 4565
    },
    {
      "entropy": 5.446862363815308,
      "epoch": 4.150840527033167,
      "grad_norm": 1.078125,
      "learning_rate": 0.00037277847823277444,
      "loss": 4.8632,
      "mean_token_accuracy": 0.22013641744852067,
      "num_tokens": 9458930.0,
      "step": 4570
    },
    {
      "entropy": 5.396658086776734,
      "epoch": 4.155383916401636,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00037246004614213164,
      "loss": 4.8416,
      "mean_token_accuracy": 0.21147270053625106,
      "num_tokens": 9469740.0,
      "step": 4575
    },
    {
      "entropy": 5.440028095245362,
      "epoch": 4.159927305770104,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003721413735785132,
      "loss": 4.8524,
      "mean_token_accuracy": 0.2148097425699234,
      "num_tokens": 9479823.0,
      "step": 4580
    },
    {
      "entropy": 5.464069128036499,
      "epoch": 4.164470695138573,
      "grad_norm": 1.078125,
      "learning_rate": 0.000371822461328212,
      "loss": 4.8471,
      "mean_token_accuracy": 0.21501293778419495,
      "num_tokens": 9489995.0,
      "step": 4585
    },
    {
      "entropy": 5.440762615203857,
      "epoch": 4.169014084507042,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00037150331017811233,
      "loss": 4.9156,
      "mean_token_accuracy": 0.20599235147237777,
      "num_tokens": 9501311.0,
      "step": 4590
    },
    {
      "entropy": 5.481850814819336,
      "epoch": 4.173557473875511,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003711839209156878,
      "loss": 4.9728,
      "mean_token_accuracy": 0.20227048993110658,
      "num_tokens": 9511901.0,
      "step": 4595
    },
    {
      "entropy": 5.486803102493286,
      "epoch": 4.17810086324398,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003708642943289999,
      "loss": 4.9173,
      "mean_token_accuracy": 0.20348752588033675,
      "num_tokens": 9522908.0,
      "step": 4600
    },
    {
      "entropy": 5.511645936965943,
      "epoch": 4.182644252612449,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003705444312066953,
      "loss": 4.9989,
      "mean_token_accuracy": 0.20353905111551285,
      "num_tokens": 9533363.0,
      "step": 4605
    },
    {
      "entropy": 5.4505181312561035,
      "epoch": 4.187187641980918,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003702243323380045,
      "loss": 4.8713,
      "mean_token_accuracy": 0.20573434084653855,
      "num_tokens": 9543302.0,
      "step": 4610
    },
    {
      "entropy": 5.434371614456177,
      "epoch": 4.191731031349387,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00036990399851273965,
      "loss": 4.9182,
      "mean_token_accuracy": 0.19429059624671935,
      "num_tokens": 9553143.0,
      "step": 4615
    },
    {
      "entropy": 5.516224193572998,
      "epoch": 4.196274420717856,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0003695834305212925,
      "loss": 4.9392,
      "mean_token_accuracy": 0.20294555723667146,
      "num_tokens": 9564236.0,
      "step": 4620
    },
    {
      "entropy": 5.443789386749268,
      "epoch": 4.200817810086324,
      "grad_norm": 1.171875,
      "learning_rate": 0.00036926262915463294,
      "loss": 4.8993,
      "mean_token_accuracy": 0.21000194549560547,
      "num_tokens": 9574363.0,
      "step": 4625
    },
    {
      "entropy": 5.403774690628052,
      "epoch": 4.205361199454793,
      "grad_norm": 1.15625,
      "learning_rate": 0.00036894159520430616,
      "loss": 4.9357,
      "mean_token_accuracy": 0.20407017469406127,
      "num_tokens": 9585631.0,
      "step": 4630
    },
    {
      "entropy": 5.480752658843994,
      "epoch": 4.209904588823262,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00036862032946243173,
      "loss": 4.9115,
      "mean_token_accuracy": 0.19986301064491271,
      "num_tokens": 9596826.0,
      "step": 4635
    },
    {
      "entropy": 5.371126937866211,
      "epoch": 4.214447978191731,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00036829883272170087,
      "loss": 4.8535,
      "mean_token_accuracy": 0.20859225541353227,
      "num_tokens": 9608158.0,
      "step": 4640
    },
    {
      "entropy": 5.415221691131592,
      "epoch": 4.2189913675602,
      "grad_norm": 1.046875,
      "learning_rate": 0.00036797710577537486,
      "loss": 4.9606,
      "mean_token_accuracy": 0.2066285640001297,
      "num_tokens": 9619437.0,
      "step": 4645
    },
    {
      "entropy": 5.5375312805175785,
      "epoch": 4.223534756928669,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00036765514941728295,
      "loss": 4.9516,
      "mean_token_accuracy": 0.211208313703537,
      "num_tokens": 9629913.0,
      "step": 4650
    },
    {
      "entropy": 5.4514195919036865,
      "epoch": 4.228078146297138,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0003673329644418204,
      "loss": 4.9233,
      "mean_token_accuracy": 0.20623777359724044,
      "num_tokens": 9640033.0,
      "step": 4655
    },
    {
      "entropy": 5.484861660003662,
      "epoch": 4.232621535665607,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00036701055164394664,
      "loss": 4.9574,
      "mean_token_accuracy": 0.20476742535829545,
      "num_tokens": 9651010.0,
      "step": 4660
    },
    {
      "entropy": 5.416854000091552,
      "epoch": 4.237164925034075,
      "grad_norm": 1.109375,
      "learning_rate": 0.00036668791181918313,
      "loss": 4.8794,
      "mean_token_accuracy": 0.2079287901520729,
      "num_tokens": 9661549.0,
      "step": 4665
    },
    {
      "entropy": 5.388582706451416,
      "epoch": 4.241708314402544,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003663650457636117,
      "loss": 4.9215,
      "mean_token_accuracy": 0.20407782196998597,
      "num_tokens": 9671619.0,
      "step": 4670
    },
    {
      "entropy": 5.55538592338562,
      "epoch": 4.246251703771013,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003660419542738722,
      "loss": 4.97,
      "mean_token_accuracy": 0.20962463021278382,
      "num_tokens": 9682582.0,
      "step": 4675
    },
    {
      "entropy": 5.405483293533325,
      "epoch": 4.250795093139482,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003657186381471608,
      "loss": 4.8863,
      "mean_token_accuracy": 0.20848942697048187,
      "num_tokens": 9691997.0,
      "step": 4680
    },
    {
      "entropy": 5.445845413208008,
      "epoch": 4.255338482507951,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00036539509818122765,
      "loss": 4.9576,
      "mean_token_accuracy": 0.20391900837421417,
      "num_tokens": 9702105.0,
      "step": 4685
    },
    {
      "entropy": 5.510810708999633,
      "epoch": 4.25988187187642,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003650713351743758,
      "loss": 4.9458,
      "mean_token_accuracy": 0.2082253873348236,
      "num_tokens": 9712347.0,
      "step": 4690
    },
    {
      "entropy": 5.439699363708496,
      "epoch": 4.264425261244889,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003647473499254581,
      "loss": 4.9324,
      "mean_token_accuracy": 0.20745539367198945,
      "num_tokens": 9723116.0,
      "step": 4695
    },
    {
      "entropy": 5.368207693099976,
      "epoch": 4.268968650613358,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000364423143233876,
      "loss": 4.8743,
      "mean_token_accuracy": 0.20722342729568483,
      "num_tokens": 9732656.0,
      "step": 4700
    },
    {
      "entropy": 5.489114665985108,
      "epoch": 4.273512039981826,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00036409871589957723,
      "loss": 4.972,
      "mean_token_accuracy": 0.2039525792002678,
      "num_tokens": 9743312.0,
      "step": 4705
    },
    {
      "entropy": 5.457533550262451,
      "epoch": 4.278055429350295,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000363774068723054,
      "loss": 4.9211,
      "mean_token_accuracy": 0.20605117827653885,
      "num_tokens": 9753358.0,
      "step": 4710
    },
    {
      "entropy": 5.445799255371094,
      "epoch": 4.282598818718764,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00036344920250534106,
      "loss": 4.8385,
      "mean_token_accuracy": 0.2128051444888115,
      "num_tokens": 9763472.0,
      "step": 4715
    },
    {
      "entropy": 5.401280355453491,
      "epoch": 4.287142208087233,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003631241180480133,
      "loss": 4.8715,
      "mean_token_accuracy": 0.20780899673700332,
      "num_tokens": 9773700.0,
      "step": 4720
    },
    {
      "entropy": 5.421959447860718,
      "epoch": 4.291685597455702,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003627988161531845,
      "loss": 4.89,
      "mean_token_accuracy": 0.20720557272434234,
      "num_tokens": 9784042.0,
      "step": 4725
    },
    {
      "entropy": 5.480333375930786,
      "epoch": 4.296228986824171,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003624732976235046,
      "loss": 5.016,
      "mean_token_accuracy": 0.20216021090745925,
      "num_tokens": 9794464.0,
      "step": 4730
    },
    {
      "entropy": 5.460859107971191,
      "epoch": 4.30077237619264,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003621475632621584,
      "loss": 4.9436,
      "mean_token_accuracy": 0.2035440057516098,
      "num_tokens": 9804459.0,
      "step": 4735
    },
    {
      "entropy": 5.458033418655395,
      "epoch": 4.305315765561109,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003618216138728627,
      "loss": 4.9281,
      "mean_token_accuracy": 0.2090199425816536,
      "num_tokens": 9814962.0,
      "step": 4740
    },
    {
      "entropy": 5.467393398284912,
      "epoch": 4.309859154929577,
      "grad_norm": 1.09375,
      "learning_rate": 0.00036149545025986567,
      "loss": 4.9673,
      "mean_token_accuracy": 0.20339298993349075,
      "num_tokens": 9825184.0,
      "step": 4745
    },
    {
      "entropy": 5.497726774215698,
      "epoch": 4.314402544298046,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00036116907322794335,
      "loss": 5.0176,
      "mean_token_accuracy": 0.1979744866490364,
      "num_tokens": 9835184.0,
      "step": 4750
    },
    {
      "entropy": 5.462279891967773,
      "epoch": 4.318945933666515,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00036084248358239857,
      "loss": 4.9237,
      "mean_token_accuracy": 0.20223242193460464,
      "num_tokens": 9845358.0,
      "step": 4755
    },
    {
      "entropy": 5.4031494140625,
      "epoch": 4.323489323034984,
      "grad_norm": 1.21875,
      "learning_rate": 0.00036051568212905897,
      "loss": 4.8598,
      "mean_token_accuracy": 0.20770094841718673,
      "num_tokens": 9856519.0,
      "step": 4760
    },
    {
      "entropy": 5.466473579406738,
      "epoch": 4.328032712403453,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003601886696742745,
      "loss": 4.9334,
      "mean_token_accuracy": 0.20561221092939377,
      "num_tokens": 9867478.0,
      "step": 4765
    },
    {
      "entropy": 5.402345752716064,
      "epoch": 4.332576101771922,
      "grad_norm": 1.15625,
      "learning_rate": 0.000359861447024916,
      "loss": 4.9147,
      "mean_token_accuracy": 0.20839653462171553,
      "num_tokens": 9877709.0,
      "step": 4770
    },
    {
      "entropy": 5.420436191558838,
      "epoch": 4.337119491140391,
      "grad_norm": 1.046875,
      "learning_rate": 0.00035953401498837285,
      "loss": 4.8765,
      "mean_token_accuracy": 0.21107325851917266,
      "num_tokens": 9888134.0,
      "step": 4775
    },
    {
      "entropy": 5.433488273620606,
      "epoch": 4.34166288050886,
      "grad_norm": 0.953125,
      "learning_rate": 0.00035920637437255096,
      "loss": 4.9227,
      "mean_token_accuracy": 0.2078911155462265,
      "num_tokens": 9899150.0,
      "step": 4780
    },
    {
      "entropy": 5.489434814453125,
      "epoch": 4.346206269877328,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000358878525985871,
      "loss": 4.9302,
      "mean_token_accuracy": 0.20458950251340866,
      "num_tokens": 9909599.0,
      "step": 4785
    },
    {
      "entropy": 5.460358619689941,
      "epoch": 4.350749659245797,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003585504706372663,
      "loss": 4.889,
      "mean_token_accuracy": 0.215658600628376,
      "num_tokens": 9919607.0,
      "step": 4790
    },
    {
      "entropy": 5.410480880737305,
      "epoch": 4.355293048614266,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003582222091361808,
      "loss": 4.9406,
      "mean_token_accuracy": 0.20467400252819062,
      "num_tokens": 9930431.0,
      "step": 4795
    },
    {
      "entropy": 5.443422651290893,
      "epoch": 4.359836437982735,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003578937422925671,
      "loss": 4.9071,
      "mean_token_accuracy": 0.2098840981721878,
      "num_tokens": 9940647.0,
      "step": 4800
    },
    {
      "entropy": 5.428092288970947,
      "epoch": 4.364379827351204,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003575650709168846,
      "loss": 4.9458,
      "mean_token_accuracy": 0.1994195982813835,
      "num_tokens": 9951110.0,
      "step": 4805
    },
    {
      "entropy": 5.456834030151367,
      "epoch": 4.368923216719673,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00035723619582009715,
      "loss": 4.9128,
      "mean_token_accuracy": 0.20758690685033798,
      "num_tokens": 9961158.0,
      "step": 4810
    },
    {
      "entropy": 5.396404361724853,
      "epoch": 4.373466606088142,
      "grad_norm": 1.21875,
      "learning_rate": 0.00035690711781367137,
      "loss": 4.9514,
      "mean_token_accuracy": 0.20395814776420593,
      "num_tokens": 9971051.0,
      "step": 4815
    },
    {
      "entropy": 5.386496353149414,
      "epoch": 4.378009995456611,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003565778377095744,
      "loss": 4.9087,
      "mean_token_accuracy": 0.2015366718173027,
      "num_tokens": 9981990.0,
      "step": 4820
    },
    {
      "entropy": 5.41032133102417,
      "epoch": 4.38255338482508,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003562483563202724,
      "loss": 4.8582,
      "mean_token_accuracy": 0.21003009378910065,
      "num_tokens": 9991668.0,
      "step": 4825
    },
    {
      "entropy": 5.382327699661255,
      "epoch": 4.387096774193548,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00035591867445872784,
      "loss": 4.8104,
      "mean_token_accuracy": 0.21280850768089293,
      "num_tokens": 10001519.0,
      "step": 4830
    },
    {
      "entropy": 5.389340877532959,
      "epoch": 4.391640163562017,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000355588792938398,
      "loss": 4.8757,
      "mean_token_accuracy": 0.21122119426727295,
      "num_tokens": 10010856.0,
      "step": 4835
    },
    {
      "entropy": 5.366349983215332,
      "epoch": 4.396183552930486,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003552587125732326,
      "loss": 4.8873,
      "mean_token_accuracy": 0.2083209902048111,
      "num_tokens": 10020647.0,
      "step": 4840
    },
    {
      "entropy": 5.427668190002441,
      "epoch": 4.400726942298955,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00035492843417767224,
      "loss": 4.8079,
      "mean_token_accuracy": 0.21706265807151795,
      "num_tokens": 10030114.0,
      "step": 4845
    },
    {
      "entropy": 5.459252786636353,
      "epoch": 4.405270331667424,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000354597958566646,
      "loss": 4.9112,
      "mean_token_accuracy": 0.21120489686727523,
      "num_tokens": 10039863.0,
      "step": 4850
    },
    {
      "entropy": 5.356090068817139,
      "epoch": 4.409813721035893,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00035426728655556954,
      "loss": 4.8692,
      "mean_token_accuracy": 0.20364532619714737,
      "num_tokens": 10049111.0,
      "step": 4855
    },
    {
      "entropy": 5.427938604354859,
      "epoch": 4.414357110404362,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003539364189603433,
      "loss": 4.9539,
      "mean_token_accuracy": 0.209016153216362,
      "num_tokens": 10058762.0,
      "step": 4860
    },
    {
      "entropy": 5.479335594177246,
      "epoch": 4.418900499772831,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00035360535659735006,
      "loss": 4.9447,
      "mean_token_accuracy": 0.20219985991716385,
      "num_tokens": 10068587.0,
      "step": 4865
    },
    {
      "entropy": 5.463398694992065,
      "epoch": 4.423443889141299,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003532741002834534,
      "loss": 4.9085,
      "mean_token_accuracy": 0.2037759393453598,
      "num_tokens": 10079386.0,
      "step": 4870
    },
    {
      "entropy": 5.383228349685669,
      "epoch": 4.427987278509768,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00035294265083599526,
      "loss": 4.8478,
      "mean_token_accuracy": 0.21351510435342788,
      "num_tokens": 10089490.0,
      "step": 4875
    },
    {
      "entropy": 5.39179368019104,
      "epoch": 4.432530667878237,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003526110090727942,
      "loss": 4.932,
      "mean_token_accuracy": 0.21079819053411483,
      "num_tokens": 10100465.0,
      "step": 4880
    },
    {
      "entropy": 5.457243776321411,
      "epoch": 4.437074057246706,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003522791758121434,
      "loss": 4.8631,
      "mean_token_accuracy": 0.21403679847717286,
      "num_tokens": 10110919.0,
      "step": 4885
    },
    {
      "entropy": 5.3957586765289305,
      "epoch": 4.441617446615175,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00035194715187280835,
      "loss": 4.913,
      "mean_token_accuracy": 0.20956196188926696,
      "num_tokens": 10121116.0,
      "step": 4890
    },
    {
      "entropy": 5.504559087753296,
      "epoch": 4.446160835983644,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003516149380740252,
      "loss": 5.011,
      "mean_token_accuracy": 0.20230667442083358,
      "num_tokens": 10132269.0,
      "step": 4895
    },
    {
      "entropy": 5.4354900360107425,
      "epoch": 4.450704225352113,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003512825352354983,
      "loss": 4.9005,
      "mean_token_accuracy": 0.2148653119802475,
      "num_tokens": 10142856.0,
      "step": 4900
    },
    {
      "entropy": 5.381114530563354,
      "epoch": 4.455247614720582,
      "grad_norm": 1.234375,
      "learning_rate": 0.00035094994417739884,
      "loss": 4.8793,
      "mean_token_accuracy": 0.20515192300081253,
      "num_tokens": 10152377.0,
      "step": 4905
    },
    {
      "entropy": 5.470276737213135,
      "epoch": 4.45979100408905,
      "grad_norm": 1.203125,
      "learning_rate": 0.000350617165720362,
      "loss": 4.9573,
      "mean_token_accuracy": 0.2039278194308281,
      "num_tokens": 10161961.0,
      "step": 4910
    },
    {
      "entropy": 5.4940601825714115,
      "epoch": 4.464334393457519,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00035028420068548567,
      "loss": 4.9135,
      "mean_token_accuracy": 0.21195923686027526,
      "num_tokens": 10171624.0,
      "step": 4915
    },
    {
      "entropy": 5.503613901138306,
      "epoch": 4.468877782825988,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003499510498943279,
      "loss": 5.0012,
      "mean_token_accuracy": 0.2024716004729271,
      "num_tokens": 10182642.0,
      "step": 4920
    },
    {
      "entropy": 5.444105958938598,
      "epoch": 4.473421172194457,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00034961771416890526,
      "loss": 4.9647,
      "mean_token_accuracy": 0.21156363785266877,
      "num_tokens": 10193307.0,
      "step": 4925
    },
    {
      "entropy": 5.404733943939209,
      "epoch": 4.477964561562926,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003492841943316903,
      "loss": 4.9478,
      "mean_token_accuracy": 0.20503314584493637,
      "num_tokens": 10203506.0,
      "step": 4930
    },
    {
      "entropy": 5.402101325988769,
      "epoch": 4.482507950931395,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0003489504912056103,
      "loss": 4.8292,
      "mean_token_accuracy": 0.2116386190056801,
      "num_tokens": 10214266.0,
      "step": 4935
    },
    {
      "entropy": 5.349980115890503,
      "epoch": 4.487051340299864,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003486166056140445,
      "loss": 4.844,
      "mean_token_accuracy": 0.21342409104108812,
      "num_tokens": 10224953.0,
      "step": 4940
    },
    {
      "entropy": 5.388110399246216,
      "epoch": 4.491594729668333,
      "grad_norm": 1.09375,
      "learning_rate": 0.00034828253838082223,
      "loss": 4.8899,
      "mean_token_accuracy": 0.20608312785625457,
      "num_tokens": 10235435.0,
      "step": 4945
    },
    {
      "entropy": 5.416355752944947,
      "epoch": 4.496138119036801,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00034794829033022134,
      "loss": 4.891,
      "mean_token_accuracy": 0.20996856838464736,
      "num_tokens": 10245677.0,
      "step": 4950
    },
    {
      "entropy": 5.322266626358032,
      "epoch": 4.50068150840527,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003476138622869657,
      "loss": 4.8584,
      "mean_token_accuracy": 0.20683515220880508,
      "num_tokens": 10256135.0,
      "step": 4955
    },
    {
      "entropy": 5.393886423110962,
      "epoch": 4.505224897773739,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003472792550762232,
      "loss": 4.9621,
      "mean_token_accuracy": 0.2065868631005287,
      "num_tokens": 10266465.0,
      "step": 4960
    },
    {
      "entropy": 5.5116486072540285,
      "epoch": 4.509768287142208,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003469444695236038,
      "loss": 4.9602,
      "mean_token_accuracy": 0.20579384714365007,
      "num_tokens": 10278123.0,
      "step": 4965
    },
    {
      "entropy": 5.386662292480469,
      "epoch": 4.514311676510677,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00034660950645515765,
      "loss": 4.9238,
      "mean_token_accuracy": 0.20658672004938125,
      "num_tokens": 10287486.0,
      "step": 4970
    },
    {
      "entropy": 5.446750974655151,
      "epoch": 4.518855065879146,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003462743666973728,
      "loss": 4.9098,
      "mean_token_accuracy": 0.21138269752264022,
      "num_tokens": 10298100.0,
      "step": 4975
    },
    {
      "entropy": 5.485393476486206,
      "epoch": 4.523398455247615,
      "grad_norm": 1.125,
      "learning_rate": 0.0003459390510771734,
      "loss": 4.9813,
      "mean_token_accuracy": 0.20802601128816606,
      "num_tokens": 10308549.0,
      "step": 4980
    },
    {
      "entropy": 5.425182723999024,
      "epoch": 4.527941844616084,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003456035604219172,
      "loss": 4.921,
      "mean_token_accuracy": 0.20278574675321578,
      "num_tokens": 10319030.0,
      "step": 4985
    },
    {
      "entropy": 5.359147644042968,
      "epoch": 4.532485233984552,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003452678955593942,
      "loss": 4.9636,
      "mean_token_accuracy": 0.20471454858779908,
      "num_tokens": 10329613.0,
      "step": 4990
    },
    {
      "entropy": 5.454889011383057,
      "epoch": 4.537028623353021,
      "grad_norm": 1.25,
      "learning_rate": 0.000344932057317824,
      "loss": 4.9621,
      "mean_token_accuracy": 0.20497691184282302,
      "num_tokens": 10339844.0,
      "step": 4995
    },
    {
      "entropy": 5.461254024505616,
      "epoch": 4.54157201272149,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003445960465258542,
      "loss": 4.9818,
      "mean_token_accuracy": 0.20175698697566985,
      "num_tokens": 10350906.0,
      "step": 5000
    },
    {
      "epoch": 4.54157201272149,
      "eval_entropy": 5.280047230387843,
      "eval_loss": 5.692351341247559,
      "eval_mean_token_accuracy": 0.17146256896992063,
      "eval_num_tokens": 10350906.0,
      "eval_runtime": 2.0158,
      "eval_samples_per_second": 1704.01,
      "eval_steps_per_second": 213.311,
      "step": 5000
    },
    {
      "entropy": 5.375666809082031,
      "epoch": 4.546115402089959,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003442598640125578,
      "loss": 4.9248,
      "mean_token_accuracy": 0.20097829401493073,
      "num_tokens": 10361574.0,
      "step": 5005
    },
    {
      "entropy": 5.469150543212891,
      "epoch": 4.550658791458428,
      "grad_norm": 1.265625,
      "learning_rate": 0.000343923510607432,
      "loss": 4.9561,
      "mean_token_accuracy": 0.20554766207933425,
      "num_tokens": 10372714.0,
      "step": 5010
    },
    {
      "entropy": 5.392849063873291,
      "epoch": 4.555202180826897,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003435869871403952,
      "loss": 4.8689,
      "mean_token_accuracy": 0.21154365092515945,
      "num_tokens": 10382143.0,
      "step": 5015
    },
    {
      "entropy": 5.402473592758179,
      "epoch": 4.559745570195366,
      "grad_norm": 1.171875,
      "learning_rate": 0.00034325029444178565,
      "loss": 4.9006,
      "mean_token_accuracy": 0.2075703486800194,
      "num_tokens": 10392040.0,
      "step": 5020
    },
    {
      "entropy": 5.440453052520752,
      "epoch": 4.564288959563835,
      "grad_norm": 1.0625,
      "learning_rate": 0.00034291343334235915,
      "loss": 4.9598,
      "mean_token_accuracy": 0.19682183265686035,
      "num_tokens": 10403230.0,
      "step": 5025
    },
    {
      "entropy": 5.309180736541748,
      "epoch": 4.568832348932304,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003425764046732869,
      "loss": 4.8454,
      "mean_token_accuracy": 0.21326805651187897,
      "num_tokens": 10413466.0,
      "step": 5030
    },
    {
      "entropy": 5.433459043502808,
      "epoch": 4.573375738300772,
      "grad_norm": 1.15625,
      "learning_rate": 0.00034223920926615376,
      "loss": 4.8965,
      "mean_token_accuracy": 0.20729638040065765,
      "num_tokens": 10422559.0,
      "step": 5035
    },
    {
      "entropy": 5.432064056396484,
      "epoch": 4.577919127669241,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00034190184795295577,
      "loss": 4.8772,
      "mean_token_accuracy": 0.212820465862751,
      "num_tokens": 10432544.0,
      "step": 5040
    },
    {
      "entropy": 5.388049554824829,
      "epoch": 4.58246251703771,
      "grad_norm": 1.078125,
      "learning_rate": 0.00034156432156609844,
      "loss": 4.9565,
      "mean_token_accuracy": 0.20192985683679582,
      "num_tokens": 10442014.0,
      "step": 5045
    },
    {
      "entropy": 5.35285062789917,
      "epoch": 4.587005906406179,
      "grad_norm": 1.171875,
      "learning_rate": 0.00034122663093839466,
      "loss": 4.8029,
      "mean_token_accuracy": 0.21481169909238815,
      "num_tokens": 10451093.0,
      "step": 5050
    },
    {
      "entropy": 5.41038875579834,
      "epoch": 4.591549295774648,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003408887769030624,
      "loss": 4.8251,
      "mean_token_accuracy": 0.21288346797227858,
      "num_tokens": 10461678.0,
      "step": 5055
    },
    {
      "entropy": 5.3476423740386965,
      "epoch": 4.596092685143117,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000340550760293723,
      "loss": 4.8845,
      "mean_token_accuracy": 0.21030484586954118,
      "num_tokens": 10471607.0,
      "step": 5060
    },
    {
      "entropy": 5.446288585662842,
      "epoch": 4.600636074511586,
      "grad_norm": 1.03125,
      "learning_rate": 0.00034021258194439876,
      "loss": 4.9401,
      "mean_token_accuracy": 0.20515215992927552,
      "num_tokens": 10482453.0,
      "step": 5065
    },
    {
      "entropy": 5.348289680480957,
      "epoch": 4.605179463880054,
      "grad_norm": 1.09375,
      "learning_rate": 0.00033987424268951117,
      "loss": 4.8337,
      "mean_token_accuracy": 0.21458347886800766,
      "num_tokens": 10492424.0,
      "step": 5070
    },
    {
      "entropy": 5.40476770401001,
      "epoch": 4.609722853248523,
      "grad_norm": 1.296875,
      "learning_rate": 0.00033953574336387867,
      "loss": 4.8798,
      "mean_token_accuracy": 0.20457347482442856,
      "num_tokens": 10500958.0,
      "step": 5075
    },
    {
      "entropy": 5.40281343460083,
      "epoch": 4.614266242616992,
      "grad_norm": 1.25,
      "learning_rate": 0.0003391970848027148,
      "loss": 4.936,
      "mean_token_accuracy": 0.2119533896446228,
      "num_tokens": 10510906.0,
      "step": 5080
    },
    {
      "entropy": 5.39763069152832,
      "epoch": 4.618809631985461,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003388582678416257,
      "loss": 4.8466,
      "mean_token_accuracy": 0.2091699317097664,
      "num_tokens": 10521832.0,
      "step": 5085
    },
    {
      "entropy": 5.40036301612854,
      "epoch": 4.62335302135393,
      "grad_norm": 1.140625,
      "learning_rate": 0.00033851929331660876,
      "loss": 4.8565,
      "mean_token_accuracy": 0.20985513776540757,
      "num_tokens": 10532310.0,
      "step": 5090
    },
    {
      "entropy": 5.387194204330444,
      "epoch": 4.627896410722399,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00033818016206404983,
      "loss": 4.8845,
      "mean_token_accuracy": 0.20544965863227843,
      "num_tokens": 10543979.0,
      "step": 5095
    },
    {
      "entropy": 5.507230806350708,
      "epoch": 4.632439800090868,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003378408749207215,
      "loss": 4.9896,
      "mean_token_accuracy": 0.19879006892442702,
      "num_tokens": 10553757.0,
      "step": 5100
    },
    {
      "entropy": 5.446568298339844,
      "epoch": 4.636983189459337,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003375014327237812,
      "loss": 4.9232,
      "mean_token_accuracy": 0.20174409151077272,
      "num_tokens": 10563575.0,
      "step": 5105
    },
    {
      "entropy": 5.452083921432495,
      "epoch": 4.641526578827806,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003371618363107686,
      "loss": 4.9558,
      "mean_token_accuracy": 0.20682297497987748,
      "num_tokens": 10572828.0,
      "step": 5110
    },
    {
      "entropy": 5.352878570556641,
      "epoch": 4.646069968196274,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00033682208651960444,
      "loss": 4.9121,
      "mean_token_accuracy": 0.20493586212396622,
      "num_tokens": 10582567.0,
      "step": 5115
    },
    {
      "entropy": 5.390731620788574,
      "epoch": 4.650613357564743,
      "grad_norm": 1.109375,
      "learning_rate": 0.00033648218418858716,
      "loss": 4.9069,
      "mean_token_accuracy": 0.2096439555287361,
      "num_tokens": 10594046.0,
      "step": 5120
    },
    {
      "entropy": 5.424234437942505,
      "epoch": 4.655156746933212,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003361421301563922,
      "loss": 4.8944,
      "mean_token_accuracy": 0.20958732217550277,
      "num_tokens": 10604343.0,
      "step": 5125
    },
    {
      "entropy": 5.388157939910888,
      "epoch": 4.659700136301681,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003358019252620691,
      "loss": 4.9804,
      "mean_token_accuracy": 0.20471182465553284,
      "num_tokens": 10615140.0,
      "step": 5130
    },
    {
      "entropy": 5.462279319763184,
      "epoch": 4.66424352567015,
      "grad_norm": 1.125,
      "learning_rate": 0.00033546157034503954,
      "loss": 4.9302,
      "mean_token_accuracy": 0.21176572144031525,
      "num_tokens": 10624768.0,
      "step": 5135
    },
    {
      "entropy": 5.399920988082886,
      "epoch": 4.668786915038619,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003351210662450955,
      "loss": 4.8401,
      "mean_token_accuracy": 0.21388722360134124,
      "num_tokens": 10635246.0,
      "step": 5140
    },
    {
      "entropy": 5.352454805374146,
      "epoch": 4.673330304407088,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003347804138023968,
      "loss": 4.8454,
      "mean_token_accuracy": 0.21708537340164186,
      "num_tokens": 10645118.0,
      "step": 5145
    },
    {
      "entropy": 5.427688550949097,
      "epoch": 4.677873693775557,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003344396138574698,
      "loss": 4.875,
      "mean_token_accuracy": 0.21377287060022354,
      "num_tokens": 10655846.0,
      "step": 5150
    },
    {
      "entropy": 5.434671354293823,
      "epoch": 4.682417083144026,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003340986672512042,
      "loss": 4.9649,
      "mean_token_accuracy": 0.20583093166351318,
      "num_tokens": 10665741.0,
      "step": 5155
    },
    {
      "entropy": 5.422264432907104,
      "epoch": 4.686960472512494,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003337575748248521,
      "loss": 4.9205,
      "mean_token_accuracy": 0.20550040453672408,
      "num_tokens": 10676120.0,
      "step": 5160
    },
    {
      "entropy": 5.399494218826294,
      "epoch": 4.691503861880963,
      "grad_norm": 1.140625,
      "learning_rate": 0.00033341633742002503,
      "loss": 4.9014,
      "mean_token_accuracy": 0.20292871743440627,
      "num_tokens": 10686195.0,
      "step": 5165
    },
    {
      "entropy": 5.385882186889648,
      "epoch": 4.696047251249432,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003330749558786924,
      "loss": 4.9276,
      "mean_token_accuracy": 0.20876037925481797,
      "num_tokens": 10697278.0,
      "step": 5170
    },
    {
      "entropy": 5.4281505107879635,
      "epoch": 4.700590640617901,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003327334310431791,
      "loss": 4.969,
      "mean_token_accuracy": 0.2098015785217285,
      "num_tokens": 10707937.0,
      "step": 5175
    },
    {
      "entropy": 5.452768087387085,
      "epoch": 4.70513402998637,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003323917637561638,
      "loss": 4.8215,
      "mean_token_accuracy": 0.20917874872684478,
      "num_tokens": 10717963.0,
      "step": 5180
    },
    {
      "entropy": 5.364978837966919,
      "epoch": 4.709677419354839,
      "grad_norm": 1.171875,
      "learning_rate": 0.00033204995486067664,
      "loss": 4.9117,
      "mean_token_accuracy": 0.20494003146886824,
      "num_tokens": 10728294.0,
      "step": 5185
    },
    {
      "entropy": 5.5251671314239506,
      "epoch": 4.714220808723308,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00033170800520009703,
      "loss": 4.9661,
      "mean_token_accuracy": 0.20230364352464675,
      "num_tokens": 10739617.0,
      "step": 5190
    },
    {
      "entropy": 5.424886655807495,
      "epoch": 4.718764198091776,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003313659156181518,
      "loss": 4.9846,
      "mean_token_accuracy": 0.20391473174095154,
      "num_tokens": 10750704.0,
      "step": 5195
    },
    {
      "entropy": 5.407444286346435,
      "epoch": 4.723307587460245,
      "grad_norm": 1.125,
      "learning_rate": 0.00033102368695891283,
      "loss": 4.8963,
      "mean_token_accuracy": 0.2175852656364441,
      "num_tokens": 10760106.0,
      "step": 5200
    },
    {
      "entropy": 5.50756607055664,
      "epoch": 4.727850976828714,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00033068132006679554,
      "loss": 5.0536,
      "mean_token_accuracy": 0.2014394983649254,
      "num_tokens": 10770963.0,
      "step": 5205
    },
    {
      "entropy": 5.442524433135986,
      "epoch": 4.732394366197183,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00033033881578655605,
      "loss": 4.976,
      "mean_token_accuracy": 0.20333271026611327,
      "num_tokens": 10781441.0,
      "step": 5210
    },
    {
      "entropy": 5.449512195587158,
      "epoch": 4.736937755565652,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003299961749632898,
      "loss": 4.9423,
      "mean_token_accuracy": 0.21045019179582597,
      "num_tokens": 10791548.0,
      "step": 5215
    },
    {
      "entropy": 5.4583837509155275,
      "epoch": 4.741481144934121,
      "grad_norm": 1.140625,
      "learning_rate": 0.00032965339844242864,
      "loss": 4.9143,
      "mean_token_accuracy": 0.20794034749269485,
      "num_tokens": 10801525.0,
      "step": 5220
    },
    {
      "entropy": 5.4070212841033936,
      "epoch": 4.74602453430259,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003293104870697399,
      "loss": 4.888,
      "mean_token_accuracy": 0.20212750732898713,
      "num_tokens": 10811907.0,
      "step": 5225
    },
    {
      "entropy": 5.374733209609985,
      "epoch": 4.750567923671059,
      "grad_norm": 1.15625,
      "learning_rate": 0.00032896744169132305,
      "loss": 4.8701,
      "mean_token_accuracy": 0.2166287124156952,
      "num_tokens": 10822996.0,
      "step": 5230
    },
    {
      "entropy": 5.393605756759643,
      "epoch": 4.755111313039528,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003286242631536087,
      "loss": 4.8328,
      "mean_token_accuracy": 0.20981537997722627,
      "num_tokens": 10832561.0,
      "step": 5235
    },
    {
      "entropy": 5.439042758941651,
      "epoch": 4.759654702407996,
      "grad_norm": 1.1875,
      "learning_rate": 0.00032828095230335555,
      "loss": 4.9261,
      "mean_token_accuracy": 0.2071788340806961,
      "num_tokens": 10842591.0,
      "step": 5240
    },
    {
      "entropy": 5.388352632522583,
      "epoch": 4.764198091776465,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003279375099876491,
      "loss": 4.948,
      "mean_token_accuracy": 0.20108623504638673,
      "num_tokens": 10852888.0,
      "step": 5245
    },
    {
      "entropy": 5.495434761047363,
      "epoch": 4.768741481144934,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003275939370538992,
      "loss": 4.971,
      "mean_token_accuracy": 0.20417943000793456,
      "num_tokens": 10863792.0,
      "step": 5250
    },
    {
      "entropy": 5.382977151870728,
      "epoch": 4.773284870513403,
      "grad_norm": 1.140625,
      "learning_rate": 0.00032725023434983767,
      "loss": 4.8354,
      "mean_token_accuracy": 0.21415474116802216,
      "num_tokens": 10873206.0,
      "step": 5255
    },
    {
      "entropy": 5.388292503356934,
      "epoch": 4.777828259881872,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00032690640272351715,
      "loss": 4.9614,
      "mean_token_accuracy": 0.20495930165052414,
      "num_tokens": 10883686.0,
      "step": 5260
    },
    {
      "entropy": 5.311230325698853,
      "epoch": 4.782371649250341,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003265624430233076,
      "loss": 4.9032,
      "mean_token_accuracy": 0.21490921974182128,
      "num_tokens": 10894416.0,
      "step": 5265
    },
    {
      "entropy": 5.491904163360596,
      "epoch": 4.78691503861881,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003262183560978956,
      "loss": 5.0035,
      "mean_token_accuracy": 0.20856187492609024,
      "num_tokens": 10904405.0,
      "step": 5270
    },
    {
      "entropy": 5.534668254852295,
      "epoch": 4.791458427987278,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003258741427962814,
      "loss": 4.9163,
      "mean_token_accuracy": 0.20928130447864532,
      "num_tokens": 10915339.0,
      "step": 5275
    },
    {
      "entropy": 5.404999160766602,
      "epoch": 4.796001817355747,
      "grad_norm": 1.078125,
      "learning_rate": 0.00032552980396777714,
      "loss": 4.9082,
      "mean_token_accuracy": 0.2078085646033287,
      "num_tokens": 10926373.0,
      "step": 5280
    },
    {
      "entropy": 5.375723743438721,
      "epoch": 4.800545206724216,
      "grad_norm": 1.09375,
      "learning_rate": 0.00032518534046200464,
      "loss": 4.9882,
      "mean_token_accuracy": 0.20805052369832994,
      "num_tokens": 10936695.0,
      "step": 5285
    },
    {
      "entropy": 5.501412010192871,
      "epoch": 4.805088596092685,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003248407531288932,
      "loss": 5.0242,
      "mean_token_accuracy": 0.19748157560825347,
      "num_tokens": 10949741.0,
      "step": 5290
    },
    {
      "entropy": 5.463061141967773,
      "epoch": 4.809631985461154,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000324496042818678,
      "loss": 4.9525,
      "mean_token_accuracy": 0.20949915647506714,
      "num_tokens": 10960786.0,
      "step": 5295
    },
    {
      "entropy": 5.427356719970703,
      "epoch": 4.814175374829623,
      "grad_norm": 1.203125,
      "learning_rate": 0.00032415121038189746,
      "loss": 4.8865,
      "mean_token_accuracy": 0.20903180837631224,
      "num_tokens": 10970423.0,
      "step": 5300
    },
    {
      "entropy": 5.391693353652954,
      "epoch": 4.818718764198092,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00032380625666939125,
      "loss": 4.9121,
      "mean_token_accuracy": 0.2097949966788292,
      "num_tokens": 10980196.0,
      "step": 5305
    },
    {
      "entropy": 5.398657369613647,
      "epoch": 4.823262153566561,
      "grad_norm": 1.171875,
      "learning_rate": 0.00032346118253229834,
      "loss": 4.9366,
      "mean_token_accuracy": 0.20603130608797074,
      "num_tokens": 10989857.0,
      "step": 5310
    },
    {
      "entropy": 5.424055480957032,
      "epoch": 4.82780554293503,
      "grad_norm": 1.15625,
      "learning_rate": 0.00032311598882205496,
      "loss": 4.9236,
      "mean_token_accuracy": 0.20240387320518494,
      "num_tokens": 11000276.0,
      "step": 5315
    },
    {
      "entropy": 5.433774328231811,
      "epoch": 4.832348932303498,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00032277067639039216,
      "loss": 4.9783,
      "mean_token_accuracy": 0.19645922333002092,
      "num_tokens": 11012502.0,
      "step": 5320
    },
    {
      "entropy": 5.426474905014038,
      "epoch": 4.836892321671967,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00032242524608933423,
      "loss": 4.8579,
      "mean_token_accuracy": 0.2027639552950859,
      "num_tokens": 11023042.0,
      "step": 5325
    },
    {
      "entropy": 5.3796224117279055,
      "epoch": 4.841435711040436,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00032207969877119587,
      "loss": 4.8828,
      "mean_token_accuracy": 0.21613171845674514,
      "num_tokens": 11032259.0,
      "step": 5330
    },
    {
      "entropy": 5.492548513412475,
      "epoch": 4.845979100408905,
      "grad_norm": 1.0625,
      "learning_rate": 0.00032173403528858093,
      "loss": 5.0329,
      "mean_token_accuracy": 0.2018111914396286,
      "num_tokens": 11042324.0,
      "step": 5335
    },
    {
      "entropy": 5.4306488037109375,
      "epoch": 4.850522489777374,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003213882564943796,
      "loss": 4.9238,
      "mean_token_accuracy": 0.2109948992729187,
      "num_tokens": 11053519.0,
      "step": 5340
    },
    {
      "entropy": 5.418545913696289,
      "epoch": 4.855065879145843,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003210423632417666,
      "loss": 4.9141,
      "mean_token_accuracy": 0.21067829728126525,
      "num_tokens": 11063413.0,
      "step": 5345
    },
    {
      "entropy": 5.318604040145874,
      "epoch": 4.859609268514312,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003206963563841993,
      "loss": 4.8731,
      "mean_token_accuracy": 0.205884213745594,
      "num_tokens": 11073232.0,
      "step": 5350
    },
    {
      "entropy": 5.352860450744629,
      "epoch": 4.86415265788278,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00032035023677541515,
      "loss": 4.8455,
      "mean_token_accuracy": 0.20415829569101335,
      "num_tokens": 11083408.0,
      "step": 5355
    },
    {
      "entropy": 5.4700439929962155,
      "epoch": 4.868696047251249,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003200040052694299,
      "loss": 4.9651,
      "mean_token_accuracy": 0.20283088833093643,
      "num_tokens": 11094125.0,
      "step": 5360
    },
    {
      "entropy": 5.4241174221038815,
      "epoch": 4.873239436619718,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00031965766272053534,
      "loss": 4.9412,
      "mean_token_accuracy": 0.20353047996759416,
      "num_tokens": 11106293.0,
      "step": 5365
    },
    {
      "entropy": 5.364050722122192,
      "epoch": 4.877782825988187,
      "grad_norm": 1.09375,
      "learning_rate": 0.00031931120998329737,
      "loss": 4.8424,
      "mean_token_accuracy": 0.2169603317975998,
      "num_tokens": 11116485.0,
      "step": 5370
    },
    {
      "entropy": 5.44035005569458,
      "epoch": 4.882326215356656,
      "grad_norm": 1.0,
      "learning_rate": 0.0003189646479125536,
      "loss": 4.9982,
      "mean_token_accuracy": 0.19936336427927018,
      "num_tokens": 11127876.0,
      "step": 5375
    },
    {
      "entropy": 5.501359367370606,
      "epoch": 4.886869604725125,
      "grad_norm": 1.125,
      "learning_rate": 0.0003186179773634115,
      "loss": 5.0463,
      "mean_token_accuracy": 0.20278658866882324,
      "num_tokens": 11138871.0,
      "step": 5380
    },
    {
      "entropy": 5.440841865539551,
      "epoch": 4.891412994093594,
      "grad_norm": 1.09375,
      "learning_rate": 0.00031827119919124614,
      "loss": 4.9371,
      "mean_token_accuracy": 0.20398538410663605,
      "num_tokens": 11148872.0,
      "step": 5385
    },
    {
      "entropy": 5.342106628417969,
      "epoch": 4.895956383462063,
      "grad_norm": 1.046875,
      "learning_rate": 0.00031792431425169826,
      "loss": 4.8107,
      "mean_token_accuracy": 0.21693755984306334,
      "num_tokens": 11159716.0,
      "step": 5390
    },
    {
      "entropy": 5.471890592575074,
      "epoch": 4.900499772830532,
      "grad_norm": 1.25,
      "learning_rate": 0.00031757732340067195,
      "loss": 4.895,
      "mean_token_accuracy": 0.21031888723373413,
      "num_tokens": 11169187.0,
      "step": 5395
    },
    {
      "entropy": 5.463786363601685,
      "epoch": 4.905043162199,
      "grad_norm": 1.140625,
      "learning_rate": 0.00031723022749433267,
      "loss": 4.8643,
      "mean_token_accuracy": 0.2104683294892311,
      "num_tokens": 11179889.0,
      "step": 5400
    },
    {
      "entropy": 5.454077625274659,
      "epoch": 4.909586551567469,
      "grad_norm": 1.234375,
      "learning_rate": 0.000316883027389105,
      "loss": 5.0353,
      "mean_token_accuracy": 0.20053693801164627,
      "num_tokens": 11189439.0,
      "step": 5405
    },
    {
      "entropy": 5.400667285919189,
      "epoch": 4.914129940935938,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003165357239416706,
      "loss": 4.9649,
      "mean_token_accuracy": 0.2015451028943062,
      "num_tokens": 11200834.0,
      "step": 5410
    },
    {
      "entropy": 5.423655796051025,
      "epoch": 4.918673330304407,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003161883180089665,
      "loss": 4.8818,
      "mean_token_accuracy": 0.21383170634508133,
      "num_tokens": 11210475.0,
      "step": 5415
    },
    {
      "entropy": 5.440661478042602,
      "epoch": 4.923216719672876,
      "grad_norm": 1.125,
      "learning_rate": 0.0003158408104481821,
      "loss": 4.9406,
      "mean_token_accuracy": 0.1992870092391968,
      "num_tokens": 11220744.0,
      "step": 5420
    },
    {
      "entropy": 5.476654720306397,
      "epoch": 4.927760109041345,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00031549320211675776,
      "loss": 5.0019,
      "mean_token_accuracy": 0.19777682721614837,
      "num_tokens": 11231108.0,
      "step": 5425
    },
    {
      "entropy": 5.4110688209533695,
      "epoch": 4.932303498409814,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00031514549387238246,
      "loss": 4.9622,
      "mean_token_accuracy": 0.20619925260543823,
      "num_tokens": 11241160.0,
      "step": 5430
    },
    {
      "entropy": 5.421690893173218,
      "epoch": 4.936846887778283,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003147976865729918,
      "loss": 4.9884,
      "mean_token_accuracy": 0.19896900802850723,
      "num_tokens": 11251711.0,
      "step": 5435
    },
    {
      "entropy": 5.423631954193115,
      "epoch": 4.941390277146752,
      "grad_norm": 1.078125,
      "learning_rate": 0.00031444978107676567,
      "loss": 4.9345,
      "mean_token_accuracy": 0.20540334433317184,
      "num_tokens": 11263025.0,
      "step": 5440
    },
    {
      "entropy": 5.4044641017913815,
      "epoch": 4.94593366651522,
      "grad_norm": 1.09375,
      "learning_rate": 0.00031410177824212635,
      "loss": 4.9239,
      "mean_token_accuracy": 0.20599578320980072,
      "num_tokens": 11273927.0,
      "step": 5445
    },
    {
      "entropy": 5.408950901031494,
      "epoch": 4.950477055883689,
      "grad_norm": 1.125,
      "learning_rate": 0.0003137536789277362,
      "loss": 4.9247,
      "mean_token_accuracy": 0.21180832087993623,
      "num_tokens": 11284108.0,
      "step": 5450
    },
    {
      "entropy": 5.428660774230957,
      "epoch": 4.955020445252158,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003134054839924957,
      "loss": 4.8384,
      "mean_token_accuracy": 0.21101424545049668,
      "num_tokens": 11294530.0,
      "step": 5455
    },
    {
      "entropy": 5.371351480484009,
      "epoch": 4.959563834620627,
      "grad_norm": 1.078125,
      "learning_rate": 0.00031305719429554124,
      "loss": 4.8923,
      "mean_token_accuracy": 0.20294477939605712,
      "num_tokens": 11304803.0,
      "step": 5460
    },
    {
      "entropy": 5.396196746826172,
      "epoch": 4.964107223989096,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003127088106962429,
      "loss": 4.9628,
      "mean_token_accuracy": 0.2048162415623665,
      "num_tokens": 11314567.0,
      "step": 5465
    },
    {
      "entropy": 5.393624639511108,
      "epoch": 4.968650613357565,
      "grad_norm": 1.03125,
      "learning_rate": 0.00031236033405420284,
      "loss": 4.9349,
      "mean_token_accuracy": 0.20752957016229628,
      "num_tokens": 11324452.0,
      "step": 5470
    },
    {
      "entropy": 5.474193096160889,
      "epoch": 4.973194002726034,
      "grad_norm": 1.109375,
      "learning_rate": 0.00031201176522925236,
      "loss": 4.9269,
      "mean_token_accuracy": 0.20911700576543807,
      "num_tokens": 11334930.0,
      "step": 5475
    },
    {
      "entropy": 5.374194717407226,
      "epoch": 4.977737392094502,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003116631050814503,
      "loss": 4.8508,
      "mean_token_accuracy": 0.21221814602613448,
      "num_tokens": 11345487.0,
      "step": 5480
    },
    {
      "entropy": 5.433488512039185,
      "epoch": 4.982280781462971,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00031131435447108116,
      "loss": 4.9379,
      "mean_token_accuracy": 0.20952683240175246,
      "num_tokens": 11355150.0,
      "step": 5485
    },
    {
      "entropy": 5.432433652877807,
      "epoch": 4.98682417083144,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003109655142586521,
      "loss": 4.8987,
      "mean_token_accuracy": 0.21029337197542192,
      "num_tokens": 11366642.0,
      "step": 5490
    },
    {
      "entropy": 5.416983222961425,
      "epoch": 4.991367560199909,
      "grad_norm": 1.109375,
      "learning_rate": 0.00031061658530489195,
      "loss": 4.9275,
      "mean_token_accuracy": 0.20745221078395842,
      "num_tokens": 11377229.0,
      "step": 5495
    },
    {
      "entropy": 5.387570095062256,
      "epoch": 4.995910949568378,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00031026756847074804,
      "loss": 4.9731,
      "mean_token_accuracy": 0.19972439855337143,
      "num_tokens": 11387444.0,
      "step": 5500
    },
    {
      "epoch": 4.995910949568378,
      "eval_entropy": 5.293585781718409,
      "eval_loss": 5.654120445251465,
      "eval_mean_token_accuracy": 0.1730309786491616,
      "eval_num_tokens": 11387444.0,
      "eval_runtime": 2.1986,
      "eval_samples_per_second": 1562.364,
      "eval_steps_per_second": 195.58,
      "step": 5500
    },
    {
      "entropy": 5.442888100941976,
      "epoch": 5.0,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0003099184646173849,
      "loss": 4.9194,
      "mean_token_accuracy": 0.2147990564505259,
      "num_tokens": 11396985.0,
      "step": 5505
    },
    {
      "entropy": 5.435710906982422,
      "epoch": 5.004543389368469,
      "grad_norm": 1.03125,
      "learning_rate": 0.00030956927460618126,
      "loss": 4.7148,
      "mean_token_accuracy": 0.22122498005628585,
      "num_tokens": 11408747.0,
      "step": 5510
    },
    {
      "entropy": 5.386406993865966,
      "epoch": 5.009086778736938,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003092199992987291,
      "loss": 4.6449,
      "mean_token_accuracy": 0.22442148476839066,
      "num_tokens": 11418695.0,
      "step": 5515
    },
    {
      "entropy": 5.333002519607544,
      "epoch": 5.013630168105407,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00030887063955683044,
      "loss": 4.7192,
      "mean_token_accuracy": 0.22673683613538742,
      "num_tokens": 11428143.0,
      "step": 5520
    },
    {
      "entropy": 5.409880685806274,
      "epoch": 5.018173557473876,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00030852119624249553,
      "loss": 4.756,
      "mean_token_accuracy": 0.22064421623945235,
      "num_tokens": 11439610.0,
      "step": 5525
    },
    {
      "entropy": 5.456471872329712,
      "epoch": 5.022716946842344,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003081716702179412,
      "loss": 4.7258,
      "mean_token_accuracy": 0.22147823721170426,
      "num_tokens": 11450035.0,
      "step": 5530
    },
    {
      "entropy": 5.34367880821228,
      "epoch": 5.027260336210813,
      "grad_norm": 1.203125,
      "learning_rate": 0.000307822062345588,
      "loss": 4.686,
      "mean_token_accuracy": 0.2318791389465332,
      "num_tokens": 11460736.0,
      "step": 5535
    },
    {
      "entropy": 5.376640892028808,
      "epoch": 5.031803725579282,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003074723734880588,
      "loss": 4.682,
      "mean_token_accuracy": 0.22599183470010759,
      "num_tokens": 11471319.0,
      "step": 5540
    },
    {
      "entropy": 5.337324857711792,
      "epoch": 5.036347114947751,
      "grad_norm": 1.1875,
      "learning_rate": 0.000307122604508176,
      "loss": 4.6837,
      "mean_token_accuracy": 0.22623700201511382,
      "num_tokens": 11481052.0,
      "step": 5545
    },
    {
      "entropy": 5.36128830909729,
      "epoch": 5.04089050431622,
      "grad_norm": 1.125,
      "learning_rate": 0.0003067727562689598,
      "loss": 4.6835,
      "mean_token_accuracy": 0.22055497914552688,
      "num_tokens": 11491166.0,
      "step": 5550
    },
    {
      "entropy": 5.385983419418335,
      "epoch": 5.045433893684689,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003064228296336259,
      "loss": 4.6736,
      "mean_token_accuracy": 0.22627367526292802,
      "num_tokens": 11501007.0,
      "step": 5555
    },
    {
      "entropy": 5.377386856079101,
      "epoch": 5.049977283053158,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003060728254655836,
      "loss": 4.747,
      "mean_token_accuracy": 0.21813812255859374,
      "num_tokens": 11512197.0,
      "step": 5560
    },
    {
      "entropy": 5.436225843429566,
      "epoch": 5.054520672421627,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00030572274462843327,
      "loss": 4.7326,
      "mean_token_accuracy": 0.2193930059671402,
      "num_tokens": 11522396.0,
      "step": 5565
    },
    {
      "entropy": 5.4139075756073,
      "epoch": 5.059064061790095,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00030537258798596465,
      "loss": 4.7478,
      "mean_token_accuracy": 0.22213096618652345,
      "num_tokens": 11533034.0,
      "step": 5570
    },
    {
      "entropy": 5.476143407821655,
      "epoch": 5.063607451158564,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003050223564021543,
      "loss": 4.7549,
      "mean_token_accuracy": 0.22342167049646378,
      "num_tokens": 11543281.0,
      "step": 5575
    },
    {
      "entropy": 5.366095972061157,
      "epoch": 5.068150840527033,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003046720507411641,
      "loss": 4.6879,
      "mean_token_accuracy": 0.22973687946796417,
      "num_tokens": 11552905.0,
      "step": 5580
    },
    {
      "entropy": 5.352806329727173,
      "epoch": 5.072694229895502,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003043216718673383,
      "loss": 4.7169,
      "mean_token_accuracy": 0.2277402937412262,
      "num_tokens": 11562397.0,
      "step": 5585
    },
    {
      "entropy": 5.444687557220459,
      "epoch": 5.077237619263971,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00030397122064520186,
      "loss": 4.7831,
      "mean_token_accuracy": 0.21618370413780214,
      "num_tokens": 11573375.0,
      "step": 5590
    },
    {
      "entropy": 5.400043296813965,
      "epoch": 5.08178100863244,
      "grad_norm": 1.171875,
      "learning_rate": 0.00030362069793945856,
      "loss": 4.7071,
      "mean_token_accuracy": 0.22284533977508544,
      "num_tokens": 11584155.0,
      "step": 5595
    },
    {
      "entropy": 5.350366592407227,
      "epoch": 5.086324398000909,
      "grad_norm": 1.25,
      "learning_rate": 0.0003032701046149881,
      "loss": 4.7018,
      "mean_token_accuracy": 0.2222864806652069,
      "num_tokens": 11593929.0,
      "step": 5600
    },
    {
      "entropy": 5.397481918334961,
      "epoch": 5.090867787369378,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003029194415368448,
      "loss": 4.7307,
      "mean_token_accuracy": 0.22585897892713547,
      "num_tokens": 11604891.0,
      "step": 5605
    },
    {
      "entropy": 5.425997066497803,
      "epoch": 5.095411176737846,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003025687095702548,
      "loss": 4.7746,
      "mean_token_accuracy": 0.21371407359838485,
      "num_tokens": 11616377.0,
      "step": 5610
    },
    {
      "entropy": 5.332108736038208,
      "epoch": 5.099954566106315,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003022179095806146,
      "loss": 4.7419,
      "mean_token_accuracy": 0.2166347920894623,
      "num_tokens": 11627079.0,
      "step": 5615
    },
    {
      "entropy": 5.349233818054199,
      "epoch": 5.104497955474784,
      "grad_norm": 1.1875,
      "learning_rate": 0.000301867042433488,
      "loss": 4.7114,
      "mean_token_accuracy": 0.22112816274166108,
      "num_tokens": 11638014.0,
      "step": 5620
    },
    {
      "entropy": 5.4628486156463625,
      "epoch": 5.109041344843253,
      "grad_norm": 1.109375,
      "learning_rate": 0.00030151610899460505,
      "loss": 4.8204,
      "mean_token_accuracy": 0.22456971555948257,
      "num_tokens": 11648759.0,
      "step": 5625
    },
    {
      "entropy": 5.38982663154602,
      "epoch": 5.113584734211722,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00030116511012985916,
      "loss": 4.7312,
      "mean_token_accuracy": 0.23063601702451705,
      "num_tokens": 11659050.0,
      "step": 5630
    },
    {
      "entropy": 5.412549161911011,
      "epoch": 5.118128123580191,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003008140467053048,
      "loss": 4.7708,
      "mean_token_accuracy": 0.2194770395755768,
      "num_tokens": 11670131.0,
      "step": 5635
    },
    {
      "entropy": 5.327593898773193,
      "epoch": 5.12267151294866,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003004629195871565,
      "loss": 4.6695,
      "mean_token_accuracy": 0.2264173671603203,
      "num_tokens": 11680564.0,
      "step": 5640
    },
    {
      "entropy": 5.450533676147461,
      "epoch": 5.127214902317129,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003001117296417852,
      "loss": 4.7991,
      "mean_token_accuracy": 0.2178507477045059,
      "num_tokens": 11691910.0,
      "step": 5645
    },
    {
      "entropy": 5.339080142974853,
      "epoch": 5.131758291685597,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002997604777357173,
      "loss": 4.7334,
      "mean_token_accuracy": 0.22220585495233536,
      "num_tokens": 11702174.0,
      "step": 5650
    },
    {
      "entropy": 5.42990345954895,
      "epoch": 5.136301681054066,
      "grad_norm": 1.125,
      "learning_rate": 0.000299409164735632,
      "loss": 4.8529,
      "mean_token_accuracy": 0.2145821273326874,
      "num_tokens": 11712841.0,
      "step": 5655
    },
    {
      "entropy": 5.440341329574585,
      "epoch": 5.140845070422535,
      "grad_norm": 1.25,
      "learning_rate": 0.0002990577915083591,
      "loss": 4.7416,
      "mean_token_accuracy": 0.2192390739917755,
      "num_tokens": 11722781.0,
      "step": 5660
    },
    {
      "entropy": 5.332843685150147,
      "epoch": 5.145388459791004,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00029870635892087716,
      "loss": 4.6345,
      "mean_token_accuracy": 0.2324478283524513,
      "num_tokens": 11732774.0,
      "step": 5665
    },
    {
      "entropy": 5.3430047035217285,
      "epoch": 5.149931849159473,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002983548678403112,
      "loss": 4.7385,
      "mean_token_accuracy": 0.22232196033000945,
      "num_tokens": 11743308.0,
      "step": 5670
    },
    {
      "entropy": 5.408922386169434,
      "epoch": 5.154475238527942,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002980033191339305,
      "loss": 4.8044,
      "mean_token_accuracy": 0.2151191607117653,
      "num_tokens": 11754572.0,
      "step": 5675
    },
    {
      "entropy": 5.3874105453491214,
      "epoch": 5.159018627896411,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002976517136691465,
      "loss": 4.7355,
      "mean_token_accuracy": 0.21856366246938705,
      "num_tokens": 11764406.0,
      "step": 5680
    },
    {
      "entropy": 5.339394807815552,
      "epoch": 5.16356201726488,
      "grad_norm": 1.171875,
      "learning_rate": 0.00029730005231351074,
      "loss": 4.7166,
      "mean_token_accuracy": 0.22120877206325532,
      "num_tokens": 11774121.0,
      "step": 5685
    },
    {
      "entropy": 5.400779771804809,
      "epoch": 5.168105406633348,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002969483359347127,
      "loss": 4.7568,
      "mean_token_accuracy": 0.22426994740962983,
      "num_tokens": 11784820.0,
      "step": 5690
    },
    {
      "entropy": 5.3218421459198,
      "epoch": 5.172648796001817,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002965965654005776,
      "loss": 4.7229,
      "mean_token_accuracy": 0.22579770982265474,
      "num_tokens": 11794891.0,
      "step": 5695
    },
    {
      "entropy": 5.375887489318847,
      "epoch": 5.177192185370286,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002962447415790642,
      "loss": 4.6878,
      "mean_token_accuracy": 0.2253068521618843,
      "num_tokens": 11804842.0,
      "step": 5700
    },
    {
      "entropy": 5.373139095306397,
      "epoch": 5.181735574738755,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00029589286533826276,
      "loss": 4.7427,
      "mean_token_accuracy": 0.22268518805503845,
      "num_tokens": 11815893.0,
      "step": 5705
    },
    {
      "entropy": 5.316330146789551,
      "epoch": 5.186278964107224,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002955409375463931,
      "loss": 4.6683,
      "mean_token_accuracy": 0.23021786510944367,
      "num_tokens": 11826332.0,
      "step": 5710
    },
    {
      "entropy": 5.384970235824585,
      "epoch": 5.190822353475693,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002951889590718019,
      "loss": 4.7907,
      "mean_token_accuracy": 0.2207020565867424,
      "num_tokens": 11836977.0,
      "step": 5715
    },
    {
      "entropy": 5.3728169918060305,
      "epoch": 5.195365742844162,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002948369307829612,
      "loss": 4.6996,
      "mean_token_accuracy": 0.22452046871185302,
      "num_tokens": 11846483.0,
      "step": 5720
    },
    {
      "entropy": 5.405947637557984,
      "epoch": 5.199909132212631,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002944848535484656,
      "loss": 4.753,
      "mean_token_accuracy": 0.22051024585962295,
      "num_tokens": 11857055.0,
      "step": 5725
    },
    {
      "entropy": 5.310382318496704,
      "epoch": 5.2044525215811,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00029413272823703085,
      "loss": 4.6595,
      "mean_token_accuracy": 0.2269337445497513,
      "num_tokens": 11866929.0,
      "step": 5730
    },
    {
      "entropy": 5.346225929260254,
      "epoch": 5.208995910949568,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002937805557174911,
      "loss": 4.7624,
      "mean_token_accuracy": 0.21509087979793548,
      "num_tokens": 11875973.0,
      "step": 5735
    },
    {
      "entropy": 5.403276205062866,
      "epoch": 5.213539300318037,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00029342833685879713,
      "loss": 4.7667,
      "mean_token_accuracy": 0.23103353530168533,
      "num_tokens": 11886273.0,
      "step": 5740
    },
    {
      "entropy": 5.484728813171387,
      "epoch": 5.218082689686506,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002930760725300137,
      "loss": 4.7899,
      "mean_token_accuracy": 0.22288512140512468,
      "num_tokens": 11897462.0,
      "step": 5745
    },
    {
      "entropy": 5.384399032592773,
      "epoch": 5.222626079054975,
      "grad_norm": 1.296875,
      "learning_rate": 0.00029272376360031833,
      "loss": 4.7428,
      "mean_token_accuracy": 0.2207630231976509,
      "num_tokens": 11908116.0,
      "step": 5750
    },
    {
      "entropy": 5.383107328414917,
      "epoch": 5.227169468423444,
      "grad_norm": 1.140625,
      "learning_rate": 0.00029237141093899804,
      "loss": 4.8046,
      "mean_token_accuracy": 0.21564611792564392,
      "num_tokens": 11918990.0,
      "step": 5755
    },
    {
      "entropy": 5.453704929351806,
      "epoch": 5.231712857791913,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00029201901541544824,
      "loss": 4.7865,
      "mean_token_accuracy": 0.223606576025486,
      "num_tokens": 11929201.0,
      "step": 5760
    },
    {
      "entropy": 5.35169792175293,
      "epoch": 5.236256247160382,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002916665778991696,
      "loss": 4.7547,
      "mean_token_accuracy": 0.21328990906476974,
      "num_tokens": 11939002.0,
      "step": 5765
    },
    {
      "entropy": 5.328910970687867,
      "epoch": 5.240799636528851,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002913140992597668,
      "loss": 4.6438,
      "mean_token_accuracy": 0.22596994191408157,
      "num_tokens": 11949255.0,
      "step": 5770
    },
    {
      "entropy": 5.409477567672729,
      "epoch": 5.245343025897319,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002909615803669458,
      "loss": 4.7095,
      "mean_token_accuracy": 0.22509807646274566,
      "num_tokens": 11959108.0,
      "step": 5775
    },
    {
      "entropy": 5.2694422721862795,
      "epoch": 5.249886415265788,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002906090220905119,
      "loss": 4.6396,
      "mean_token_accuracy": 0.22823198288679122,
      "num_tokens": 11969196.0,
      "step": 5780
    },
    {
      "entropy": 5.268312454223633,
      "epoch": 5.254429804634257,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002902564253003677,
      "loss": 4.6728,
      "mean_token_accuracy": 0.2274135544896126,
      "num_tokens": 11979101.0,
      "step": 5785
    },
    {
      "entropy": 5.38636155128479,
      "epoch": 5.258973194002726,
      "grad_norm": 1.25,
      "learning_rate": 0.0002899037908665106,
      "loss": 4.735,
      "mean_token_accuracy": 0.22178169041872026,
      "num_tokens": 11988396.0,
      "step": 5790
    },
    {
      "entropy": 5.390290355682373,
      "epoch": 5.263516583371195,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002895511196590312,
      "loss": 4.7203,
      "mean_token_accuracy": 0.2280610218644142,
      "num_tokens": 11998355.0,
      "step": 5795
    },
    {
      "entropy": 5.334471940994263,
      "epoch": 5.268059972739664,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002891984125481104,
      "loss": 4.7202,
      "mean_token_accuracy": 0.22561107575893402,
      "num_tokens": 12008938.0,
      "step": 5800
    },
    {
      "entropy": 5.372486114501953,
      "epoch": 5.272603362108133,
      "grad_norm": 1.0625,
      "learning_rate": 0.00028884567040401814,
      "loss": 4.7798,
      "mean_token_accuracy": 0.21562401801347733,
      "num_tokens": 12019875.0,
      "step": 5805
    },
    {
      "entropy": 5.410190057754517,
      "epoch": 5.277146751476602,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002884928940971105,
      "loss": 4.7413,
      "mean_token_accuracy": 0.22387544959783554,
      "num_tokens": 12029368.0,
      "step": 5810
    },
    {
      "entropy": 5.354491376876831,
      "epoch": 5.28169014084507,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000288140084497828,
      "loss": 4.7029,
      "mean_token_accuracy": 0.22088606059551238,
      "num_tokens": 12040191.0,
      "step": 5815
    },
    {
      "entropy": 5.320126295089722,
      "epoch": 5.286233530213539,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002877872424766932,
      "loss": 4.7044,
      "mean_token_accuracy": 0.23195876479148864,
      "num_tokens": 12050087.0,
      "step": 5820
    },
    {
      "entropy": 5.297035121917725,
      "epoch": 5.290776919582008,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002874343689043086,
      "loss": 4.7327,
      "mean_token_accuracy": 0.2318143665790558,
      "num_tokens": 12061014.0,
      "step": 5825
    },
    {
      "entropy": 5.374315309524536,
      "epoch": 5.295320308950477,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000287081464651355,
      "loss": 4.7361,
      "mean_token_accuracy": 0.22517008185386658,
      "num_tokens": 12071055.0,
      "step": 5830
    },
    {
      "entropy": 5.335392141342163,
      "epoch": 5.299863698318946,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002867285305885882,
      "loss": 4.7458,
      "mean_token_accuracy": 0.22201600968837737,
      "num_tokens": 12081323.0,
      "step": 5835
    },
    {
      "entropy": 5.304869747161865,
      "epoch": 5.304407087687415,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00028637556758683816,
      "loss": 4.7008,
      "mean_token_accuracy": 0.2302267774939537,
      "num_tokens": 12091482.0,
      "step": 5840
    },
    {
      "entropy": 5.273191595077515,
      "epoch": 5.308950477055884,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002860225765170058,
      "loss": 4.6356,
      "mean_token_accuracy": 0.23417919576168061,
      "num_tokens": 12101960.0,
      "step": 5845
    },
    {
      "entropy": 5.397185850143432,
      "epoch": 5.313493866424353,
      "grad_norm": 1.28125,
      "learning_rate": 0.00028566955825006174,
      "loss": 4.7406,
      "mean_token_accuracy": 0.2199667513370514,
      "num_tokens": 12111654.0,
      "step": 5850
    },
    {
      "entropy": 5.4260204315185545,
      "epoch": 5.318037255792821,
      "grad_norm": 1.0625,
      "learning_rate": 0.00028531651365704325,
      "loss": 4.816,
      "mean_token_accuracy": 0.2194522961974144,
      "num_tokens": 12123096.0,
      "step": 5855
    },
    {
      "entropy": 5.316342401504516,
      "epoch": 5.32258064516129,
      "grad_norm": 1.109375,
      "learning_rate": 0.00028496344360905287,
      "loss": 4.7243,
      "mean_token_accuracy": 0.21936907917261123,
      "num_tokens": 12133064.0,
      "step": 5860
    },
    {
      "entropy": 5.3128745555877686,
      "epoch": 5.327124034529759,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00028461034897725587,
      "loss": 4.7412,
      "mean_token_accuracy": 0.22360700219869614,
      "num_tokens": 12144386.0,
      "step": 5865
    },
    {
      "entropy": 5.383149003982544,
      "epoch": 5.331667423898228,
      "grad_norm": 1.0625,
      "learning_rate": 0.00028425723063287803,
      "loss": 4.7907,
      "mean_token_accuracy": 0.2185806304216385,
      "num_tokens": 12156111.0,
      "step": 5870
    },
    {
      "entropy": 5.4445630550384525,
      "epoch": 5.336210813266697,
      "grad_norm": 1.234375,
      "learning_rate": 0.00028390408944720385,
      "loss": 4.8268,
      "mean_token_accuracy": 0.20818151384592057,
      "num_tokens": 12166462.0,
      "step": 5875
    },
    {
      "entropy": 5.341730403900146,
      "epoch": 5.340754202635166,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002835509262915741,
      "loss": 4.8555,
      "mean_token_accuracy": 0.20754481106996536,
      "num_tokens": 12178635.0,
      "step": 5880
    },
    {
      "entropy": 5.316912221908569,
      "epoch": 5.345297592003635,
      "grad_norm": 1.15625,
      "learning_rate": 0.00028319774203738375,
      "loss": 4.6903,
      "mean_token_accuracy": 0.22445664405822754,
      "num_tokens": 12188880.0,
      "step": 5885
    },
    {
      "entropy": 5.331257295608521,
      "epoch": 5.349840981372104,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002828445375560798,
      "loss": 4.711,
      "mean_token_accuracy": 0.21752217710018157,
      "num_tokens": 12199733.0,
      "step": 5890
    },
    {
      "entropy": 5.334118223190307,
      "epoch": 5.354384370740572,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00028249131371915927,
      "loss": 4.7598,
      "mean_token_accuracy": 0.21755870282649994,
      "num_tokens": 12209925.0,
      "step": 5895
    },
    {
      "entropy": 5.471984767913819,
      "epoch": 5.358927760109041,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002821380713981668,
      "loss": 4.8015,
      "mean_token_accuracy": 0.22108588367700577,
      "num_tokens": 12219870.0,
      "step": 5900
    },
    {
      "entropy": 5.3149021625518795,
      "epoch": 5.36347114947751,
      "grad_norm": 1.25,
      "learning_rate": 0.00028178481146469273,
      "loss": 4.7254,
      "mean_token_accuracy": 0.2290690526366234,
      "num_tokens": 12230295.0,
      "step": 5905
    },
    {
      "entropy": 5.382998514175415,
      "epoch": 5.368014538845979,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002814315347903709,
      "loss": 4.744,
      "mean_token_accuracy": 0.2276427775621414,
      "num_tokens": 12239700.0,
      "step": 5910
    },
    {
      "entropy": 5.408007049560547,
      "epoch": 5.372557928214448,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002810782422468763,
      "loss": 4.793,
      "mean_token_accuracy": 0.21218898743391038,
      "num_tokens": 12249529.0,
      "step": 5915
    },
    {
      "entropy": 5.374137210845947,
      "epoch": 5.377101317582917,
      "grad_norm": 1.25,
      "learning_rate": 0.00028072493470592317,
      "loss": 4.7003,
      "mean_token_accuracy": 0.2291147768497467,
      "num_tokens": 12259774.0,
      "step": 5920
    },
    {
      "entropy": 5.379475259780884,
      "epoch": 5.381644706951386,
      "grad_norm": 1.125,
      "learning_rate": 0.00028037161303926265,
      "loss": 4.7352,
      "mean_token_accuracy": 0.22256742268800736,
      "num_tokens": 12270484.0,
      "step": 5925
    },
    {
      "entropy": 5.350165462493896,
      "epoch": 5.386188096319855,
      "grad_norm": 1.21875,
      "learning_rate": 0.000280018278118681,
      "loss": 4.7256,
      "mean_token_accuracy": 0.22498584538698196,
      "num_tokens": 12281195.0,
      "step": 5930
    },
    {
      "entropy": 5.342803716659546,
      "epoch": 5.390731485688324,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000279664930815997,
      "loss": 4.6988,
      "mean_token_accuracy": 0.2238556519150734,
      "num_tokens": 12292052.0,
      "step": 5935
    },
    {
      "entropy": 5.349413824081421,
      "epoch": 5.395274875056792,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002793115720030599,
      "loss": 4.7748,
      "mean_token_accuracy": 0.22351041287183762,
      "num_tokens": 12302562.0,
      "step": 5940
    },
    {
      "entropy": 5.381225538253784,
      "epoch": 5.399818264425261,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002789582025517474,
      "loss": 4.7675,
      "mean_token_accuracy": 0.21973980218172073,
      "num_tokens": 12313875.0,
      "step": 5945
    },
    {
      "entropy": 5.37748646736145,
      "epoch": 5.40436165379373,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002786048233339637,
      "loss": 4.7671,
      "mean_token_accuracy": 0.2165626898407936,
      "num_tokens": 12324500.0,
      "step": 5950
    },
    {
      "entropy": 5.339924716949463,
      "epoch": 5.408905043162199,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00027825143522163675,
      "loss": 4.7482,
      "mean_token_accuracy": 0.21991816610097886,
      "num_tokens": 12335474.0,
      "step": 5955
    },
    {
      "entropy": 5.3630767345428465,
      "epoch": 5.413448432530668,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002778980390867165,
      "loss": 4.7105,
      "mean_token_accuracy": 0.22762758582830428,
      "num_tokens": 12346174.0,
      "step": 5960
    },
    {
      "entropy": 5.386571645736694,
      "epoch": 5.417991821899137,
      "grad_norm": 1.171875,
      "learning_rate": 0.00027754463580117306,
      "loss": 4.7384,
      "mean_token_accuracy": 0.22584440112113952,
      "num_tokens": 12356108.0,
      "step": 5965
    },
    {
      "entropy": 5.366361045837403,
      "epoch": 5.422535211267606,
      "grad_norm": 1.125,
      "learning_rate": 0.00027719122623699353,
      "loss": 4.7798,
      "mean_token_accuracy": 0.22111901938915252,
      "num_tokens": 12366699.0,
      "step": 5970
    },
    {
      "entropy": 5.477452516555786,
      "epoch": 5.427078600636074,
      "grad_norm": 1.125,
      "learning_rate": 0.0002768378112661813,
      "loss": 4.7739,
      "mean_token_accuracy": 0.21721669882535935,
      "num_tokens": 12377225.0,
      "step": 5975
    },
    {
      "entropy": 5.418661403656006,
      "epoch": 5.431621990004543,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002764843917607523,
      "loss": 4.7858,
      "mean_token_accuracy": 0.22524149417877198,
      "num_tokens": 12388936.0,
      "step": 5980
    },
    {
      "entropy": 5.345610761642456,
      "epoch": 5.436165379373012,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002761309685927343,
      "loss": 4.7301,
      "mean_token_accuracy": 0.22234415411949157,
      "num_tokens": 12399130.0,
      "step": 5985
    },
    {
      "entropy": 5.386108636856079,
      "epoch": 5.440708768741481,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002757775426341636,
      "loss": 4.7265,
      "mean_token_accuracy": 0.22986256033182145,
      "num_tokens": 12410181.0,
      "step": 5990
    },
    {
      "entropy": 5.371101236343383,
      "epoch": 5.44525215810995,
      "grad_norm": 1.140625,
      "learning_rate": 0.00027542411475708385,
      "loss": 4.8131,
      "mean_token_accuracy": 0.21388424187898636,
      "num_tokens": 12421173.0,
      "step": 5995
    },
    {
      "entropy": 5.4222126483917235,
      "epoch": 5.449795547478419,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00027507068583354304,
      "loss": 4.8264,
      "mean_token_accuracy": 0.20639564245939254,
      "num_tokens": 12433195.0,
      "step": 6000
    },
    {
      "epoch": 5.449795547478419,
      "eval_entropy": 5.2894156589064485,
      "eval_loss": 5.6739726066589355,
      "eval_mean_token_accuracy": 0.1733105929698362,
      "eval_num_tokens": 12433195.0,
      "eval_runtime": 2.0129,
      "eval_samples_per_second": 1706.468,
      "eval_steps_per_second": 213.619,
      "step": 6000
    },
    {
      "entropy": 5.393551540374756,
      "epoch": 5.454338936846888,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00027471725673559205,
      "loss": 4.7382,
      "mean_token_accuracy": 0.22173366248607634,
      "num_tokens": 12442944.0,
      "step": 6005
    },
    {
      "entropy": 5.324278163909912,
      "epoch": 5.458882326215357,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00027436382833528184,
      "loss": 4.7062,
      "mean_token_accuracy": 0.22851437777280809,
      "num_tokens": 12453133.0,
      "step": 6010
    },
    {
      "entropy": 5.322636365890503,
      "epoch": 5.463425715583826,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002740104015046621,
      "loss": 4.8128,
      "mean_token_accuracy": 0.21134943813085555,
      "num_tokens": 12464606.0,
      "step": 6015
    },
    {
      "entropy": 5.347986173629761,
      "epoch": 5.467969104952294,
      "grad_norm": 1.25,
      "learning_rate": 0.00027365697711577815,
      "loss": 4.7163,
      "mean_token_accuracy": 0.2192281723022461,
      "num_tokens": 12474490.0,
      "step": 6020
    },
    {
      "entropy": 5.363819885253906,
      "epoch": 5.472512494320763,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002733035560406697,
      "loss": 4.7134,
      "mean_token_accuracy": 0.22690630704164505,
      "num_tokens": 12485158.0,
      "step": 6025
    },
    {
      "entropy": 5.282220792770386,
      "epoch": 5.477055883689232,
      "grad_norm": 1.109375,
      "learning_rate": 0.00027295013915136805,
      "loss": 4.7682,
      "mean_token_accuracy": 0.21705720126628875,
      "num_tokens": 12496265.0,
      "step": 6030
    },
    {
      "entropy": 5.312582683563233,
      "epoch": 5.481599273057701,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00027259672731989436,
      "loss": 4.657,
      "mean_token_accuracy": 0.22460297346115113,
      "num_tokens": 12506719.0,
      "step": 6035
    },
    {
      "entropy": 5.397791719436645,
      "epoch": 5.48614266242617,
      "grad_norm": 1.15625,
      "learning_rate": 0.000272243321418257,
      "loss": 4.746,
      "mean_token_accuracy": 0.22697338908910752,
      "num_tokens": 12515990.0,
      "step": 6040
    },
    {
      "entropy": 5.371173667907715,
      "epoch": 5.490686051794639,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002718899223184501,
      "loss": 4.7479,
      "mean_token_accuracy": 0.21903583109378816,
      "num_tokens": 12526557.0,
      "step": 6045
    },
    {
      "entropy": 5.339066696166992,
      "epoch": 5.495229441163108,
      "grad_norm": 1.234375,
      "learning_rate": 0.00027153653089245067,
      "loss": 4.6954,
      "mean_token_accuracy": 0.23325262516736983,
      "num_tokens": 12536657.0,
      "step": 6050
    },
    {
      "entropy": 5.3603599071502686,
      "epoch": 5.499772830531577,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000271183148012217,
      "loss": 4.7101,
      "mean_token_accuracy": 0.22533073723316194,
      "num_tokens": 12546958.0,
      "step": 6055
    },
    {
      "entropy": 5.32589430809021,
      "epoch": 5.504316219900045,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002708297745496861,
      "loss": 4.7277,
      "mean_token_accuracy": 0.2209014117717743,
      "num_tokens": 12556978.0,
      "step": 6060
    },
    {
      "entropy": 5.363377618789673,
      "epoch": 5.508859609268514,
      "grad_norm": 1.15625,
      "learning_rate": 0.00027047641137677213,
      "loss": 4.694,
      "mean_token_accuracy": 0.22603297978639603,
      "num_tokens": 12566954.0,
      "step": 6065
    },
    {
      "entropy": 5.352986812591553,
      "epoch": 5.513402998636983,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002701230593653632,
      "loss": 4.7292,
      "mean_token_accuracy": 0.22671225816011428,
      "num_tokens": 12577129.0,
      "step": 6070
    },
    {
      "entropy": 5.3487790584564205,
      "epoch": 5.517946388005452,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002697697193873207,
      "loss": 4.7362,
      "mean_token_accuracy": 0.21662198305130004,
      "num_tokens": 12586802.0,
      "step": 6075
    },
    {
      "entropy": 5.383929252624512,
      "epoch": 5.522489777373921,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00026941639231447547,
      "loss": 4.7853,
      "mean_token_accuracy": 0.2112003281712532,
      "num_tokens": 12597732.0,
      "step": 6080
    },
    {
      "entropy": 5.3602245330810545,
      "epoch": 5.52703316674239,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002690630790186273,
      "loss": 4.6731,
      "mean_token_accuracy": 0.22521342039108277,
      "num_tokens": 12607185.0,
      "step": 6085
    },
    {
      "entropy": 5.376276588439941,
      "epoch": 5.531576556110859,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00026870978037154137,
      "loss": 4.7647,
      "mean_token_accuracy": 0.21941367536783218,
      "num_tokens": 12616821.0,
      "step": 6090
    },
    {
      "entropy": 5.385237836837769,
      "epoch": 5.536119945479328,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002683564972449472,
      "loss": 4.8038,
      "mean_token_accuracy": 0.21444914788007735,
      "num_tokens": 12626698.0,
      "step": 6095
    },
    {
      "entropy": 5.353044700622559,
      "epoch": 5.540663334847796,
      "grad_norm": 1.375,
      "learning_rate": 0.0002680032305105355,
      "loss": 4.7232,
      "mean_token_accuracy": 0.22827369570732117,
      "num_tokens": 12636588.0,
      "step": 6100
    },
    {
      "entropy": 5.416423702239991,
      "epoch": 5.545206724216265,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002676499810399572,
      "loss": 4.7695,
      "mean_token_accuracy": 0.22056723088026048,
      "num_tokens": 12647196.0,
      "step": 6105
    },
    {
      "entropy": 5.360438585281372,
      "epoch": 5.549750113584734,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00026729674970481985,
      "loss": 4.7552,
      "mean_token_accuracy": 0.22090166360139846,
      "num_tokens": 12658631.0,
      "step": 6110
    },
    {
      "entropy": 5.357418060302734,
      "epoch": 5.554293502953203,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00026694353737668705,
      "loss": 4.7655,
      "mean_token_accuracy": 0.22060160487890243,
      "num_tokens": 12668905.0,
      "step": 6115
    },
    {
      "entropy": 5.333030700683594,
      "epoch": 5.558836892321672,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00026659034492707473,
      "loss": 4.7178,
      "mean_token_accuracy": 0.22780750095844268,
      "num_tokens": 12678270.0,
      "step": 6120
    },
    {
      "entropy": 5.320735168457031,
      "epoch": 5.563380281690141,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002662371732274504,
      "loss": 4.7166,
      "mean_token_accuracy": 0.2272453114390373,
      "num_tokens": 12688248.0,
      "step": 6125
    },
    {
      "entropy": 5.377007627487183,
      "epoch": 5.56792367105861,
      "grad_norm": 1.140625,
      "learning_rate": 0.00026588402314923,
      "loss": 4.8243,
      "mean_token_accuracy": 0.21789301186800003,
      "num_tokens": 12698418.0,
      "step": 6130
    },
    {
      "entropy": 5.286288738250732,
      "epoch": 5.572467060427079,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002655308955637763,
      "loss": 4.6548,
      "mean_token_accuracy": 0.2268152967095375,
      "num_tokens": 12708578.0,
      "step": 6135
    },
    {
      "entropy": 5.332411861419677,
      "epoch": 5.577010449795548,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002651777913423965,
      "loss": 4.781,
      "mean_token_accuracy": 0.21195392608642577,
      "num_tokens": 12718899.0,
      "step": 6140
    },
    {
      "entropy": 5.345312166213989,
      "epoch": 5.581553839164016,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002648247113563402,
      "loss": 4.7032,
      "mean_token_accuracy": 0.21822802871465682,
      "num_tokens": 12728342.0,
      "step": 6145
    },
    {
      "entropy": 5.440801620483398,
      "epoch": 5.586097228532485,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00026447165647679703,
      "loss": 4.7513,
      "mean_token_accuracy": 0.2281322181224823,
      "num_tokens": 12739337.0,
      "step": 6150
    },
    {
      "entropy": 5.345335340499878,
      "epoch": 5.590640617900954,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000264118627574895,
      "loss": 4.747,
      "mean_token_accuracy": 0.22101138532161713,
      "num_tokens": 12749650.0,
      "step": 6155
    },
    {
      "entropy": 5.330507278442383,
      "epoch": 5.595184007269423,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00026376562552169776,
      "loss": 4.809,
      "mean_token_accuracy": 0.21345535069704055,
      "num_tokens": 12760573.0,
      "step": 6160
    },
    {
      "entropy": 5.4569206714630125,
      "epoch": 5.599727396637892,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00026341265118820284,
      "loss": 4.8207,
      "mean_token_accuracy": 0.21778863221406936,
      "num_tokens": 12771070.0,
      "step": 6165
    },
    {
      "entropy": 5.412123823165894,
      "epoch": 5.604270786006361,
      "grad_norm": 1.125,
      "learning_rate": 0.00026305970544533903,
      "loss": 4.75,
      "mean_token_accuracy": 0.2222387373447418,
      "num_tokens": 12781896.0,
      "step": 6170
    },
    {
      "entropy": 5.368235731124878,
      "epoch": 5.60881417537483,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00026270678916396527,
      "loss": 4.7893,
      "mean_token_accuracy": 0.21919185370206834,
      "num_tokens": 12792675.0,
      "step": 6175
    },
    {
      "entropy": 5.3204442977905275,
      "epoch": 5.613357564743298,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00026235390321486705,
      "loss": 4.684,
      "mean_token_accuracy": 0.2302582949399948,
      "num_tokens": 12802371.0,
      "step": 6180
    },
    {
      "entropy": 5.399893665313721,
      "epoch": 5.617900954111767,
      "grad_norm": 1.171875,
      "learning_rate": 0.00026200104846875574,
      "loss": 4.7488,
      "mean_token_accuracy": 0.22201360911130905,
      "num_tokens": 12813281.0,
      "step": 6185
    },
    {
      "entropy": 5.335555982589722,
      "epoch": 5.622444343480236,
      "grad_norm": 1.15625,
      "learning_rate": 0.000261648225796265,
      "loss": 4.7765,
      "mean_token_accuracy": 0.21601221859455108,
      "num_tokens": 12823859.0,
      "step": 6190
    },
    {
      "entropy": 5.348474836349487,
      "epoch": 5.626987732848705,
      "grad_norm": 1.125,
      "learning_rate": 0.0002612954360679499,
      "loss": 4.6767,
      "mean_token_accuracy": 0.23114795088768006,
      "num_tokens": 12833288.0,
      "step": 6195
    },
    {
      "entropy": 5.336826992034912,
      "epoch": 5.631531122217174,
      "grad_norm": 1.296875,
      "learning_rate": 0.000260942680154284,
      "loss": 4.7257,
      "mean_token_accuracy": 0.22137789130210878,
      "num_tokens": 12843698.0,
      "step": 6200
    },
    {
      "entropy": 5.336512756347656,
      "epoch": 5.636074511585643,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002605899589256574,
      "loss": 4.6614,
      "mean_token_accuracy": 0.23357352763414382,
      "num_tokens": 12853956.0,
      "step": 6205
    },
    {
      "entropy": 5.377992057800293,
      "epoch": 5.640617900954112,
      "grad_norm": 1.203125,
      "learning_rate": 0.00026023727325237463,
      "loss": 4.7696,
      "mean_token_accuracy": 0.21965580880641938,
      "num_tokens": 12864680.0,
      "step": 6210
    },
    {
      "entropy": 5.35068998336792,
      "epoch": 5.645161290322581,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002598846240046527,
      "loss": 4.6753,
      "mean_token_accuracy": 0.2214706778526306,
      "num_tokens": 12873790.0,
      "step": 6215
    },
    {
      "entropy": 5.36745867729187,
      "epoch": 5.64970467969105,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002595320120526183,
      "loss": 4.7372,
      "mean_token_accuracy": 0.22803843319416045,
      "num_tokens": 12884559.0,
      "step": 6220
    },
    {
      "entropy": 5.34061975479126,
      "epoch": 5.654248069059518,
      "grad_norm": 1.109375,
      "learning_rate": 0.00025917943826630666,
      "loss": 4.7837,
      "mean_token_accuracy": 0.22199500501155853,
      "num_tokens": 12896176.0,
      "step": 6225
    },
    {
      "entropy": 5.371632862091064,
      "epoch": 5.658791458427987,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002588269035156584,
      "loss": 4.7345,
      "mean_token_accuracy": 0.21303083300590514,
      "num_tokens": 12906381.0,
      "step": 6230
    },
    {
      "entropy": 5.380237293243408,
      "epoch": 5.663334847796456,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00025847440867051796,
      "loss": 4.713,
      "mean_token_accuracy": 0.21560619175434112,
      "num_tokens": 12915738.0,
      "step": 6235
    },
    {
      "entropy": 5.353420639038086,
      "epoch": 5.667878237164925,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002581219546006314,
      "loss": 4.722,
      "mean_token_accuracy": 0.224865560233593,
      "num_tokens": 12926059.0,
      "step": 6240
    },
    {
      "entropy": 5.342691850662232,
      "epoch": 5.672421626533394,
      "grad_norm": 1.25,
      "learning_rate": 0.0002577695421756441,
      "loss": 4.7557,
      "mean_token_accuracy": 0.22004517316818237,
      "num_tokens": 12936730.0,
      "step": 6245
    },
    {
      "entropy": 5.366034507751465,
      "epoch": 5.676965015901863,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002574171722650986,
      "loss": 4.7689,
      "mean_token_accuracy": 0.21681433022022248,
      "num_tokens": 12945687.0,
      "step": 6250
    },
    {
      "entropy": 5.426312589645386,
      "epoch": 5.681508405270332,
      "grad_norm": 1.296875,
      "learning_rate": 0.00025706484573843273,
      "loss": 4.7648,
      "mean_token_accuracy": 0.2270517960190773,
      "num_tokens": 12955087.0,
      "step": 6255
    },
    {
      "entropy": 5.347537612915039,
      "epoch": 5.6860517946388,
      "grad_norm": 1.125,
      "learning_rate": 0.000256712563464977,
      "loss": 4.735,
      "mean_token_accuracy": 0.217491415143013,
      "num_tokens": 12965497.0,
      "step": 6260
    },
    {
      "entropy": 5.43813681602478,
      "epoch": 5.690595184007269,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002563603263139532,
      "loss": 4.8941,
      "mean_token_accuracy": 0.20858675241470337,
      "num_tokens": 12976520.0,
      "step": 6265
    },
    {
      "entropy": 5.342008304595947,
      "epoch": 5.695138573375738,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002560081351544711,
      "loss": 4.7391,
      "mean_token_accuracy": 0.2242509663105011,
      "num_tokens": 12987271.0,
      "step": 6270
    },
    {
      "entropy": 5.353102684020996,
      "epoch": 5.699681962744207,
      "grad_norm": 1.109375,
      "learning_rate": 0.00025565599085552776,
      "loss": 4.7568,
      "mean_token_accuracy": 0.21733236610889434,
      "num_tokens": 12998170.0,
      "step": 6275
    },
    {
      "entropy": 5.343435764312744,
      "epoch": 5.704225352112676,
      "grad_norm": 1.265625,
      "learning_rate": 0.00025530389428600394,
      "loss": 4.7555,
      "mean_token_accuracy": 0.2193263590335846,
      "num_tokens": 13008059.0,
      "step": 6280
    },
    {
      "entropy": 5.353769159317016,
      "epoch": 5.708768741481145,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002549518463146631,
      "loss": 4.7203,
      "mean_token_accuracy": 0.22279399484395981,
      "num_tokens": 13017904.0,
      "step": 6285
    },
    {
      "entropy": 5.419822645187378,
      "epoch": 5.713312130849614,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002545998478101485,
      "loss": 4.7642,
      "mean_token_accuracy": 0.22119873613119126,
      "num_tokens": 13028399.0,
      "step": 6290
    },
    {
      "entropy": 5.371592950820923,
      "epoch": 5.717855520218083,
      "grad_norm": 1.28125,
      "learning_rate": 0.00025424789964098154,
      "loss": 4.7663,
      "mean_token_accuracy": 0.21555964052677154,
      "num_tokens": 13038420.0,
      "step": 6295
    },
    {
      "entropy": 5.446383953094482,
      "epoch": 5.722398909586552,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002538960026755593,
      "loss": 4.7553,
      "mean_token_accuracy": 0.222581684589386,
      "num_tokens": 13048514.0,
      "step": 6300
    },
    {
      "entropy": 5.382921838760376,
      "epoch": 5.72694229895502,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00025354415778215255,
      "loss": 4.8148,
      "mean_token_accuracy": 0.21639999002218246,
      "num_tokens": 13059728.0,
      "step": 6305
    },
    {
      "entropy": 5.3707221984863285,
      "epoch": 5.731485688323489,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00025319236582890364,
      "loss": 4.833,
      "mean_token_accuracy": 0.2158609077334404,
      "num_tokens": 13069535.0,
      "step": 6310
    },
    {
      "entropy": 5.369030714035034,
      "epoch": 5.736029077691958,
      "grad_norm": 1.234375,
      "learning_rate": 0.00025284062768382424,
      "loss": 4.7133,
      "mean_token_accuracy": 0.21941834688186646,
      "num_tokens": 13079012.0,
      "step": 6315
    },
    {
      "entropy": 5.404089832305909,
      "epoch": 5.740572467060427,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002524889442147932,
      "loss": 4.7398,
      "mean_token_accuracy": 0.22943668067455292,
      "num_tokens": 13089067.0,
      "step": 6320
    },
    {
      "entropy": 5.345050382614136,
      "epoch": 5.745115856428896,
      "grad_norm": 1.09375,
      "learning_rate": 0.00025213731628955454,
      "loss": 4.7622,
      "mean_token_accuracy": 0.2183064579963684,
      "num_tokens": 13100576.0,
      "step": 6325
    },
    {
      "entropy": 5.352456855773926,
      "epoch": 5.749659245797365,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00025178574477571525,
      "loss": 4.8999,
      "mean_token_accuracy": 0.208744552731514,
      "num_tokens": 13111926.0,
      "step": 6330
    },
    {
      "entropy": 5.3606664657592775,
      "epoch": 5.754202635165834,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00025143423054074305,
      "loss": 4.7111,
      "mean_token_accuracy": 0.2209509015083313,
      "num_tokens": 13122362.0,
      "step": 6335
    },
    {
      "entropy": 5.345755290985108,
      "epoch": 5.758746024534303,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002510827744519644,
      "loss": 4.714,
      "mean_token_accuracy": 0.224467496573925,
      "num_tokens": 13133372.0,
      "step": 6340
    },
    {
      "entropy": 5.333174562454223,
      "epoch": 5.763289413902772,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00025073137737656226,
      "loss": 4.7191,
      "mean_token_accuracy": 0.22743240147829055,
      "num_tokens": 13143440.0,
      "step": 6345
    },
    {
      "entropy": 5.3817990779876705,
      "epoch": 5.76783280327124,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002503800401815739,
      "loss": 4.7512,
      "mean_token_accuracy": 0.2192935585975647,
      "num_tokens": 13153564.0,
      "step": 6350
    },
    {
      "entropy": 5.378074264526367,
      "epoch": 5.772376192639709,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00025002876373388913,
      "loss": 4.8249,
      "mean_token_accuracy": 0.21348915249109268,
      "num_tokens": 13163679.0,
      "step": 6355
    },
    {
      "entropy": 5.398235559463501,
      "epoch": 5.776919582008178,
      "grad_norm": 1.171875,
      "learning_rate": 0.00024967754890024737,
      "loss": 4.7632,
      "mean_token_accuracy": 0.21926371455192567,
      "num_tokens": 13173930.0,
      "step": 6360
    },
    {
      "entropy": 5.411173152923584,
      "epoch": 5.781462971376647,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002493263965472366,
      "loss": 4.7468,
      "mean_token_accuracy": 0.22991445362567903,
      "num_tokens": 13184805.0,
      "step": 6365
    },
    {
      "entropy": 5.32814736366272,
      "epoch": 5.786006360745116,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002489753075412901,
      "loss": 4.7809,
      "mean_token_accuracy": 0.21886532306671141,
      "num_tokens": 13196229.0,
      "step": 6370
    },
    {
      "entropy": 5.344338369369507,
      "epoch": 5.790549750113585,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00024862428274868533,
      "loss": 4.6855,
      "mean_token_accuracy": 0.2290743887424469,
      "num_tokens": 13205906.0,
      "step": 6375
    },
    {
      "entropy": 5.33489351272583,
      "epoch": 5.795093139482054,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002482733230355408,
      "loss": 4.6984,
      "mean_token_accuracy": 0.22368897944688798,
      "num_tokens": 13215076.0,
      "step": 6380
    },
    {
      "entropy": 5.443288516998291,
      "epoch": 5.799636528850522,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00024792242926781495,
      "loss": 4.8256,
      "mean_token_accuracy": 0.21821932792663573,
      "num_tokens": 13225761.0,
      "step": 6385
    },
    {
      "entropy": 5.404580354690552,
      "epoch": 5.804179918218991,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00024757160231130313,
      "loss": 4.819,
      "mean_token_accuracy": 0.2202747568488121,
      "num_tokens": 13236342.0,
      "step": 6390
    },
    {
      "entropy": 5.318978643417358,
      "epoch": 5.80872330758746,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00024722084303163607,
      "loss": 4.711,
      "mean_token_accuracy": 0.22624929696321489,
      "num_tokens": 13244989.0,
      "step": 6395
    },
    {
      "entropy": 5.3740006446838375,
      "epoch": 5.813266696955929,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002468701522942774,
      "loss": 4.7403,
      "mean_token_accuracy": 0.22168441563844682,
      "num_tokens": 13255495.0,
      "step": 6400
    },
    {
      "entropy": 5.428449869155884,
      "epoch": 5.817810086324398,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00024651953096452165,
      "loss": 4.7978,
      "mean_token_accuracy": 0.2143199175596237,
      "num_tokens": 13266189.0,
      "step": 6405
    },
    {
      "entropy": 5.352286338806152,
      "epoch": 5.822353475692867,
      "grad_norm": 1.265625,
      "learning_rate": 0.00024616897990749196,
      "loss": 4.7627,
      "mean_token_accuracy": 0.21914055794477463,
      "num_tokens": 13276194.0,
      "step": 6410
    },
    {
      "entropy": 5.347740554809571,
      "epoch": 5.826896865061336,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002458184999881385,
      "loss": 4.7804,
      "mean_token_accuracy": 0.21630769520998,
      "num_tokens": 13286188.0,
      "step": 6415
    },
    {
      "entropy": 5.393495559692383,
      "epoch": 5.831440254429805,
      "grad_norm": 1.296875,
      "learning_rate": 0.00024546809207123534,
      "loss": 4.7293,
      "mean_token_accuracy": 0.2255068376660347,
      "num_tokens": 13295926.0,
      "step": 6420
    },
    {
      "entropy": 5.4190332889556885,
      "epoch": 5.835983643798274,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002451177570213794,
      "loss": 4.7994,
      "mean_token_accuracy": 0.21933480948209763,
      "num_tokens": 13307373.0,
      "step": 6425
    },
    {
      "entropy": 5.357099914550782,
      "epoch": 5.840527033166742,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00024476749570298746,
      "loss": 4.7524,
      "mean_token_accuracy": 0.2192711591720581,
      "num_tokens": 13316984.0,
      "step": 6430
    },
    {
      "entropy": 5.313825559616089,
      "epoch": 5.845070422535211,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00024441730898029456,
      "loss": 4.7196,
      "mean_token_accuracy": 0.2256878435611725,
      "num_tokens": 13327188.0,
      "step": 6435
    },
    {
      "entropy": 5.474205493927002,
      "epoch": 5.84961381190368,
      "grad_norm": 1.140625,
      "learning_rate": 0.00024406719771735165,
      "loss": 4.8161,
      "mean_token_accuracy": 0.2214495062828064,
      "num_tokens": 13338108.0,
      "step": 6440
    },
    {
      "entropy": 5.356754541397095,
      "epoch": 5.854157201272149,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002437171627780234,
      "loss": 4.7194,
      "mean_token_accuracy": 0.2229494959115982,
      "num_tokens": 13347898.0,
      "step": 6445
    },
    {
      "entropy": 5.377622461318969,
      "epoch": 5.858700590640618,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00024336720502598624,
      "loss": 4.7873,
      "mean_token_accuracy": 0.22328079491853714,
      "num_tokens": 13358649.0,
      "step": 6450
    },
    {
      "entropy": 5.318341398239136,
      "epoch": 5.863243980009087,
      "grad_norm": 1.25,
      "learning_rate": 0.0002430173253247262,
      "loss": 4.7781,
      "mean_token_accuracy": 0.21938917636871338,
      "num_tokens": 13368447.0,
      "step": 6455
    },
    {
      "entropy": 5.323021316528321,
      "epoch": 5.867787369377556,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00024266752453753655,
      "loss": 4.7419,
      "mean_token_accuracy": 0.21933239251375197,
      "num_tokens": 13377809.0,
      "step": 6460
    },
    {
      "entropy": 5.407270812988282,
      "epoch": 5.872330758746024,
      "grad_norm": 1.109375,
      "learning_rate": 0.00024231780352751615,
      "loss": 4.8528,
      "mean_token_accuracy": 0.21235859394073486,
      "num_tokens": 13389352.0,
      "step": 6465
    },
    {
      "entropy": 5.4242559432983395,
      "epoch": 5.876874148114493,
      "grad_norm": 1.140625,
      "learning_rate": 0.00024196816315756665,
      "loss": 4.7562,
      "mean_token_accuracy": 0.22412494271993638,
      "num_tokens": 13399929.0,
      "step": 6470
    },
    {
      "entropy": 5.37782735824585,
      "epoch": 5.881417537482962,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002416186042903909,
      "loss": 4.6612,
      "mean_token_accuracy": 0.22580741792917253,
      "num_tokens": 13410096.0,
      "step": 6475
    },
    {
      "entropy": 5.3449465274810795,
      "epoch": 5.885960926851431,
      "grad_norm": 1.25,
      "learning_rate": 0.00024126912778849074,
      "loss": 4.8527,
      "mean_token_accuracy": 0.21656646579504013,
      "num_tokens": 13420384.0,
      "step": 6480
    },
    {
      "entropy": 5.402365732192993,
      "epoch": 5.8905043162199,
      "grad_norm": 1.203125,
      "learning_rate": 0.00024091973451416472,
      "loss": 4.7393,
      "mean_token_accuracy": 0.22209289222955703,
      "num_tokens": 13430887.0,
      "step": 6485
    },
    {
      "entropy": 5.379916334152222,
      "epoch": 5.895047705588369,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00024057042532950597,
      "loss": 4.7364,
      "mean_token_accuracy": 0.2256981000304222,
      "num_tokens": 13440914.0,
      "step": 6490
    },
    {
      "entropy": 5.354023504257202,
      "epoch": 5.899591094956838,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002402212010964001,
      "loss": 4.8085,
      "mean_token_accuracy": 0.21201860159635544,
      "num_tokens": 13452476.0,
      "step": 6495
    },
    {
      "entropy": 5.360160684585571,
      "epoch": 5.904134484325307,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00023987206267652334,
      "loss": 4.7272,
      "mean_token_accuracy": 0.22050591558218002,
      "num_tokens": 13463779.0,
      "step": 6500
    },
    {
      "epoch": 5.904134484325307,
      "eval_entropy": 5.244270818178044,
      "eval_loss": 5.649238109588623,
      "eval_mean_token_accuracy": 0.17540705186683078,
      "eval_num_tokens": 13463779.0,
      "eval_runtime": 2.2039,
      "eval_samples_per_second": 1558.586,
      "eval_steps_per_second": 195.107,
      "step": 6500
    },
    {
      "entropy": 5.388013744354248,
      "epoch": 5.908677873693776,
      "grad_norm": 1.09375,
      "learning_rate": 0.00023952301093133993,
      "loss": 4.8478,
      "mean_token_accuracy": 0.21418342143297195,
      "num_tokens": 13475467.0,
      "step": 6505
    },
    {
      "entropy": 5.353434610366821,
      "epoch": 5.913221263062244,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00023917404672210046,
      "loss": 4.7051,
      "mean_token_accuracy": 0.22004712522029876,
      "num_tokens": 13485484.0,
      "step": 6510
    },
    {
      "entropy": 5.3657999515533445,
      "epoch": 5.917764652430713,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002388251709098392,
      "loss": 4.7196,
      "mean_token_accuracy": 0.22641577571630478,
      "num_tokens": 13494941.0,
      "step": 6515
    },
    {
      "entropy": 5.428310203552246,
      "epoch": 5.922308041799182,
      "grad_norm": 1.15625,
      "learning_rate": 0.00023847638435537283,
      "loss": 4.7888,
      "mean_token_accuracy": 0.22599727511405945,
      "num_tokens": 13504762.0,
      "step": 6520
    },
    {
      "entropy": 5.383725976943969,
      "epoch": 5.926851431167651,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00023812768791929724,
      "loss": 4.7457,
      "mean_token_accuracy": 0.21856015920639038,
      "num_tokens": 13515436.0,
      "step": 6525
    },
    {
      "entropy": 5.356895971298218,
      "epoch": 5.93139482053612,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00023777908246198643,
      "loss": 4.7606,
      "mean_token_accuracy": 0.2206914559006691,
      "num_tokens": 13525211.0,
      "step": 6530
    },
    {
      "entropy": 5.371611833572388,
      "epoch": 5.935938209904589,
      "grad_norm": 1.140625,
      "learning_rate": 0.00023743056884358954,
      "loss": 4.7355,
      "mean_token_accuracy": 0.21972739845514297,
      "num_tokens": 13534954.0,
      "step": 6535
    },
    {
      "entropy": 5.37831711769104,
      "epoch": 5.940481599273058,
      "grad_norm": 1.234375,
      "learning_rate": 0.00023708214792402942,
      "loss": 4.7672,
      "mean_token_accuracy": 0.23271822631359101,
      "num_tokens": 13545459.0,
      "step": 6540
    },
    {
      "entropy": 5.341162919998169,
      "epoch": 5.945024988641527,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00023673382056299997,
      "loss": 4.7435,
      "mean_token_accuracy": 0.22186996638774872,
      "num_tokens": 13554541.0,
      "step": 6545
    },
    {
      "entropy": 5.395187473297119,
      "epoch": 5.949568378009996,
      "grad_norm": 1.1875,
      "learning_rate": 0.00023638558761996436,
      "loss": 4.7575,
      "mean_token_accuracy": 0.21707545667886735,
      "num_tokens": 13564810.0,
      "step": 6550
    },
    {
      "entropy": 5.367066335678101,
      "epoch": 5.954111767378464,
      "grad_norm": 1.171875,
      "learning_rate": 0.00023603744995415277,
      "loss": 4.7287,
      "mean_token_accuracy": 0.2215479016304016,
      "num_tokens": 13574756.0,
      "step": 6555
    },
    {
      "entropy": 5.416759777069092,
      "epoch": 5.958655156746933,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00023568940842456026,
      "loss": 4.7942,
      "mean_token_accuracy": 0.218600732088089,
      "num_tokens": 13584346.0,
      "step": 6560
    },
    {
      "entropy": 5.419164323806763,
      "epoch": 5.963198546115402,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002353414638899447,
      "loss": 4.7844,
      "mean_token_accuracy": 0.2216697856783867,
      "num_tokens": 13595713.0,
      "step": 6565
    },
    {
      "entropy": 5.261274719238282,
      "epoch": 5.967741935483871,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00023499361720882468,
      "loss": 4.6525,
      "mean_token_accuracy": 0.23123864978551864,
      "num_tokens": 13605065.0,
      "step": 6570
    },
    {
      "entropy": 5.334798765182495,
      "epoch": 5.97228532485234,
      "grad_norm": 1.21875,
      "learning_rate": 0.00023464586923947718,
      "loss": 4.8051,
      "mean_token_accuracy": 0.2213533639907837,
      "num_tokens": 13615028.0,
      "step": 6575
    },
    {
      "entropy": 5.344036054611206,
      "epoch": 5.976828714220809,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00023429822083993596,
      "loss": 4.7227,
      "mean_token_accuracy": 0.22255554050207138,
      "num_tokens": 13625752.0,
      "step": 6580
    },
    {
      "entropy": 5.417845153808594,
      "epoch": 5.981372103589278,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00023395067286798872,
      "loss": 4.7017,
      "mean_token_accuracy": 0.22743663042783738,
      "num_tokens": 13635094.0,
      "step": 6585
    },
    {
      "entropy": 5.347240877151489,
      "epoch": 5.985915492957746,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00023360322618117563,
      "loss": 4.7546,
      "mean_token_accuracy": 0.2272810846567154,
      "num_tokens": 13645007.0,
      "step": 6590
    },
    {
      "entropy": 5.289091873168945,
      "epoch": 5.990458882326215,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00023325588163678674,
      "loss": 4.6514,
      "mean_token_accuracy": 0.22483912706375123,
      "num_tokens": 13655101.0,
      "step": 6595
    },
    {
      "entropy": 5.344776630401611,
      "epoch": 5.995002271694684,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00023290864009186036,
      "loss": 4.7146,
      "mean_token_accuracy": 0.22480927109718324,
      "num_tokens": 13665289.0,
      "step": 6600
    },
    {
      "entropy": 5.404370880126953,
      "epoch": 5.999545661063153,
      "grad_norm": 1.296875,
      "learning_rate": 0.00023256150240318025,
      "loss": 4.8198,
      "mean_token_accuracy": 0.21757276356220245,
      "num_tokens": 13675384.0,
      "step": 6605
    },
    {
      "entropy": 5.439683437347412,
      "epoch": 6.003634711494775,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00023221446942727438,
      "loss": 4.7206,
      "mean_token_accuracy": 0.23997605674796635,
      "num_tokens": 13683918.0,
      "step": 6610
    },
    {
      "entropy": 5.393720436096191,
      "epoch": 6.008178100863244,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002318675420204119,
      "loss": 4.6408,
      "mean_token_accuracy": 0.22987898141145707,
      "num_tokens": 13693999.0,
      "step": 6615
    },
    {
      "entropy": 5.4327880382537845,
      "epoch": 6.012721490231713,
      "grad_norm": 1.046875,
      "learning_rate": 0.00023152072103860188,
      "loss": 4.703,
      "mean_token_accuracy": 0.2277260974049568,
      "num_tokens": 13705465.0,
      "step": 6620
    },
    {
      "entropy": 5.339075422286987,
      "epoch": 6.017264879600182,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00023117400733759053,
      "loss": 4.5784,
      "mean_token_accuracy": 0.2356556236743927,
      "num_tokens": 13715858.0,
      "step": 6625
    },
    {
      "entropy": 5.374523448944092,
      "epoch": 6.021808268968651,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002308274017728595,
      "loss": 4.5997,
      "mean_token_accuracy": 0.23941086381673812,
      "num_tokens": 13725639.0,
      "step": 6630
    },
    {
      "entropy": 5.27228798866272,
      "epoch": 6.026351658337119,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002304809051996235,
      "loss": 4.5109,
      "mean_token_accuracy": 0.24581624269485475,
      "num_tokens": 13736015.0,
      "step": 6635
    },
    {
      "entropy": 5.192969036102295,
      "epoch": 6.030895047705588,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002301345184728284,
      "loss": 4.488,
      "mean_token_accuracy": 0.2312484011054039,
      "num_tokens": 13745433.0,
      "step": 6640
    },
    {
      "entropy": 5.367278671264648,
      "epoch": 6.035438437074057,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00022978824244714903,
      "loss": 4.5914,
      "mean_token_accuracy": 0.2444334879517555,
      "num_tokens": 13754970.0,
      "step": 6645
    },
    {
      "entropy": 5.366037273406983,
      "epoch": 6.039981826442526,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00022944207797698707,
      "loss": 4.681,
      "mean_token_accuracy": 0.22910215705633163,
      "num_tokens": 13766356.0,
      "step": 6650
    },
    {
      "entropy": 5.344854927062988,
      "epoch": 6.044525215810995,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002290960259164689,
      "loss": 4.6468,
      "mean_token_accuracy": 0.2350903794169426,
      "num_tokens": 13776503.0,
      "step": 6655
    },
    {
      "entropy": 5.397882652282715,
      "epoch": 6.049068605179464,
      "grad_norm": 1.078125,
      "learning_rate": 0.00022875008711944355,
      "loss": 4.6488,
      "mean_token_accuracy": 0.2274888575077057,
      "num_tokens": 13787918.0,
      "step": 6660
    },
    {
      "entropy": 5.270745134353637,
      "epoch": 6.053611994547933,
      "grad_norm": 1.1875,
      "learning_rate": 0.00022840426243948077,
      "loss": 4.5971,
      "mean_token_accuracy": 0.23010750561952592,
      "num_tokens": 13798305.0,
      "step": 6665
    },
    {
      "entropy": 5.336856555938721,
      "epoch": 6.058155383916402,
      "grad_norm": 1.1875,
      "learning_rate": 0.00022805855272986842,
      "loss": 4.6181,
      "mean_token_accuracy": 0.22882436066865922,
      "num_tokens": 13808867.0,
      "step": 6670
    },
    {
      "entropy": 5.297554397583008,
      "epoch": 6.062698773284871,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00022771295884361088,
      "loss": 4.4891,
      "mean_token_accuracy": 0.24593140184879303,
      "num_tokens": 13820073.0,
      "step": 6675
    },
    {
      "entropy": 5.372827625274658,
      "epoch": 6.067242162653339,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002273674816334268,
      "loss": 4.6509,
      "mean_token_accuracy": 0.23335450738668442,
      "num_tokens": 13830318.0,
      "step": 6680
    },
    {
      "entropy": 5.253058338165284,
      "epoch": 6.071785552021808,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00022702212195174664,
      "loss": 4.5856,
      "mean_token_accuracy": 0.23185911476612092,
      "num_tokens": 13840847.0,
      "step": 6685
    },
    {
      "entropy": 5.357430362701416,
      "epoch": 6.076328941390277,
      "grad_norm": 1.296875,
      "learning_rate": 0.00022667688065071146,
      "loss": 4.5857,
      "mean_token_accuracy": 0.23659006059169768,
      "num_tokens": 13850478.0,
      "step": 6690
    },
    {
      "entropy": 5.377923107147216,
      "epoch": 6.080872330758746,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00022633175858216936,
      "loss": 4.6557,
      "mean_token_accuracy": 0.2307317554950714,
      "num_tokens": 13860749.0,
      "step": 6695
    },
    {
      "entropy": 5.365042734146118,
      "epoch": 6.085415720127215,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00022598675659767515,
      "loss": 4.6498,
      "mean_token_accuracy": 0.23305918276309967,
      "num_tokens": 13869909.0,
      "step": 6700
    },
    {
      "entropy": 5.2986362934112545,
      "epoch": 6.089959109495684,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00022564187554848663,
      "loss": 4.5941,
      "mean_token_accuracy": 0.23669322431087494,
      "num_tokens": 13880235.0,
      "step": 6705
    },
    {
      "entropy": 5.229275703430176,
      "epoch": 6.094502498864153,
      "grad_norm": 1.25,
      "learning_rate": 0.00022529711628556387,
      "loss": 4.5,
      "mean_token_accuracy": 0.24297849386930465,
      "num_tokens": 13890324.0,
      "step": 6710
    },
    {
      "entropy": 5.3842566967010494,
      "epoch": 6.099045888232622,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00022495247965956567,
      "loss": 4.6381,
      "mean_token_accuracy": 0.2314253717660904,
      "num_tokens": 13900250.0,
      "step": 6715
    },
    {
      "entropy": 5.361855316162109,
      "epoch": 6.10358927760109,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002246079665208489,
      "loss": 4.5968,
      "mean_token_accuracy": 0.226557394862175,
      "num_tokens": 13909736.0,
      "step": 6720
    },
    {
      "entropy": 5.3018059730529785,
      "epoch": 6.108132666969559,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00022426357771946544,
      "loss": 4.6194,
      "mean_token_accuracy": 0.2350020706653595,
      "num_tokens": 13920187.0,
      "step": 6725
    },
    {
      "entropy": 5.3208760738372805,
      "epoch": 6.112676056338028,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00022391931410516036,
      "loss": 4.5723,
      "mean_token_accuracy": 0.22392480224370956,
      "num_tokens": 13930858.0,
      "step": 6730
    },
    {
      "entropy": 5.27270097732544,
      "epoch": 6.117219445706497,
      "grad_norm": 1.296875,
      "learning_rate": 0.00022357517652736992,
      "loss": 4.5282,
      "mean_token_accuracy": 0.2422218218445778,
      "num_tokens": 13940771.0,
      "step": 6735
    },
    {
      "entropy": 5.310018587112427,
      "epoch": 6.121762835074966,
      "grad_norm": 1.34375,
      "learning_rate": 0.00022323116583521948,
      "loss": 4.6729,
      "mean_token_accuracy": 0.22430850565433502,
      "num_tokens": 13951312.0,
      "step": 6740
    },
    {
      "entropy": 5.32427077293396,
      "epoch": 6.126306224443435,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00022288728287752109,
      "loss": 4.6403,
      "mean_token_accuracy": 0.229435558617115,
      "num_tokens": 13960862.0,
      "step": 6745
    },
    {
      "entropy": 5.343790578842163,
      "epoch": 6.130849613811904,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002225435285027719,
      "loss": 4.6079,
      "mean_token_accuracy": 0.23076928853988649,
      "num_tokens": 13972228.0,
      "step": 6750
    },
    {
      "entropy": 5.299166107177735,
      "epoch": 6.135393003180373,
      "grad_norm": 1.21875,
      "learning_rate": 0.00022219990355915148,
      "loss": 4.5761,
      "mean_token_accuracy": 0.23780259490013123,
      "num_tokens": 13982977.0,
      "step": 6755
    },
    {
      "entropy": 5.395422410964966,
      "epoch": 6.139936392548841,
      "grad_norm": 1.328125,
      "learning_rate": 0.00022185640889452036,
      "loss": 4.55,
      "mean_token_accuracy": 0.24228204637765885,
      "num_tokens": 13991645.0,
      "step": 6760
    },
    {
      "entropy": 5.404045677185058,
      "epoch": 6.14447978191731,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00022151304535641747,
      "loss": 4.6132,
      "mean_token_accuracy": 0.22810261249542235,
      "num_tokens": 14001313.0,
      "step": 6765
    },
    {
      "entropy": 5.3323915004730225,
      "epoch": 6.149023171285779,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002211698137920582,
      "loss": 4.6318,
      "mean_token_accuracy": 0.22985336780548096,
      "num_tokens": 14011248.0,
      "step": 6770
    },
    {
      "entropy": 5.341980266571045,
      "epoch": 6.153566560654248,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00022082671504833223,
      "loss": 4.5995,
      "mean_token_accuracy": 0.23081060498952866,
      "num_tokens": 14022290.0,
      "step": 6775
    },
    {
      "entropy": 5.3025165557861325,
      "epoch": 6.158109950022717,
      "grad_norm": 1.171875,
      "learning_rate": 0.00022048374997180175,
      "loss": 4.5845,
      "mean_token_accuracy": 0.23648117035627364,
      "num_tokens": 14032899.0,
      "step": 6780
    },
    {
      "entropy": 5.382287979125977,
      "epoch": 6.162653339391186,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00022014091940869885,
      "loss": 4.6075,
      "mean_token_accuracy": 0.2361038625240326,
      "num_tokens": 14043065.0,
      "step": 6785
    },
    {
      "entropy": 5.29147162437439,
      "epoch": 6.167196728759655,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00021979822420492402,
      "loss": 4.6401,
      "mean_token_accuracy": 0.2312309294939041,
      "num_tokens": 14054594.0,
      "step": 6790
    },
    {
      "entropy": 5.440401363372803,
      "epoch": 6.171740118128124,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00021945566520604337,
      "loss": 4.7335,
      "mean_token_accuracy": 0.22482556700706482,
      "num_tokens": 14064211.0,
      "step": 6795
    },
    {
      "entropy": 5.31172513961792,
      "epoch": 6.176283507496592,
      "grad_norm": 1.25,
      "learning_rate": 0.0002191132432572875,
      "loss": 4.6176,
      "mean_token_accuracy": 0.23840446323156356,
      "num_tokens": 14073946.0,
      "step": 6800
    },
    {
      "entropy": 5.332688856124878,
      "epoch": 6.180826896865061,
      "grad_norm": 1.265625,
      "learning_rate": 0.00021877095920354818,
      "loss": 4.538,
      "mean_token_accuracy": 0.2388283669948578,
      "num_tokens": 14083303.0,
      "step": 6805
    },
    {
      "entropy": 5.3314307689666744,
      "epoch": 6.18537028623353,
      "grad_norm": 1.28125,
      "learning_rate": 0.00021842881388937764,
      "loss": 4.5422,
      "mean_token_accuracy": 0.240011565387249,
      "num_tokens": 14093475.0,
      "step": 6810
    },
    {
      "entropy": 5.383239316940307,
      "epoch": 6.189913675601999,
      "grad_norm": 1.1875,
      "learning_rate": 0.00021808680815898508,
      "loss": 4.6738,
      "mean_token_accuracy": 0.23169527053833008,
      "num_tokens": 14104117.0,
      "step": 6815
    },
    {
      "entropy": 5.26419997215271,
      "epoch": 6.194457064970468,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00021774494285623598,
      "loss": 4.5615,
      "mean_token_accuracy": 0.23605556637048722,
      "num_tokens": 14115480.0,
      "step": 6820
    },
    {
      "entropy": 5.265815544128418,
      "epoch": 6.199000454338937,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002174032188246487,
      "loss": 4.5539,
      "mean_token_accuracy": 0.23568375110626222,
      "num_tokens": 14125827.0,
      "step": 6825
    },
    {
      "entropy": 5.318867969512939,
      "epoch": 6.203543843707406,
      "grad_norm": 1.203125,
      "learning_rate": 0.00021706163690739344,
      "loss": 4.6135,
      "mean_token_accuracy": 0.23609348088502885,
      "num_tokens": 14136424.0,
      "step": 6830
    },
    {
      "entropy": 5.3959362506866455,
      "epoch": 6.208087233075875,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00021672019794728963,
      "loss": 4.6957,
      "mean_token_accuracy": 0.22257688790559768,
      "num_tokens": 14147478.0,
      "step": 6835
    },
    {
      "entropy": 5.416971349716187,
      "epoch": 6.212630622444343,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00021637890278680406,
      "loss": 4.6849,
      "mean_token_accuracy": 0.23246722370386125,
      "num_tokens": 14159044.0,
      "step": 6840
    },
    {
      "entropy": 5.327849054336548,
      "epoch": 6.217174011812812,
      "grad_norm": 1.234375,
      "learning_rate": 0.00021603775226804845,
      "loss": 4.6241,
      "mean_token_accuracy": 0.23269544392824174,
      "num_tokens": 14169370.0,
      "step": 6845
    },
    {
      "entropy": 5.3461650848388675,
      "epoch": 6.221717401181281,
      "grad_norm": 1.34375,
      "learning_rate": 0.00021569674723277783,
      "loss": 4.6469,
      "mean_token_accuracy": 0.23357320874929427,
      "num_tokens": 14179774.0,
      "step": 6850
    },
    {
      "entropy": 5.374762344360351,
      "epoch": 6.22626079054975,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00021535588852238823,
      "loss": 4.5978,
      "mean_token_accuracy": 0.2405007392168045,
      "num_tokens": 14190182.0,
      "step": 6855
    },
    {
      "entropy": 5.308204221725464,
      "epoch": 6.230804179918219,
      "grad_norm": 1.3125,
      "learning_rate": 0.00021501517697791479,
      "loss": 4.5263,
      "mean_token_accuracy": 0.24233802258968354,
      "num_tokens": 14199505.0,
      "step": 6860
    },
    {
      "entropy": 5.338066625595093,
      "epoch": 6.235347569286688,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002146746134400291,
      "loss": 4.6407,
      "mean_token_accuracy": 0.22296755611896515,
      "num_tokens": 14209677.0,
      "step": 6865
    },
    {
      "entropy": 5.304502534866333,
      "epoch": 6.239890958655157,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00021433419874903803,
      "loss": 4.5817,
      "mean_token_accuracy": 0.2319676846265793,
      "num_tokens": 14219356.0,
      "step": 6870
    },
    {
      "entropy": 5.276384401321411,
      "epoch": 6.244434348023626,
      "grad_norm": 1.40625,
      "learning_rate": 0.00021399393374488097,
      "loss": 4.5325,
      "mean_token_accuracy": 0.247141295671463,
      "num_tokens": 14227992.0,
      "step": 6875
    },
    {
      "entropy": 5.379275798797607,
      "epoch": 6.248977737392094,
      "grad_norm": 1.171875,
      "learning_rate": 0.00021365381926712797,
      "loss": 4.5927,
      "mean_token_accuracy": 0.23550449311733246,
      "num_tokens": 14238354.0,
      "step": 6880
    },
    {
      "entropy": 5.3425108909606935,
      "epoch": 6.253521126760563,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00021331385615497756,
      "loss": 4.6348,
      "mean_token_accuracy": 0.23240153342485428,
      "num_tokens": 14249560.0,
      "step": 6885
    },
    {
      "entropy": 5.3338686466217045,
      "epoch": 6.258064516129032,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002129740452472552,
      "loss": 4.6622,
      "mean_token_accuracy": 0.2253888338804245,
      "num_tokens": 14259372.0,
      "step": 6890
    },
    {
      "entropy": 5.3283140659332275,
      "epoch": 6.262607905497501,
      "grad_norm": 1.15625,
      "learning_rate": 0.00021263438738241026,
      "loss": 4.5755,
      "mean_token_accuracy": 0.23725361973047257,
      "num_tokens": 14270837.0,
      "step": 6895
    },
    {
      "entropy": 5.330993127822876,
      "epoch": 6.26715129486597,
      "grad_norm": 1.21875,
      "learning_rate": 0.00021229488339851498,
      "loss": 4.5839,
      "mean_token_accuracy": 0.24013753235340118,
      "num_tokens": 14280138.0,
      "step": 6900
    },
    {
      "entropy": 5.366893482208252,
      "epoch": 6.271694684234439,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00021195553413326146,
      "loss": 4.6947,
      "mean_token_accuracy": 0.22456787824630736,
      "num_tokens": 14290131.0,
      "step": 6905
    },
    {
      "entropy": 5.283655118942261,
      "epoch": 6.276238073602908,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002116163404239605,
      "loss": 4.6241,
      "mean_token_accuracy": 0.23082166761159897,
      "num_tokens": 14299961.0,
      "step": 6910
    },
    {
      "entropy": 5.304340171813965,
      "epoch": 6.280781462971377,
      "grad_norm": 1.265625,
      "learning_rate": 0.00021127730310753866,
      "loss": 4.5723,
      "mean_token_accuracy": 0.23955017775297166,
      "num_tokens": 14310529.0,
      "step": 6915
    },
    {
      "entropy": 5.289507722854614,
      "epoch": 6.285324852339846,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00021093842302053706,
      "loss": 4.5187,
      "mean_token_accuracy": 0.2466626822948456,
      "num_tokens": 14320220.0,
      "step": 6920
    },
    {
      "entropy": 5.42295093536377,
      "epoch": 6.289868241708314,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00021059970099910835,
      "loss": 4.6626,
      "mean_token_accuracy": 0.23541614711284636,
      "num_tokens": 14331700.0,
      "step": 6925
    },
    {
      "entropy": 5.330759716033936,
      "epoch": 6.294411631076783,
      "grad_norm": 1.25,
      "learning_rate": 0.00021026113787901568,
      "loss": 4.6009,
      "mean_token_accuracy": 0.2364975020289421,
      "num_tokens": 14341163.0,
      "step": 6930
    },
    {
      "entropy": 5.303437566757202,
      "epoch": 6.298955020445252,
      "grad_norm": 1.171875,
      "learning_rate": 0.00020992273449562966,
      "loss": 4.6093,
      "mean_token_accuracy": 0.2332180082798004,
      "num_tokens": 14351398.0,
      "step": 6935
    },
    {
      "entropy": 5.36496524810791,
      "epoch": 6.303498409813721,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00020958449168392723,
      "loss": 4.608,
      "mean_token_accuracy": 0.23479930758476258,
      "num_tokens": 14361920.0,
      "step": 6940
    },
    {
      "entropy": 5.2969990253448485,
      "epoch": 6.30804179918219,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002092464102784888,
      "loss": 4.604,
      "mean_token_accuracy": 0.23360318392515184,
      "num_tokens": 14371980.0,
      "step": 6945
    },
    {
      "entropy": 5.300631856918335,
      "epoch": 6.312585188550659,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00020890849111349663,
      "loss": 4.6283,
      "mean_token_accuracy": 0.22778421491384507,
      "num_tokens": 14381397.0,
      "step": 6950
    },
    {
      "entropy": 5.2807502269744875,
      "epoch": 6.317128577919128,
      "grad_norm": 1.21875,
      "learning_rate": 0.00020857073502273266,
      "loss": 4.6434,
      "mean_token_accuracy": 0.2269931674003601,
      "num_tokens": 14393476.0,
      "step": 6955
    },
    {
      "entropy": 5.343697929382325,
      "epoch": 6.321671967287597,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002082331428395765,
      "loss": 4.6321,
      "mean_token_accuracy": 0.2337341398000717,
      "num_tokens": 14404801.0,
      "step": 6960
    },
    {
      "entropy": 5.346200704574585,
      "epoch": 6.326215356656065,
      "grad_norm": 1.1875,
      "learning_rate": 0.00020789571539700326,
      "loss": 4.6404,
      "mean_token_accuracy": 0.23234931975603104,
      "num_tokens": 14415921.0,
      "step": 6965
    },
    {
      "entropy": 5.2825743675231935,
      "epoch": 6.330758746024534,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00020755845352758175,
      "loss": 4.6458,
      "mean_token_accuracy": 0.22893086820840836,
      "num_tokens": 14426455.0,
      "step": 6970
    },
    {
      "entropy": 5.3493248462677006,
      "epoch": 6.335302135393003,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002072213580634718,
      "loss": 4.6162,
      "mean_token_accuracy": 0.2365632951259613,
      "num_tokens": 14438720.0,
      "step": 6975
    },
    {
      "entropy": 5.325234651565552,
      "epoch": 6.339845524761472,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002068844298364233,
      "loss": 4.6495,
      "mean_token_accuracy": 0.23035131692886351,
      "num_tokens": 14449644.0,
      "step": 6980
    },
    {
      "entropy": 5.347111797332763,
      "epoch": 6.344388914129941,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00020654766967777288,
      "loss": 4.6817,
      "mean_token_accuracy": 0.2235606387257576,
      "num_tokens": 14459712.0,
      "step": 6985
    },
    {
      "entropy": 5.368547201156616,
      "epoch": 6.34893230349841,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002062110784184431,
      "loss": 4.6358,
      "mean_token_accuracy": 0.22421020418405532,
      "num_tokens": 14469440.0,
      "step": 6990
    },
    {
      "entropy": 5.331425523757934,
      "epoch": 6.353475692866879,
      "grad_norm": 1.1875,
      "learning_rate": 0.00020587465688893904,
      "loss": 4.671,
      "mean_token_accuracy": 0.22269476801156998,
      "num_tokens": 14480587.0,
      "step": 6995
    },
    {
      "entropy": 5.3906364917755125,
      "epoch": 6.358019082235348,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00020553840591934765,
      "loss": 4.6838,
      "mean_token_accuracy": 0.22819068282842636,
      "num_tokens": 14490360.0,
      "step": 7000
    },
    {
      "epoch": 6.358019082235348,
      "eval_entropy": 5.239281795191211,
      "eval_loss": 5.6642351150512695,
      "eval_mean_token_accuracy": 0.17438331284495287,
      "eval_num_tokens": 14490360.0,
      "eval_runtime": 2.0498,
      "eval_samples_per_second": 1675.745,
      "eval_steps_per_second": 209.773,
      "step": 7000
    },
    {
      "entropy": 5.333258104324341,
      "epoch": 6.362562471603816,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002052023263393346,
      "loss": 4.6029,
      "mean_token_accuracy": 0.2357860714197159,
      "num_tokens": 14501135.0,
      "step": 7005
    },
    {
      "entropy": 5.341643953323365,
      "epoch": 6.367105860972285,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00020486641897814295,
      "loss": 4.6256,
      "mean_token_accuracy": 0.2311222404241562,
      "num_tokens": 14511828.0,
      "step": 7010
    },
    {
      "entropy": 5.299260377883911,
      "epoch": 6.371649250340754,
      "grad_norm": 1.390625,
      "learning_rate": 0.00020453068466459067,
      "loss": 4.6073,
      "mean_token_accuracy": 0.235933555662632,
      "num_tokens": 14521973.0,
      "step": 7015
    },
    {
      "entropy": 5.378217363357544,
      "epoch": 6.376192639709223,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00020419512422706893,
      "loss": 4.6736,
      "mean_token_accuracy": 0.23339507281780242,
      "num_tokens": 14532515.0,
      "step": 7020
    },
    {
      "entropy": 5.354877662658692,
      "epoch": 6.380736029077692,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00020385973849353957,
      "loss": 4.6073,
      "mean_token_accuracy": 0.23221782147884368,
      "num_tokens": 14542577.0,
      "step": 7025
    },
    {
      "entropy": 5.307950115203857,
      "epoch": 6.385279418446161,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00020352452829153369,
      "loss": 4.5412,
      "mean_token_accuracy": 0.23309696912765504,
      "num_tokens": 14552472.0,
      "step": 7030
    },
    {
      "entropy": 5.291578054428101,
      "epoch": 6.38982280781463,
      "grad_norm": 1.125,
      "learning_rate": 0.000203189494448149,
      "loss": 4.6042,
      "mean_token_accuracy": 0.235651595890522,
      "num_tokens": 14562615.0,
      "step": 7035
    },
    {
      "entropy": 5.302771091461182,
      "epoch": 6.394366197183099,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00020285463779004837,
      "loss": 4.6,
      "mean_token_accuracy": 0.24091863185167312,
      "num_tokens": 14573200.0,
      "step": 7040
    },
    {
      "entropy": 5.381127977371216,
      "epoch": 6.398909586551568,
      "grad_norm": 1.28125,
      "learning_rate": 0.00020251995914345722,
      "loss": 4.6329,
      "mean_token_accuracy": 0.23823562562465667,
      "num_tokens": 14583042.0,
      "step": 7045
    },
    {
      "entropy": 5.4001991748809814,
      "epoch": 6.403452975920036,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00020218545933416181,
      "loss": 4.6494,
      "mean_token_accuracy": 0.23713090568780898,
      "num_tokens": 14594409.0,
      "step": 7050
    },
    {
      "entropy": 5.3736546516418455,
      "epoch": 6.407996365288505,
      "grad_norm": 1.125,
      "learning_rate": 0.00020185113918750724,
      "loss": 4.709,
      "mean_token_accuracy": 0.22900959104299545,
      "num_tokens": 14605370.0,
      "step": 7055
    },
    {
      "entropy": 5.353535127639771,
      "epoch": 6.412539754656974,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002015169995283952,
      "loss": 4.6412,
      "mean_token_accuracy": 0.23177941590547563,
      "num_tokens": 14615056.0,
      "step": 7060
    },
    {
      "entropy": 5.442641019821167,
      "epoch": 6.417083144025443,
      "grad_norm": 1.171875,
      "learning_rate": 0.00020118304118128211,
      "loss": 4.7026,
      "mean_token_accuracy": 0.22277091592550277,
      "num_tokens": 14624959.0,
      "step": 7065
    },
    {
      "entropy": 5.334510850906372,
      "epoch": 6.421626533393912,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00020084926497017703,
      "loss": 4.5682,
      "mean_token_accuracy": 0.24437476694583893,
      "num_tokens": 14635348.0,
      "step": 7070
    },
    {
      "entropy": 5.288582992553711,
      "epoch": 6.426169922762381,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00020051567171863944,
      "loss": 4.628,
      "mean_token_accuracy": 0.22681427001953125,
      "num_tokens": 14645232.0,
      "step": 7075
    },
    {
      "entropy": 5.355676460266113,
      "epoch": 6.43071331213085,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002001822622497777,
      "loss": 4.6053,
      "mean_token_accuracy": 0.23929642140865326,
      "num_tokens": 14656194.0,
      "step": 7080
    },
    {
      "entropy": 5.372528696060181,
      "epoch": 6.435256701499318,
      "grad_norm": 1.171875,
      "learning_rate": 0.00019984903738624632,
      "loss": 4.6235,
      "mean_token_accuracy": 0.2363289102911949,
      "num_tokens": 14667019.0,
      "step": 7085
    },
    {
      "entropy": 5.370216655731201,
      "epoch": 6.439800090867787,
      "grad_norm": 1.546875,
      "learning_rate": 0.00019951599795024478,
      "loss": 4.6208,
      "mean_token_accuracy": 0.23888255208730697,
      "num_tokens": 14676339.0,
      "step": 7090
    },
    {
      "entropy": 5.37955904006958,
      "epoch": 6.444343480236256,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00019918314476351462,
      "loss": 4.6375,
      "mean_token_accuracy": 0.23400551676750184,
      "num_tokens": 14688206.0,
      "step": 7095
    },
    {
      "entropy": 5.299837350845337,
      "epoch": 6.448886869604725,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00019885047864733803,
      "loss": 4.5949,
      "mean_token_accuracy": 0.2374642163515091,
      "num_tokens": 14698397.0,
      "step": 7100
    },
    {
      "entropy": 5.321491765975952,
      "epoch": 6.453430258973194,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00019851800042253553,
      "loss": 4.5682,
      "mean_token_accuracy": 0.24060127884149551,
      "num_tokens": 14708140.0,
      "step": 7105
    },
    {
      "entropy": 5.326788473129272,
      "epoch": 6.457973648341663,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001981857109094642,
      "loss": 4.6189,
      "mean_token_accuracy": 0.24014702886343003,
      "num_tokens": 14718710.0,
      "step": 7110
    },
    {
      "entropy": 5.348369932174682,
      "epoch": 6.462517037710132,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00019785361092801524,
      "loss": 4.6429,
      "mean_token_accuracy": 0.22898958176374434,
      "num_tokens": 14729143.0,
      "step": 7115
    },
    {
      "entropy": 5.344326686859131,
      "epoch": 6.467060427078601,
      "grad_norm": 1.234375,
      "learning_rate": 0.00019752170129761244,
      "loss": 4.6239,
      "mean_token_accuracy": 0.23978679925203322,
      "num_tokens": 14739386.0,
      "step": 7120
    },
    {
      "entropy": 5.3245179653167725,
      "epoch": 6.47160381644707,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001971899828372098,
      "loss": 4.6405,
      "mean_token_accuracy": 0.2365340396761894,
      "num_tokens": 14749209.0,
      "step": 7125
    },
    {
      "entropy": 5.3714910507202145,
      "epoch": 6.476147205815538,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00019685845636528972,
      "loss": 4.6373,
      "mean_token_accuracy": 0.23980695605278016,
      "num_tokens": 14759351.0,
      "step": 7130
    },
    {
      "entropy": 5.367058277130127,
      "epoch": 6.480690595184007,
      "grad_norm": 1.390625,
      "learning_rate": 0.00019652712269986074,
      "loss": 4.626,
      "mean_token_accuracy": 0.22997657507658004,
      "num_tokens": 14769069.0,
      "step": 7135
    },
    {
      "entropy": 5.377463626861572,
      "epoch": 6.485233984552476,
      "grad_norm": 1.359375,
      "learning_rate": 0.00019619598265845556,
      "loss": 4.6373,
      "mean_token_accuracy": 0.23722168654203415,
      "num_tokens": 14778563.0,
      "step": 7140
    },
    {
      "entropy": 5.2979395389556885,
      "epoch": 6.489777373920945,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00019586503705812969,
      "loss": 4.6135,
      "mean_token_accuracy": 0.23287311792373658,
      "num_tokens": 14788305.0,
      "step": 7145
    },
    {
      "entropy": 5.34933910369873,
      "epoch": 6.494320763289414,
      "grad_norm": 1.125,
      "learning_rate": 0.0001955342867154582,
      "loss": 4.678,
      "mean_token_accuracy": 0.22737507224082948,
      "num_tokens": 14799463.0,
      "step": 7150
    },
    {
      "entropy": 5.289685583114624,
      "epoch": 6.498864152657883,
      "grad_norm": 1.296875,
      "learning_rate": 0.00019520373244653484,
      "loss": 4.5751,
      "mean_token_accuracy": 0.24694059044122696,
      "num_tokens": 14809516.0,
      "step": 7155
    },
    {
      "entropy": 5.3292967796325685,
      "epoch": 6.503407542026352,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001948733750669693,
      "loss": 4.6868,
      "mean_token_accuracy": 0.22316508442163469,
      "num_tokens": 14819605.0,
      "step": 7160
    },
    {
      "entropy": 5.337313079833985,
      "epoch": 6.50795093139482,
      "grad_norm": 1.203125,
      "learning_rate": 0.00019454321539188568,
      "loss": 4.625,
      "mean_token_accuracy": 0.230569426715374,
      "num_tokens": 14830241.0,
      "step": 7165
    },
    {
      "entropy": 5.3634786128997805,
      "epoch": 6.512494320763289,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00019421325423592007,
      "loss": 4.6411,
      "mean_token_accuracy": 0.2359793961048126,
      "num_tokens": 14840294.0,
      "step": 7170
    },
    {
      "entropy": 5.381575202941894,
      "epoch": 6.517037710131758,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001938834924132189,
      "loss": 4.6705,
      "mean_token_accuracy": 0.23382510840892792,
      "num_tokens": 14850156.0,
      "step": 7175
    },
    {
      "entropy": 5.28847336769104,
      "epoch": 6.521581099500227,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0001935539307374366,
      "loss": 4.5744,
      "mean_token_accuracy": 0.24195554256439208,
      "num_tokens": 14859889.0,
      "step": 7180
    },
    {
      "entropy": 5.37569522857666,
      "epoch": 6.526124488868696,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00019322457002173395,
      "loss": 4.6437,
      "mean_token_accuracy": 0.23413251489400863,
      "num_tokens": 14871659.0,
      "step": 7185
    },
    {
      "entropy": 5.370726299285889,
      "epoch": 6.530667878237165,
      "grad_norm": 1.28125,
      "learning_rate": 0.00019289541107877562,
      "loss": 4.6208,
      "mean_token_accuracy": 0.23519658595323562,
      "num_tokens": 14882041.0,
      "step": 7190
    },
    {
      "entropy": 5.359445667266845,
      "epoch": 6.535211267605634,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00019256645472072876,
      "loss": 4.595,
      "mean_token_accuracy": 0.23978508710861207,
      "num_tokens": 14891709.0,
      "step": 7195
    },
    {
      "entropy": 5.332190942764282,
      "epoch": 6.539754656974103,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00019223770175926032,
      "loss": 4.6412,
      "mean_token_accuracy": 0.2357003942131996,
      "num_tokens": 14901898.0,
      "step": 7200
    },
    {
      "entropy": 5.290914678573609,
      "epoch": 6.544298046342572,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00019190915300553574,
      "loss": 4.5594,
      "mean_token_accuracy": 0.23793593943119049,
      "num_tokens": 14911910.0,
      "step": 7205
    },
    {
      "entropy": 5.330206251144409,
      "epoch": 6.54884143571104,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0001915808092702162,
      "loss": 4.6039,
      "mean_token_accuracy": 0.2280516430735588,
      "num_tokens": 14922055.0,
      "step": 7210
    },
    {
      "entropy": 5.294210052490234,
      "epoch": 6.553384825079509,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001912526713634574,
      "loss": 4.5856,
      "mean_token_accuracy": 0.23525553643703462,
      "num_tokens": 14931841.0,
      "step": 7215
    },
    {
      "entropy": 5.322956657409668,
      "epoch": 6.557928214447978,
      "grad_norm": 1.234375,
      "learning_rate": 0.0001909247400949069,
      "loss": 4.5982,
      "mean_token_accuracy": 0.2352904662489891,
      "num_tokens": 14942356.0,
      "step": 7220
    },
    {
      "entropy": 5.3171087265014645,
      "epoch": 6.562471603816447,
      "grad_norm": 1.34375,
      "learning_rate": 0.00019059701627370263,
      "loss": 4.5997,
      "mean_token_accuracy": 0.23545732200145722,
      "num_tokens": 14952552.0,
      "step": 7225
    },
    {
      "entropy": 5.2721601009368895,
      "epoch": 6.567014993184916,
      "grad_norm": 1.265625,
      "learning_rate": 0.00019026950070847025,
      "loss": 4.6051,
      "mean_token_accuracy": 0.23758504837751387,
      "num_tokens": 14962728.0,
      "step": 7230
    },
    {
      "entropy": 5.269742012023926,
      "epoch": 6.571558382553385,
      "grad_norm": 1.234375,
      "learning_rate": 0.00018994219420732228,
      "loss": 4.5834,
      "mean_token_accuracy": 0.24116588532924652,
      "num_tokens": 14974071.0,
      "step": 7235
    },
    {
      "entropy": 5.387656211853027,
      "epoch": 6.576101771921854,
      "grad_norm": 1.21875,
      "learning_rate": 0.00018961509757785462,
      "loss": 4.686,
      "mean_token_accuracy": 0.23256265372037888,
      "num_tokens": 14983567.0,
      "step": 7240
    },
    {
      "entropy": 5.27859845161438,
      "epoch": 6.580645161290323,
      "grad_norm": 1.40625,
      "learning_rate": 0.00018928821162714587,
      "loss": 4.6083,
      "mean_token_accuracy": 0.24086989611387252,
      "num_tokens": 14994172.0,
      "step": 7245
    },
    {
      "entropy": 5.33209867477417,
      "epoch": 6.585188550658792,
      "grad_norm": 1.171875,
      "learning_rate": 0.00018896153716175458,
      "loss": 4.5954,
      "mean_token_accuracy": 0.2391119807958603,
      "num_tokens": 15004094.0,
      "step": 7250
    },
    {
      "entropy": 5.313361263275146,
      "epoch": 6.58973194002726,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001886350749877175,
      "loss": 4.6037,
      "mean_token_accuracy": 0.23330151438713073,
      "num_tokens": 15013986.0,
      "step": 7255
    },
    {
      "entropy": 5.336466455459595,
      "epoch": 6.594275329395729,
      "grad_norm": 1.25,
      "learning_rate": 0.0001883088259105476,
      "loss": 4.6297,
      "mean_token_accuracy": 0.23669154793024064,
      "num_tokens": 15024764.0,
      "step": 7260
    },
    {
      "entropy": 5.3500075340271,
      "epoch": 6.598818718764198,
      "grad_norm": 1.21875,
      "learning_rate": 0.00018798279073523204,
      "loss": 4.6388,
      "mean_token_accuracy": 0.22523172646760942,
      "num_tokens": 15034971.0,
      "step": 7265
    },
    {
      "entropy": 5.341850233078003,
      "epoch": 6.603362108132667,
      "grad_norm": 1.296875,
      "learning_rate": 0.00018765697026623017,
      "loss": 4.7098,
      "mean_token_accuracy": 0.22534977197647094,
      "num_tokens": 15045750.0,
      "step": 7270
    },
    {
      "entropy": 5.381553316116333,
      "epoch": 6.607905497501136,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0001873313653074718,
      "loss": 4.7481,
      "mean_token_accuracy": 0.21793261170387268,
      "num_tokens": 15056840.0,
      "step": 7275
    },
    {
      "entropy": 5.287941789627075,
      "epoch": 6.612448886869605,
      "grad_norm": 1.390625,
      "learning_rate": 0.00018700597666235452,
      "loss": 4.5853,
      "mean_token_accuracy": 0.23458266407251357,
      "num_tokens": 15067451.0,
      "step": 7280
    },
    {
      "entropy": 5.372293949127197,
      "epoch": 6.616992276238074,
      "grad_norm": 1.25,
      "learning_rate": 0.00018668080513374268,
      "loss": 4.6406,
      "mean_token_accuracy": 0.2354448229074478,
      "num_tokens": 15077780.0,
      "step": 7285
    },
    {
      "entropy": 5.3711278438568115,
      "epoch": 6.621535665606542,
      "grad_norm": 1.21875,
      "learning_rate": 0.00018635585152396457,
      "loss": 4.7122,
      "mean_token_accuracy": 0.22054645419120789,
      "num_tokens": 15089100.0,
      "step": 7290
    },
    {
      "entropy": 5.319313621520996,
      "epoch": 6.626079054975011,
      "grad_norm": 1.203125,
      "learning_rate": 0.000186031116634811,
      "loss": 4.6217,
      "mean_token_accuracy": 0.23007019609212875,
      "num_tokens": 15099420.0,
      "step": 7295
    },
    {
      "entropy": 5.379638147354126,
      "epoch": 6.63062244434348,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00018570660126753291,
      "loss": 4.6492,
      "mean_token_accuracy": 0.22793243825435638,
      "num_tokens": 15109884.0,
      "step": 7300
    },
    {
      "entropy": 5.318063735961914,
      "epoch": 6.635165833711949,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001853823062228399,
      "loss": 4.5686,
      "mean_token_accuracy": 0.24668001383543015,
      "num_tokens": 15119761.0,
      "step": 7305
    },
    {
      "entropy": 5.267759227752686,
      "epoch": 6.639709223080418,
      "grad_norm": 1.359375,
      "learning_rate": 0.00018505823230089758,
      "loss": 4.5557,
      "mean_token_accuracy": 0.24956131726503372,
      "num_tokens": 15129159.0,
      "step": 7310
    },
    {
      "entropy": 5.333029460906983,
      "epoch": 6.644252612448887,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00018473438030132633,
      "loss": 4.5734,
      "mean_token_accuracy": 0.23725423216819763,
      "num_tokens": 15138855.0,
      "step": 7315
    },
    {
      "entropy": 5.330069732666016,
      "epoch": 6.648796001817356,
      "grad_norm": 1.1875,
      "learning_rate": 0.0001844107510231985,
      "loss": 4.6051,
      "mean_token_accuracy": 0.23148168474435807,
      "num_tokens": 15149957.0,
      "step": 7320
    },
    {
      "entropy": 5.365602111816406,
      "epoch": 6.653339391185825,
      "grad_norm": 1.40625,
      "learning_rate": 0.00018408734526503744,
      "loss": 4.6297,
      "mean_token_accuracy": 0.2324664831161499,
      "num_tokens": 15159110.0,
      "step": 7325
    },
    {
      "entropy": 5.368468618392944,
      "epoch": 6.657882780554294,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00018376416382481442,
      "loss": 4.6921,
      "mean_token_accuracy": 0.22820616215467454,
      "num_tokens": 15169444.0,
      "step": 7330
    },
    {
      "entropy": 5.473860263824463,
      "epoch": 6.662426169922762,
      "grad_norm": 1.265625,
      "learning_rate": 0.00018344120749994787,
      "loss": 4.814,
      "mean_token_accuracy": 0.21581710278987884,
      "num_tokens": 15180246.0,
      "step": 7335
    },
    {
      "entropy": 5.2863749980926515,
      "epoch": 6.666969559291231,
      "grad_norm": 1.21875,
      "learning_rate": 0.00018311847708730005,
      "loss": 4.5992,
      "mean_token_accuracy": 0.2326633781194687,
      "num_tokens": 15191578.0,
      "step": 7340
    },
    {
      "entropy": 5.305369710922241,
      "epoch": 6.6715129486597,
      "grad_norm": 1.1875,
      "learning_rate": 0.0001827959733831765,
      "loss": 4.6374,
      "mean_token_accuracy": 0.22414492070674896,
      "num_tokens": 15202380.0,
      "step": 7345
    },
    {
      "entropy": 5.284613990783692,
      "epoch": 6.676056338028169,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00018247369718332274,
      "loss": 4.6075,
      "mean_token_accuracy": 0.23796446621418,
      "num_tokens": 15214073.0,
      "step": 7350
    },
    {
      "entropy": 5.311483907699585,
      "epoch": 6.680599727396638,
      "grad_norm": 1.28125,
      "learning_rate": 0.00018215164928292354,
      "loss": 4.5634,
      "mean_token_accuracy": 0.24120257943868637,
      "num_tokens": 15223655.0,
      "step": 7355
    },
    {
      "entropy": 5.395785665512085,
      "epoch": 6.685143116765107,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00018182983047659984,
      "loss": 4.7332,
      "mean_token_accuracy": 0.2241150051355362,
      "num_tokens": 15235567.0,
      "step": 7360
    },
    {
      "entropy": 5.307887411117553,
      "epoch": 6.689686506133576,
      "grad_norm": 1.234375,
      "learning_rate": 0.00018150824155840778,
      "loss": 4.6238,
      "mean_token_accuracy": 0.23455462604761124,
      "num_tokens": 15246349.0,
      "step": 7365
    },
    {
      "entropy": 5.381241178512573,
      "epoch": 6.694229895502044,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0001811868833218358,
      "loss": 4.6812,
      "mean_token_accuracy": 0.23463217169046402,
      "num_tokens": 15256443.0,
      "step": 7370
    },
    {
      "entropy": 5.326403045654297,
      "epoch": 6.698773284870513,
      "grad_norm": 1.1875,
      "learning_rate": 0.00018086575655980376,
      "loss": 4.6158,
      "mean_token_accuracy": 0.23594296872615814,
      "num_tokens": 15267392.0,
      "step": 7375
    },
    {
      "entropy": 5.314075899124146,
      "epoch": 6.703316674238982,
      "grad_norm": 1.328125,
      "learning_rate": 0.00018054486206465952,
      "loss": 4.5877,
      "mean_token_accuracy": 0.23718660175800324,
      "num_tokens": 15277371.0,
      "step": 7380
    },
    {
      "entropy": 5.367406511306763,
      "epoch": 6.707860063607451,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001802242006281788,
      "loss": 4.5992,
      "mean_token_accuracy": 0.23754711598157882,
      "num_tokens": 15287490.0,
      "step": 7385
    },
    {
      "entropy": 5.286325454711914,
      "epoch": 6.71240345297592,
      "grad_norm": 1.25,
      "learning_rate": 0.00017990377304156165,
      "loss": 4.5966,
      "mean_token_accuracy": 0.2363538235425949,
      "num_tokens": 15298061.0,
      "step": 7390
    },
    {
      "entropy": 5.274067497253418,
      "epoch": 6.716946842344389,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001795835800954312,
      "loss": 4.5972,
      "mean_token_accuracy": 0.23638513684272766,
      "num_tokens": 15307862.0,
      "step": 7395
    },
    {
      "entropy": 5.250682735443116,
      "epoch": 6.721490231712858,
      "grad_norm": 1.15625,
      "learning_rate": 0.00017926362257983174,
      "loss": 4.6482,
      "mean_token_accuracy": 0.2363750234246254,
      "num_tokens": 15317693.0,
      "step": 7400
    },
    {
      "entropy": 5.325537538528442,
      "epoch": 6.726033621081327,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0001789439012842269,
      "loss": 4.6818,
      "mean_token_accuracy": 0.23202817738056183,
      "num_tokens": 15329000.0,
      "step": 7405
    },
    {
      "entropy": 5.380777311325073,
      "epoch": 6.730577010449796,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00017862441699749687,
      "loss": 4.6589,
      "mean_token_accuracy": 0.23581815510988235,
      "num_tokens": 15339967.0,
      "step": 7410
    },
    {
      "entropy": 5.379747915267944,
      "epoch": 6.735120399818264,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0001783051705079375,
      "loss": 4.6704,
      "mean_token_accuracy": 0.23266936093568802,
      "num_tokens": 15350478.0,
      "step": 7415
    },
    {
      "entropy": 5.386323261260986,
      "epoch": 6.739663789186733,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00017798616260325767,
      "loss": 4.6368,
      "mean_token_accuracy": 0.2262338027358055,
      "num_tokens": 15360728.0,
      "step": 7420
    },
    {
      "entropy": 5.338350296020508,
      "epoch": 6.744207178555202,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001776673940705778,
      "loss": 4.6482,
      "mean_token_accuracy": 0.23617025911808015,
      "num_tokens": 15370689.0,
      "step": 7425
    },
    {
      "entropy": 5.278313636779785,
      "epoch": 6.748750567923671,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001773488656964276,
      "loss": 4.5613,
      "mean_token_accuracy": 0.23841255903244019,
      "num_tokens": 15380987.0,
      "step": 7430
    },
    {
      "entropy": 5.3969635486602785,
      "epoch": 6.75329395729214,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000177030578266744,
      "loss": 4.6422,
      "mean_token_accuracy": 0.23722471594810485,
      "num_tokens": 15391989.0,
      "step": 7435
    },
    {
      "entropy": 5.376013994216919,
      "epoch": 6.757837346660609,
      "grad_norm": 1.25,
      "learning_rate": 0.00017671253256686963,
      "loss": 4.6576,
      "mean_token_accuracy": 0.23246419578790664,
      "num_tokens": 15402376.0,
      "step": 7440
    },
    {
      "entropy": 5.3169938087463375,
      "epoch": 6.762380736029078,
      "grad_norm": 1.328125,
      "learning_rate": 0.00017639472938155075,
      "loss": 4.6106,
      "mean_token_accuracy": 0.22831575125455855,
      "num_tokens": 15413034.0,
      "step": 7445
    },
    {
      "entropy": 5.263149213790894,
      "epoch": 6.766924125397547,
      "grad_norm": 1.15625,
      "learning_rate": 0.00017607716949493514,
      "loss": 4.5761,
      "mean_token_accuracy": 0.2286290094256401,
      "num_tokens": 15423378.0,
      "step": 7450
    },
    {
      "entropy": 5.28085765838623,
      "epoch": 6.771467514766016,
      "grad_norm": 1.171875,
      "learning_rate": 0.00017575985369057008,
      "loss": 4.5721,
      "mean_token_accuracy": 0.23402269631624223,
      "num_tokens": 15434092.0,
      "step": 7455
    },
    {
      "entropy": 5.281781816482544,
      "epoch": 6.776010904134484,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0001754427827514008,
      "loss": 4.5858,
      "mean_token_accuracy": 0.23148021399974822,
      "num_tokens": 15445178.0,
      "step": 7460
    },
    {
      "entropy": 5.356577062606812,
      "epoch": 6.780554293502953,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00017512595745976855,
      "loss": 4.6461,
      "mean_token_accuracy": 0.23317207098007203,
      "num_tokens": 15455456.0,
      "step": 7465
    },
    {
      "entropy": 5.290500020980835,
      "epoch": 6.785097682871422,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00017480937859740793,
      "loss": 4.6013,
      "mean_token_accuracy": 0.24214208722114564,
      "num_tokens": 15465805.0,
      "step": 7470
    },
    {
      "entropy": 5.362384557723999,
      "epoch": 6.789641072239891,
      "grad_norm": 1.1875,
      "learning_rate": 0.00017449304694544608,
      "loss": 4.6566,
      "mean_token_accuracy": 0.23121968358755113,
      "num_tokens": 15476590.0,
      "step": 7475
    },
    {
      "entropy": 5.319317865371704,
      "epoch": 6.79418446160836,
      "grad_norm": 1.1875,
      "learning_rate": 0.00017417696328439974,
      "loss": 4.594,
      "mean_token_accuracy": 0.2356454536318779,
      "num_tokens": 15486791.0,
      "step": 7480
    },
    {
      "entropy": 5.335692977905273,
      "epoch": 6.798727850976829,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00017386112839417402,
      "loss": 4.5989,
      "mean_token_accuracy": 0.24261886924505233,
      "num_tokens": 15497116.0,
      "step": 7485
    },
    {
      "entropy": 5.334956169128418,
      "epoch": 6.803271240345298,
      "grad_norm": 1.171875,
      "learning_rate": 0.00017354554305405994,
      "loss": 4.6225,
      "mean_token_accuracy": 0.23621838092803954,
      "num_tokens": 15506731.0,
      "step": 7490
    },
    {
      "entropy": 5.285718631744385,
      "epoch": 6.807814629713766,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001732302080427332,
      "loss": 4.6546,
      "mean_token_accuracy": 0.23435537666082382,
      "num_tokens": 15517333.0,
      "step": 7495
    },
    {
      "entropy": 5.27352123260498,
      "epoch": 6.812358019082235,
      "grad_norm": 1.21875,
      "learning_rate": 0.00017291512413825148,
      "loss": 4.6193,
      "mean_token_accuracy": 0.23730704486370086,
      "num_tokens": 15526995.0,
      "step": 7500
    },
    {
      "epoch": 6.812358019082235,
      "eval_entropy": 5.217221853344939,
      "eval_loss": 5.651242733001709,
      "eval_mean_token_accuracy": 0.1758443866358247,
      "eval_num_tokens": 15526995.0,
      "eval_runtime": 2.0324,
      "eval_samples_per_second": 1690.123,
      "eval_steps_per_second": 211.573,
      "step": 7500
    },
    {
      "entropy": 5.333926773071289,
      "epoch": 6.816901408450704,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00017260029211805297,
      "loss": 4.5868,
      "mean_token_accuracy": 0.23154065608978272,
      "num_tokens": 15536686.0,
      "step": 7505
    },
    {
      "entropy": 5.364414978027344,
      "epoch": 6.821444797819173,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00017228571275895434,
      "loss": 4.7095,
      "mean_token_accuracy": 0.22141506373882294,
      "num_tokens": 15547565.0,
      "step": 7510
    },
    {
      "entropy": 5.253812026977539,
      "epoch": 6.825988187187642,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00017197138683714902,
      "loss": 4.5873,
      "mean_token_accuracy": 0.23176054060459136,
      "num_tokens": 15557962.0,
      "step": 7515
    },
    {
      "entropy": 5.318046808242798,
      "epoch": 6.830531576556111,
      "grad_norm": 1.375,
      "learning_rate": 0.0001716573151282049,
      "loss": 4.5944,
      "mean_token_accuracy": 0.23960042744874954,
      "num_tokens": 15567164.0,
      "step": 7520
    },
    {
      "entropy": 5.292877578735352,
      "epoch": 6.83507496592458,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0001713434984070627,
      "loss": 4.5855,
      "mean_token_accuracy": 0.23072621822357178,
      "num_tokens": 15577917.0,
      "step": 7525
    },
    {
      "entropy": 5.362806367874145,
      "epoch": 6.839618355293049,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00017102993744803392,
      "loss": 4.6735,
      "mean_token_accuracy": 0.22852707803249359,
      "num_tokens": 15589106.0,
      "step": 7530
    },
    {
      "entropy": 5.345282125473022,
      "epoch": 6.844161744661518,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00017071663302479923,
      "loss": 4.6874,
      "mean_token_accuracy": 0.22209153622388839,
      "num_tokens": 15600368.0,
      "step": 7535
    },
    {
      "entropy": 5.350090885162354,
      "epoch": 6.848705134029986,
      "grad_norm": 1.265625,
      "learning_rate": 0.00017040358591040606,
      "loss": 4.5963,
      "mean_token_accuracy": 0.2324017733335495,
      "num_tokens": 15610671.0,
      "step": 7540
    },
    {
      "entropy": 5.230645036697387,
      "epoch": 6.853248523398455,
      "grad_norm": 1.234375,
      "learning_rate": 0.000170090796877267,
      "loss": 4.5588,
      "mean_token_accuracy": 0.2421345368027687,
      "num_tokens": 15620403.0,
      "step": 7545
    },
    {
      "entropy": 5.3569831371307375,
      "epoch": 6.857791912766924,
      "grad_norm": 1.265625,
      "learning_rate": 0.00016977826669715783,
      "loss": 4.6281,
      "mean_token_accuracy": 0.23967080563306808,
      "num_tokens": 15631260.0,
      "step": 7550
    },
    {
      "entropy": 5.358388137817383,
      "epoch": 6.862335302135393,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000169465996141216,
      "loss": 4.6127,
      "mean_token_accuracy": 0.2347136601805687,
      "num_tokens": 15641604.0,
      "step": 7555
    },
    {
      "entropy": 5.3474236011505125,
      "epoch": 6.866878691503862,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001691539859799377,
      "loss": 4.624,
      "mean_token_accuracy": 0.23391496539115905,
      "num_tokens": 15652469.0,
      "step": 7560
    },
    {
      "entropy": 5.29263858795166,
      "epoch": 6.871422080872331,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00016884223698317722,
      "loss": 4.6327,
      "mean_token_accuracy": 0.22715542912483216,
      "num_tokens": 15662615.0,
      "step": 7565
    },
    {
      "entropy": 5.3103269100189205,
      "epoch": 6.8759654702408,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0001685307499201441,
      "loss": 4.5866,
      "mean_token_accuracy": 0.22439027428627015,
      "num_tokens": 15674868.0,
      "step": 7570
    },
    {
      "entropy": 5.322635316848755,
      "epoch": 6.880508859609268,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0001682195255594019,
      "loss": 4.6791,
      "mean_token_accuracy": 0.23622027039527893,
      "num_tokens": 15686246.0,
      "step": 7575
    },
    {
      "entropy": 5.314529705047607,
      "epoch": 6.885052248977737,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00016790856466886544,
      "loss": 4.6131,
      "mean_token_accuracy": 0.2340119108557701,
      "num_tokens": 15697025.0,
      "step": 7580
    },
    {
      "entropy": 5.354397201538086,
      "epoch": 6.889595638346206,
      "grad_norm": 1.265625,
      "learning_rate": 0.00016759786801580018,
      "loss": 4.676,
      "mean_token_accuracy": 0.22606811076402664,
      "num_tokens": 15706636.0,
      "step": 7585
    },
    {
      "entropy": 5.3054282665252686,
      "epoch": 6.894139027714675,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00016728743636681897,
      "loss": 4.601,
      "mean_token_accuracy": 0.24114653766155242,
      "num_tokens": 15717319.0,
      "step": 7590
    },
    {
      "entropy": 5.320992040634155,
      "epoch": 6.898682417083144,
      "grad_norm": 1.375,
      "learning_rate": 0.00016697727048788118,
      "loss": 4.549,
      "mean_token_accuracy": 0.2416537031531334,
      "num_tokens": 15726546.0,
      "step": 7595
    },
    {
      "entropy": 5.299712610244751,
      "epoch": 6.903225806451613,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00016666737114429017,
      "loss": 4.6515,
      "mean_token_accuracy": 0.22778331488370895,
      "num_tokens": 15737176.0,
      "step": 7600
    },
    {
      "entropy": 5.311338424682617,
      "epoch": 6.907769195820082,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001663577391006919,
      "loss": 4.6183,
      "mean_token_accuracy": 0.24120287597179413,
      "num_tokens": 15747422.0,
      "step": 7605
    },
    {
      "entropy": 5.340631055831909,
      "epoch": 6.912312585188551,
      "grad_norm": 1.234375,
      "learning_rate": 0.00016604837512107263,
      "loss": 4.6487,
      "mean_token_accuracy": 0.22721280604600907,
      "num_tokens": 15757591.0,
      "step": 7610
    },
    {
      "entropy": 5.357366275787354,
      "epoch": 6.91685597455702,
      "grad_norm": 1.203125,
      "learning_rate": 0.00016573927996875714,
      "loss": 4.6793,
      "mean_token_accuracy": 0.22870995104312897,
      "num_tokens": 15768375.0,
      "step": 7615
    },
    {
      "entropy": 5.323020648956299,
      "epoch": 6.921399363925488,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00016543045440640697,
      "loss": 4.5641,
      "mean_token_accuracy": 0.23723311126232147,
      "num_tokens": 15778935.0,
      "step": 7620
    },
    {
      "entropy": 5.329158735275269,
      "epoch": 6.925942753293957,
      "grad_norm": 1.21875,
      "learning_rate": 0.00016512189919601865,
      "loss": 4.5868,
      "mean_token_accuracy": 0.24685502499341966,
      "num_tokens": 15788955.0,
      "step": 7625
    },
    {
      "entropy": 5.341229057312011,
      "epoch": 6.930486142662426,
      "grad_norm": 1.21875,
      "learning_rate": 0.0001648136150989214,
      "loss": 4.5722,
      "mean_token_accuracy": 0.23779306262731553,
      "num_tokens": 15798795.0,
      "step": 7630
    },
    {
      "entropy": 5.28424859046936,
      "epoch": 6.935029532030895,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001645056028757756,
      "loss": 4.6463,
      "mean_token_accuracy": 0.23225992321968078,
      "num_tokens": 15809251.0,
      "step": 7635
    },
    {
      "entropy": 5.300722122192383,
      "epoch": 6.939572921399364,
      "grad_norm": 1.234375,
      "learning_rate": 0.00016419786328657066,
      "loss": 4.5523,
      "mean_token_accuracy": 0.24180503636598588,
      "num_tokens": 15819418.0,
      "step": 7640
    },
    {
      "entropy": 5.3286542892456055,
      "epoch": 6.944116310767833,
      "grad_norm": 1.203125,
      "learning_rate": 0.00016389039709062376,
      "loss": 4.6062,
      "mean_token_accuracy": 0.23045997619628905,
      "num_tokens": 15829933.0,
      "step": 7645
    },
    {
      "entropy": 5.28349814414978,
      "epoch": 6.948659700136302,
      "grad_norm": 1.140625,
      "learning_rate": 0.00016358320504657677,
      "loss": 4.6294,
      "mean_token_accuracy": 0.2308754712343216,
      "num_tokens": 15841032.0,
      "step": 7650
    },
    {
      "entropy": 5.354779386520386,
      "epoch": 6.95320308950477,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00016327628791239576,
      "loss": 4.6787,
      "mean_token_accuracy": 0.22500632852315902,
      "num_tokens": 15851126.0,
      "step": 7655
    },
    {
      "entropy": 5.36037015914917,
      "epoch": 6.957746478873239,
      "grad_norm": 1.21875,
      "learning_rate": 0.00016296964644536816,
      "loss": 4.643,
      "mean_token_accuracy": 0.23086386322975158,
      "num_tokens": 15861832.0,
      "step": 7660
    },
    {
      "entropy": 5.3764831066131595,
      "epoch": 6.962289868241708,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001626632814021015,
      "loss": 4.6348,
      "mean_token_accuracy": 0.2333345592021942,
      "num_tokens": 15872295.0,
      "step": 7665
    },
    {
      "entropy": 5.360340213775634,
      "epoch": 6.966833257610177,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00016235719353852079,
      "loss": 4.6388,
      "mean_token_accuracy": 0.23226008862257003,
      "num_tokens": 15882453.0,
      "step": 7670
    },
    {
      "entropy": 5.32211537361145,
      "epoch": 6.971376646978646,
      "grad_norm": 1.359375,
      "learning_rate": 0.00016205138360986753,
      "loss": 4.6618,
      "mean_token_accuracy": 0.2305155336856842,
      "num_tokens": 15891962.0,
      "step": 7675
    },
    {
      "entropy": 5.313541650772095,
      "epoch": 6.975920036347115,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00016174585237069738,
      "loss": 4.5703,
      "mean_token_accuracy": 0.23474858254194259,
      "num_tokens": 15902309.0,
      "step": 7680
    },
    {
      "entropy": 5.296520233154297,
      "epoch": 6.980463425715584,
      "grad_norm": 1.171875,
      "learning_rate": 0.00016144060057487824,
      "loss": 4.5936,
      "mean_token_accuracy": 0.2288145825266838,
      "num_tokens": 15912789.0,
      "step": 7685
    },
    {
      "entropy": 5.37167706489563,
      "epoch": 6.985006815084053,
      "grad_norm": 1.234375,
      "learning_rate": 0.00016113562897558849,
      "loss": 4.6698,
      "mean_token_accuracy": 0.229607093334198,
      "num_tokens": 15922722.0,
      "step": 7690
    },
    {
      "entropy": 5.295284748077393,
      "epoch": 6.989550204452522,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0001608309383253154,
      "loss": 4.5449,
      "mean_token_accuracy": 0.24083372354507446,
      "num_tokens": 15933069.0,
      "step": 7695
    },
    {
      "entropy": 5.381824493408203,
      "epoch": 6.99409359382099,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001605265293758528,
      "loss": 4.6398,
      "mean_token_accuracy": 0.2289839953184128,
      "num_tokens": 15942427.0,
      "step": 7700
    },
    {
      "entropy": 5.337258052825928,
      "epoch": 6.998636983189459,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00016022240287829947,
      "loss": 4.6153,
      "mean_token_accuracy": 0.22795629352331162,
      "num_tokens": 15952765.0,
      "step": 7705
    },
    {
      "entropy": 5.268085532718235,
      "epoch": 7.002726033621081,
      "grad_norm": 1.21875,
      "learning_rate": 0.00015991855958305724,
      "loss": 4.5904,
      "mean_token_accuracy": 0.2403159240881602,
      "num_tokens": 15963025.0,
      "step": 7710
    },
    {
      "entropy": 5.285863876342773,
      "epoch": 7.00726942298955,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00015961500023982945,
      "loss": 4.5221,
      "mean_token_accuracy": 0.24898781776428222,
      "num_tokens": 15973425.0,
      "step": 7715
    },
    {
      "entropy": 5.3499805450439455,
      "epoch": 7.011812812358019,
      "grad_norm": 1.25,
      "learning_rate": 0.00015931172559761848,
      "loss": 4.557,
      "mean_token_accuracy": 0.2467520073056221,
      "num_tokens": 15984064.0,
      "step": 7720
    },
    {
      "entropy": 5.3726812362670895,
      "epoch": 7.016356201726488,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001590087364047243,
      "loss": 4.5329,
      "mean_token_accuracy": 0.23927863985300063,
      "num_tokens": 15994512.0,
      "step": 7725
    },
    {
      "entropy": 5.333837175369263,
      "epoch": 7.020899591094957,
      "grad_norm": 1.28125,
      "learning_rate": 0.00015870603340874267,
      "loss": 4.4804,
      "mean_token_accuracy": 0.2470007672905922,
      "num_tokens": 16004140.0,
      "step": 7730
    },
    {
      "entropy": 5.268793725967408,
      "epoch": 7.025442980463426,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00015840361735656333,
      "loss": 4.4615,
      "mean_token_accuracy": 0.243349152803421,
      "num_tokens": 16013710.0,
      "step": 7735
    },
    {
      "entropy": 5.384647846221924,
      "epoch": 7.029986369831895,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00015810148899436747,
      "loss": 4.5329,
      "mean_token_accuracy": 0.2446380838751793,
      "num_tokens": 16023846.0,
      "step": 7740
    },
    {
      "entropy": 5.308625602722168,
      "epoch": 7.034529759200363,
      "grad_norm": 1.21875,
      "learning_rate": 0.00015779964906762706,
      "loss": 4.4798,
      "mean_token_accuracy": 0.2413966178894043,
      "num_tokens": 16034327.0,
      "step": 7745
    },
    {
      "entropy": 5.308808660507202,
      "epoch": 7.039073148568832,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00015749809832110197,
      "loss": 4.5178,
      "mean_token_accuracy": 0.24670595228672026,
      "num_tokens": 16045055.0,
      "step": 7750
    },
    {
      "entropy": 5.274844169616699,
      "epoch": 7.043616537937301,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001571968374988389,
      "loss": 4.5237,
      "mean_token_accuracy": 0.23863110393285752,
      "num_tokens": 16055198.0,
      "step": 7755
    },
    {
      "entropy": 5.278604173660279,
      "epoch": 7.04815992730577,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00015689586734416876,
      "loss": 4.4692,
      "mean_token_accuracy": 0.25401477664709093,
      "num_tokens": 16065030.0,
      "step": 7760
    },
    {
      "entropy": 5.355061769485474,
      "epoch": 7.052703316674239,
      "grad_norm": 1.234375,
      "learning_rate": 0.00015659518859970545,
      "loss": 4.5663,
      "mean_token_accuracy": 0.2451522469520569,
      "num_tokens": 16075974.0,
      "step": 7765
    },
    {
      "entropy": 5.357634878158569,
      "epoch": 7.057246706042708,
      "grad_norm": 1.25,
      "learning_rate": 0.0001562948020073441,
      "loss": 4.568,
      "mean_token_accuracy": 0.2439857840538025,
      "num_tokens": 16086835.0,
      "step": 7770
    },
    {
      "entropy": 5.321579313278198,
      "epoch": 7.061790095411177,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00015599470830825868,
      "loss": 4.5435,
      "mean_token_accuracy": 0.24020788371562957,
      "num_tokens": 16098116.0,
      "step": 7775
    },
    {
      "entropy": 5.343589019775391,
      "epoch": 7.066333484779646,
      "grad_norm": 1.28125,
      "learning_rate": 0.00015569490824290054,
      "loss": 4.5639,
      "mean_token_accuracy": 0.22766242027282715,
      "num_tokens": 16109185.0,
      "step": 7780
    },
    {
      "entropy": 5.292613124847412,
      "epoch": 7.070876874148114,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00015539540255099643,
      "loss": 4.5124,
      "mean_token_accuracy": 0.2427526280283928,
      "num_tokens": 16120025.0,
      "step": 7785
    },
    {
      "entropy": 5.3121559619903564,
      "epoch": 7.075420263516583,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00015509619197154705,
      "loss": 4.4974,
      "mean_token_accuracy": 0.24864935278892517,
      "num_tokens": 16131145.0,
      "step": 7790
    },
    {
      "entropy": 5.382412910461426,
      "epoch": 7.079963652885052,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00015479727724282476,
      "loss": 4.5573,
      "mean_token_accuracy": 0.23704738467931746,
      "num_tokens": 16141063.0,
      "step": 7795
    },
    {
      "entropy": 5.330224704742432,
      "epoch": 7.084507042253521,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00015449865910237184,
      "loss": 4.5094,
      "mean_token_accuracy": 0.24340772330760957,
      "num_tokens": 16151059.0,
      "step": 7800
    },
    {
      "entropy": 5.309969425201416,
      "epoch": 7.08905043162199,
      "grad_norm": 1.328125,
      "learning_rate": 0.00015420033828699882,
      "loss": 4.5064,
      "mean_token_accuracy": 0.2527839601039886,
      "num_tokens": 16160679.0,
      "step": 7805
    },
    {
      "entropy": 5.290482997894287,
      "epoch": 7.093593820990459,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001539023155327829,
      "loss": 4.5235,
      "mean_token_accuracy": 0.23772758543491362,
      "num_tokens": 16170269.0,
      "step": 7810
    },
    {
      "entropy": 5.296168994903565,
      "epoch": 7.098137210358928,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00015360459157506525,
      "loss": 4.4987,
      "mean_token_accuracy": 0.24687401205301285,
      "num_tokens": 16180496.0,
      "step": 7815
    },
    {
      "entropy": 5.3271105766296385,
      "epoch": 7.102680599727397,
      "grad_norm": 1.21875,
      "learning_rate": 0.00015330716714845048,
      "loss": 4.5325,
      "mean_token_accuracy": 0.23923449069261551,
      "num_tokens": 16191043.0,
      "step": 7820
    },
    {
      "entropy": 5.3813300132751465,
      "epoch": 7.107223989095866,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001530100429868035,
      "loss": 4.6822,
      "mean_token_accuracy": 0.23118032813072203,
      "num_tokens": 16201551.0,
      "step": 7825
    },
    {
      "entropy": 5.353553676605225,
      "epoch": 7.111767378464334,
      "grad_norm": 1.171875,
      "learning_rate": 0.00015271321982324892,
      "loss": 4.5437,
      "mean_token_accuracy": 0.2366670235991478,
      "num_tokens": 16212928.0,
      "step": 7830
    },
    {
      "entropy": 5.285678052902222,
      "epoch": 7.116310767832803,
      "grad_norm": 1.265625,
      "learning_rate": 0.00015241669839016818,
      "loss": 4.5197,
      "mean_token_accuracy": 0.24427150338888168,
      "num_tokens": 16223233.0,
      "step": 7835
    },
    {
      "entropy": 5.305990934371948,
      "epoch": 7.120854157201272,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00015212047941919854,
      "loss": 4.6108,
      "mean_token_accuracy": 0.23421745300292968,
      "num_tokens": 16234902.0,
      "step": 7840
    },
    {
      "entropy": 5.334067106246948,
      "epoch": 7.125397546569741,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00015182456364123086,
      "loss": 4.5615,
      "mean_token_accuracy": 0.24063838124275208,
      "num_tokens": 16244846.0,
      "step": 7845
    },
    {
      "entropy": 5.32163553237915,
      "epoch": 7.12994093593821,
      "grad_norm": 1.40625,
      "learning_rate": 0.00015152895178640788,
      "loss": 4.5243,
      "mean_token_accuracy": 0.24841463416814805,
      "num_tokens": 16255067.0,
      "step": 7850
    },
    {
      "entropy": 5.342627000808716,
      "epoch": 7.134484325306679,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001512336445841224,
      "loss": 4.5578,
      "mean_token_accuracy": 0.23349680304527282,
      "num_tokens": 16265482.0,
      "step": 7855
    },
    {
      "entropy": 5.369259452819824,
      "epoch": 7.139027714675148,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00015093864276301573,
      "loss": 4.5587,
      "mean_token_accuracy": 0.24331467896699904,
      "num_tokens": 16275319.0,
      "step": 7860
    },
    {
      "entropy": 5.278583526611328,
      "epoch": 7.143571104043617,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00015064394705097546,
      "loss": 4.5318,
      "mean_token_accuracy": 0.23822576850652694,
      "num_tokens": 16284889.0,
      "step": 7865
    },
    {
      "entropy": 5.349502182006836,
      "epoch": 7.148114493412085,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00015034955817513397,
      "loss": 4.5533,
      "mean_token_accuracy": 0.24058006703853607,
      "num_tokens": 16295259.0,
      "step": 7870
    },
    {
      "entropy": 5.261496353149414,
      "epoch": 7.152657882780554,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001500554768618665,
      "loss": 4.5303,
      "mean_token_accuracy": 0.2406397596001625,
      "num_tokens": 16305489.0,
      "step": 7875
    },
    {
      "entropy": 5.295884132385254,
      "epoch": 7.157201272149023,
      "grad_norm": 1.421875,
      "learning_rate": 0.00014976170383678954,
      "loss": 4.4927,
      "mean_token_accuracy": 0.254702752828598,
      "num_tokens": 16315167.0,
      "step": 7880
    },
    {
      "entropy": 5.296162939071655,
      "epoch": 7.161744661517492,
      "grad_norm": 1.25,
      "learning_rate": 0.00014946823982475883,
      "loss": 4.4834,
      "mean_token_accuracy": 0.25267650485038756,
      "num_tokens": 16325059.0,
      "step": 7885
    },
    {
      "entropy": 5.353860473632812,
      "epoch": 7.166288050885961,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00014917508554986754,
      "loss": 4.5972,
      "mean_token_accuracy": 0.23747731149196624,
      "num_tokens": 16336647.0,
      "step": 7890
    },
    {
      "entropy": 5.290485525131226,
      "epoch": 7.17083144025443,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00014888224173544476,
      "loss": 4.5183,
      "mean_token_accuracy": 0.2372318759560585,
      "num_tokens": 16346146.0,
      "step": 7895
    },
    {
      "entropy": 5.3550129413604735,
      "epoch": 7.175374829622899,
      "grad_norm": 1.53125,
      "learning_rate": 0.00014858970910405368,
      "loss": 4.6217,
      "mean_token_accuracy": 0.2355572149157524,
      "num_tokens": 16356124.0,
      "step": 7900
    },
    {
      "entropy": 5.332889413833618,
      "epoch": 7.179918218991368,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00014829748837748906,
      "loss": 4.5646,
      "mean_token_accuracy": 0.23635377436876298,
      "num_tokens": 16367197.0,
      "step": 7905
    },
    {
      "entropy": 5.31535120010376,
      "epoch": 7.184461608359836,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001480055802767769,
      "loss": 4.5261,
      "mean_token_accuracy": 0.24338396042585372,
      "num_tokens": 16376757.0,
      "step": 7910
    },
    {
      "entropy": 5.271803665161133,
      "epoch": 7.189004997728305,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00014771398552217114,
      "loss": 4.4544,
      "mean_token_accuracy": 0.256404834985733,
      "num_tokens": 16387454.0,
      "step": 7915
    },
    {
      "entropy": 5.322522401809692,
      "epoch": 7.193548387096774,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00014742270483315319,
      "loss": 4.5465,
      "mean_token_accuracy": 0.23790270686149598,
      "num_tokens": 16397852.0,
      "step": 7920
    },
    {
      "entropy": 5.269116401672363,
      "epoch": 7.198091776465243,
      "grad_norm": 1.3125,
      "learning_rate": 0.00014713173892842878,
      "loss": 4.515,
      "mean_token_accuracy": 0.24470317363739014,
      "num_tokens": 16407752.0,
      "step": 7925
    },
    {
      "entropy": 5.264682817459106,
      "epoch": 7.202635165833712,
      "grad_norm": 1.203125,
      "learning_rate": 0.00014684108852592765,
      "loss": 4.5469,
      "mean_token_accuracy": 0.23103222995996475,
      "num_tokens": 16418500.0,
      "step": 7930
    },
    {
      "entropy": 5.338938283920288,
      "epoch": 7.207178555202181,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00014655075434280064,
      "loss": 4.5549,
      "mean_token_accuracy": 0.2417147547006607,
      "num_tokens": 16428706.0,
      "step": 7935
    },
    {
      "entropy": 5.390133428573608,
      "epoch": 7.21172194457065,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00014626073709541853,
      "loss": 4.5534,
      "mean_token_accuracy": 0.24172492027282716,
      "num_tokens": 16438184.0,
      "step": 7940
    },
    {
      "entropy": 5.299207067489624,
      "epoch": 7.216265333939119,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00014597103749937,
      "loss": 4.5119,
      "mean_token_accuracy": 0.24391905665397645,
      "num_tokens": 16449340.0,
      "step": 7945
    },
    {
      "entropy": 5.32843770980835,
      "epoch": 7.220808723307587,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00014568165626946012,
      "loss": 4.5145,
      "mean_token_accuracy": 0.23822322785854338,
      "num_tokens": 16459604.0,
      "step": 7950
    },
    {
      "entropy": 5.379015398025513,
      "epoch": 7.225352112676056,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00014539259411970823,
      "loss": 4.6406,
      "mean_token_accuracy": 0.2340823546051979,
      "num_tokens": 16470150.0,
      "step": 7955
    },
    {
      "entropy": 5.341042613983154,
      "epoch": 7.229895502044525,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001451038517633465,
      "loss": 4.5699,
      "mean_token_accuracy": 0.24702875167131425,
      "num_tokens": 16480797.0,
      "step": 7960
    },
    {
      "entropy": 5.363569450378418,
      "epoch": 7.234438891412994,
      "grad_norm": 1.234375,
      "learning_rate": 0.00014481542991281782,
      "loss": 4.6118,
      "mean_token_accuracy": 0.23775914311408997,
      "num_tokens": 16492629.0,
      "step": 7965
    },
    {
      "entropy": 5.308783531188965,
      "epoch": 7.238982280781463,
      "grad_norm": 1.265625,
      "learning_rate": 0.00014452732927977465,
      "loss": 4.4922,
      "mean_token_accuracy": 0.24166162461042404,
      "num_tokens": 16503225.0,
      "step": 7970
    },
    {
      "entropy": 5.336436605453491,
      "epoch": 7.243525670149932,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001442395505750766,
      "loss": 4.4901,
      "mean_token_accuracy": 0.2524169161915779,
      "num_tokens": 16512446.0,
      "step": 7975
    },
    {
      "entropy": 5.311456537246704,
      "epoch": 7.248069059518401,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00014395209450878899,
      "loss": 4.5316,
      "mean_token_accuracy": 0.24552505761384963,
      "num_tokens": 16522944.0,
      "step": 7980
    },
    {
      "entropy": 5.283830976486206,
      "epoch": 7.25261244888687,
      "grad_norm": 1.328125,
      "learning_rate": 0.000143664961790181,
      "loss": 4.5174,
      "mean_token_accuracy": 0.24236050695180894,
      "num_tokens": 16533100.0,
      "step": 7985
    },
    {
      "entropy": 5.3482073783874515,
      "epoch": 7.257155838255338,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001433781531277243,
      "loss": 4.603,
      "mean_token_accuracy": 0.23179831206798554,
      "num_tokens": 16542893.0,
      "step": 7990
    },
    {
      "entropy": 5.318871784210205,
      "epoch": 7.261699227623807,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001430916692290905,
      "loss": 4.5402,
      "mean_token_accuracy": 0.2466399848461151,
      "num_tokens": 16553209.0,
      "step": 7995
    },
    {
      "entropy": 5.350064611434936,
      "epoch": 7.266242616992276,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00014280551080115032,
      "loss": 4.5546,
      "mean_token_accuracy": 0.24203133285045625,
      "num_tokens": 16564574.0,
      "step": 8000
    },
    {
      "epoch": 7.266242616992276,
      "eval_entropy": 5.213549815776736,
      "eval_loss": 5.6612443923950195,
      "eval_mean_token_accuracy": 0.17497689264804817,
      "eval_num_tokens": 16564574.0,
      "eval_runtime": 2.0308,
      "eval_samples_per_second": 1691.435,
      "eval_steps_per_second": 211.737,
      "step": 8000
    },
    {
      "entropy": 5.393393898010254,
      "epoch": 7.270786006360745,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00014251967854997106,
      "loss": 4.5574,
      "mean_token_accuracy": 0.242436283826828,
      "num_tokens": 16574517.0,
      "step": 8005
    },
    {
      "entropy": 5.347425889968872,
      "epoch": 7.275329395729214,
      "grad_norm": 1.25,
      "learning_rate": 0.00014223417318081572,
      "loss": 4.5674,
      "mean_token_accuracy": 0.23733702749013902,
      "num_tokens": 16585311.0,
      "step": 8010
    },
    {
      "entropy": 5.2701653957366945,
      "epoch": 7.279872785097683,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00014194899539814,
      "loss": 4.5515,
      "mean_token_accuracy": 0.23599171340465547,
      "num_tokens": 16596055.0,
      "step": 8015
    },
    {
      "entropy": 5.246894693374633,
      "epoch": 7.284416174466152,
      "grad_norm": 1.234375,
      "learning_rate": 0.00014166414590559197,
      "loss": 4.495,
      "mean_token_accuracy": 0.2418297052383423,
      "num_tokens": 16606076.0,
      "step": 8020
    },
    {
      "entropy": 5.3322216987609865,
      "epoch": 7.288959563834621,
      "grad_norm": 1.390625,
      "learning_rate": 0.00014137962540600944,
      "loss": 4.5513,
      "mean_token_accuracy": 0.2384777382016182,
      "num_tokens": 16616740.0,
      "step": 8025
    },
    {
      "entropy": 5.3182238101959225,
      "epoch": 7.29350295320309,
      "grad_norm": 1.28125,
      "learning_rate": 0.00014109543460141837,
      "loss": 4.5005,
      "mean_token_accuracy": 0.24887488931417465,
      "num_tokens": 16627412.0,
      "step": 8030
    },
    {
      "entropy": 5.292384481430053,
      "epoch": 7.298046342571558,
      "grad_norm": 1.1875,
      "learning_rate": 0.0001408115741930313,
      "loss": 4.5192,
      "mean_token_accuracy": 0.24316384196281432,
      "num_tokens": 16637032.0,
      "step": 8035
    },
    {
      "entropy": 5.282077407836914,
      "epoch": 7.302589731940027,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00014052804488124575,
      "loss": 4.5135,
      "mean_token_accuracy": 0.2439026191830635,
      "num_tokens": 16646596.0,
      "step": 8040
    },
    {
      "entropy": 5.3078954219818115,
      "epoch": 7.307133121308496,
      "grad_norm": 1.28125,
      "learning_rate": 0.00014024484736564196,
      "loss": 4.4747,
      "mean_token_accuracy": 0.24844290018081666,
      "num_tokens": 16656404.0,
      "step": 8045
    },
    {
      "entropy": 5.327663087844849,
      "epoch": 7.311676510676965,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0001399619823449817,
      "loss": 4.5339,
      "mean_token_accuracy": 0.24329713732004166,
      "num_tokens": 16666372.0,
      "step": 8050
    },
    {
      "entropy": 5.305889892578125,
      "epoch": 7.316219900045434,
      "grad_norm": 1.21875,
      "learning_rate": 0.00013967945051720614,
      "loss": 4.5201,
      "mean_token_accuracy": 0.2474939152598381,
      "num_tokens": 16677150.0,
      "step": 8055
    },
    {
      "entropy": 5.307173442840576,
      "epoch": 7.320763289413903,
      "grad_norm": 1.125,
      "learning_rate": 0.00013939725257943475,
      "loss": 4.6086,
      "mean_token_accuracy": 0.23767561614513397,
      "num_tokens": 16688712.0,
      "step": 8060
    },
    {
      "entropy": 5.315055418014526,
      "epoch": 7.325306678782372,
      "grad_norm": 1.140625,
      "learning_rate": 0.00013911538922796275,
      "loss": 4.5265,
      "mean_token_accuracy": 0.24296061247587203,
      "num_tokens": 16699214.0,
      "step": 8065
    },
    {
      "entropy": 5.3228474140167235,
      "epoch": 7.32985006815084,
      "grad_norm": 1.328125,
      "learning_rate": 0.00013883386115825984,
      "loss": 4.5419,
      "mean_token_accuracy": 0.24119144678115845,
      "num_tokens": 16709294.0,
      "step": 8070
    },
    {
      "entropy": 5.3132078647613525,
      "epoch": 7.334393457519309,
      "grad_norm": 1.359375,
      "learning_rate": 0.00013855266906496867,
      "loss": 4.4913,
      "mean_token_accuracy": 0.24817643761634828,
      "num_tokens": 16718761.0,
      "step": 8075
    },
    {
      "entropy": 5.331232261657715,
      "epoch": 7.338936846887778,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00013827181364190288,
      "loss": 4.6174,
      "mean_token_accuracy": 0.2306366443634033,
      "num_tokens": 16729075.0,
      "step": 8080
    },
    {
      "entropy": 5.307704973220825,
      "epoch": 7.343480236256247,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00013799129558204505,
      "loss": 4.4786,
      "mean_token_accuracy": 0.2467479318380356,
      "num_tokens": 16739293.0,
      "step": 8085
    },
    {
      "entropy": 5.282048606872559,
      "epoch": 7.348023625624716,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00013771111557754577,
      "loss": 4.5535,
      "mean_token_accuracy": 0.237929630279541,
      "num_tokens": 16749214.0,
      "step": 8090
    },
    {
      "entropy": 5.259844064712524,
      "epoch": 7.352567014993185,
      "grad_norm": 1.3125,
      "learning_rate": 0.00013743127431972124,
      "loss": 4.495,
      "mean_token_accuracy": 0.24802033603191376,
      "num_tokens": 16758715.0,
      "step": 8095
    },
    {
      "entropy": 5.343987703323364,
      "epoch": 7.357110404361654,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001371517724990522,
      "loss": 4.5971,
      "mean_token_accuracy": 0.2386366531252861,
      "num_tokens": 16768965.0,
      "step": 8100
    },
    {
      "entropy": 5.3121928691864015,
      "epoch": 7.361653793730123,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001368726108051813,
      "loss": 4.4472,
      "mean_token_accuracy": 0.2526306644082069,
      "num_tokens": 16778368.0,
      "step": 8105
    },
    {
      "entropy": 5.347832822799683,
      "epoch": 7.366197183098592,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00013659378992691244,
      "loss": 4.5669,
      "mean_token_accuracy": 0.23978629261255263,
      "num_tokens": 16789396.0,
      "step": 8110
    },
    {
      "entropy": 5.348227262496948,
      "epoch": 7.37074057246706,
      "grad_norm": 1.28125,
      "learning_rate": 0.00013631531055220835,
      "loss": 4.5517,
      "mean_token_accuracy": 0.24278114438056947,
      "num_tokens": 16800063.0,
      "step": 8115
    },
    {
      "entropy": 5.295643997192383,
      "epoch": 7.375283961835529,
      "grad_norm": 1.234375,
      "learning_rate": 0.00013603717336818934,
      "loss": 4.5373,
      "mean_token_accuracy": 0.24183764606714248,
      "num_tokens": 16810344.0,
      "step": 8120
    },
    {
      "entropy": 5.252744913101196,
      "epoch": 7.379827351203998,
      "grad_norm": 1.3125,
      "learning_rate": 0.00013575937906113095,
      "loss": 4.4853,
      "mean_token_accuracy": 0.24864417165517808,
      "num_tokens": 16820233.0,
      "step": 8125
    },
    {
      "entropy": 5.336929559707642,
      "epoch": 7.384370740572467,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001354819283164634,
      "loss": 4.5802,
      "mean_token_accuracy": 0.2416494980454445,
      "num_tokens": 16829761.0,
      "step": 8130
    },
    {
      "entropy": 5.3588134765625,
      "epoch": 7.388914129940936,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001352048218187686,
      "loss": 4.6331,
      "mean_token_accuracy": 0.23698078095912933,
      "num_tokens": 16841068.0,
      "step": 8135
    },
    {
      "entropy": 5.331047105789184,
      "epoch": 7.393457519309405,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00013492806025177935,
      "loss": 4.5046,
      "mean_token_accuracy": 0.24379788488149642,
      "num_tokens": 16850733.0,
      "step": 8140
    },
    {
      "entropy": 5.306151437759399,
      "epoch": 7.398000908677874,
      "grad_norm": 1.359375,
      "learning_rate": 0.00013465164429837716,
      "loss": 4.5388,
      "mean_token_accuracy": 0.24108296632766724,
      "num_tokens": 16860416.0,
      "step": 8145
    },
    {
      "entropy": 5.35842113494873,
      "epoch": 7.402544298046343,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0001343755746405912,
      "loss": 4.5479,
      "mean_token_accuracy": 0.2501911878585815,
      "num_tokens": 16870662.0,
      "step": 8150
    },
    {
      "entropy": 5.31092267036438,
      "epoch": 7.407087687414811,
      "grad_norm": 1.3125,
      "learning_rate": 0.00013409985195959574,
      "loss": 4.6015,
      "mean_token_accuracy": 0.23063389509916304,
      "num_tokens": 16881758.0,
      "step": 8155
    },
    {
      "entropy": 5.330673265457153,
      "epoch": 7.41163107678328,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00013382447693570908,
      "loss": 4.5596,
      "mean_token_accuracy": 0.23482524752616882,
      "num_tokens": 16892399.0,
      "step": 8160
    },
    {
      "entropy": 5.309145784378051,
      "epoch": 7.416174466151749,
      "grad_norm": 1.34375,
      "learning_rate": 0.00013354945024839166,
      "loss": 4.5912,
      "mean_token_accuracy": 0.23965176641941072,
      "num_tokens": 16903696.0,
      "step": 8165
    },
    {
      "entropy": 5.305388927459717,
      "epoch": 7.420717855520218,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00013327477257624466,
      "loss": 4.5571,
      "mean_token_accuracy": 0.23953332751989365,
      "num_tokens": 16914326.0,
      "step": 8170
    },
    {
      "entropy": 5.301921367645264,
      "epoch": 7.425261244888687,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001330004445970078,
      "loss": 4.4717,
      "mean_token_accuracy": 0.25014871507883074,
      "num_tokens": 16924533.0,
      "step": 8175
    },
    {
      "entropy": 5.315178155899048,
      "epoch": 7.429804634257156,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00013272646698755816,
      "loss": 4.4964,
      "mean_token_accuracy": 0.24694067388772964,
      "num_tokens": 16934157.0,
      "step": 8180
    },
    {
      "entropy": 5.245381641387939,
      "epoch": 7.434348023625625,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00013245284042390814,
      "loss": 4.4622,
      "mean_token_accuracy": 0.24792222678661346,
      "num_tokens": 16944148.0,
      "step": 8185
    },
    {
      "entropy": 5.3258641242980955,
      "epoch": 7.438891412994094,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00013217956558120436,
      "loss": 4.6143,
      "mean_token_accuracy": 0.23548694849014282,
      "num_tokens": 16956388.0,
      "step": 8190
    },
    {
      "entropy": 5.302282619476318,
      "epoch": 7.443434802362562,
      "grad_norm": 1.484375,
      "learning_rate": 0.000131906643133725,
      "loss": 4.5636,
      "mean_token_accuracy": 0.24130917489528655,
      "num_tokens": 16965216.0,
      "step": 8195
    },
    {
      "entropy": 5.283529329299927,
      "epoch": 7.447978191731031,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00013163407375487926,
      "loss": 4.5382,
      "mean_token_accuracy": 0.24347199350595475,
      "num_tokens": 16976250.0,
      "step": 8200
    },
    {
      "entropy": 5.281372022628784,
      "epoch": 7.4525215810995,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00013136185811720496,
      "loss": 4.5736,
      "mean_token_accuracy": 0.23704758286476135,
      "num_tokens": 16987375.0,
      "step": 8205
    },
    {
      "entropy": 5.245068788528442,
      "epoch": 7.457064970467969,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001310899968923671,
      "loss": 4.5755,
      "mean_token_accuracy": 0.2469884544610977,
      "num_tokens": 16997526.0,
      "step": 8210
    },
    {
      "entropy": 5.32345404624939,
      "epoch": 7.461608359836438,
      "grad_norm": 1.25,
      "learning_rate": 0.0001308184907511562,
      "loss": 4.584,
      "mean_token_accuracy": 0.24014361649751664,
      "num_tokens": 17008342.0,
      "step": 8215
    },
    {
      "entropy": 5.327227783203125,
      "epoch": 7.466151749204907,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00013054734036348675,
      "loss": 4.568,
      "mean_token_accuracy": 0.24140361398458482,
      "num_tokens": 17019555.0,
      "step": 8220
    },
    {
      "entropy": 5.340827131271363,
      "epoch": 7.470695138573376,
      "grad_norm": 1.1875,
      "learning_rate": 0.00013027654639839536,
      "loss": 4.5805,
      "mean_token_accuracy": 0.23932118117809295,
      "num_tokens": 17030686.0,
      "step": 8225
    },
    {
      "entropy": 5.34961895942688,
      "epoch": 7.475238527941845,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00013000610952403917,
      "loss": 4.5326,
      "mean_token_accuracy": 0.24906973540782928,
      "num_tokens": 17041089.0,
      "step": 8230
    },
    {
      "entropy": 5.259250736236572,
      "epoch": 7.479781917310314,
      "grad_norm": 1.375,
      "learning_rate": 0.00012973603040769427,
      "loss": 4.4979,
      "mean_token_accuracy": 0.2420292839407921,
      "num_tokens": 17050194.0,
      "step": 8235
    },
    {
      "entropy": 5.311214303970337,
      "epoch": 7.484325306678782,
      "grad_norm": 1.484375,
      "learning_rate": 0.00012946630971575414,
      "loss": 4.5331,
      "mean_token_accuracy": 0.2388460010290146,
      "num_tokens": 17060596.0,
      "step": 8240
    },
    {
      "entropy": 5.34862756729126,
      "epoch": 7.488868696047251,
      "grad_norm": 1.296875,
      "learning_rate": 0.00012919694811372768,
      "loss": 4.5845,
      "mean_token_accuracy": 0.23341052532196044,
      "num_tokens": 17069763.0,
      "step": 8245
    },
    {
      "entropy": 5.265694570541382,
      "epoch": 7.49341208541572,
      "grad_norm": 1.25,
      "learning_rate": 0.00012892794626623793,
      "loss": 4.5118,
      "mean_token_accuracy": 0.24516717940568925,
      "num_tokens": 17080031.0,
      "step": 8250
    },
    {
      "entropy": 5.311520528793335,
      "epoch": 7.497955474784189,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001286593048370201,
      "loss": 4.5403,
      "mean_token_accuracy": 0.2432851806282997,
      "num_tokens": 17091339.0,
      "step": 8255
    },
    {
      "entropy": 5.2760899543762205,
      "epoch": 7.502498864152658,
      "grad_norm": 1.203125,
      "learning_rate": 0.00012839102448892033,
      "loss": 4.5017,
      "mean_token_accuracy": 0.24523537755012512,
      "num_tokens": 17102264.0,
      "step": 8260
    },
    {
      "entropy": 5.311911296844483,
      "epoch": 7.507042253521127,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00012812310588389367,
      "loss": 4.5489,
      "mean_token_accuracy": 0.23519348800182344,
      "num_tokens": 17112493.0,
      "step": 8265
    },
    {
      "entropy": 5.294253349304199,
      "epoch": 7.511585642889596,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00012785554968300266,
      "loss": 4.5046,
      "mean_token_accuracy": 0.24863037317991257,
      "num_tokens": 17122037.0,
      "step": 8270
    },
    {
      "entropy": 5.310962772369384,
      "epoch": 7.516129032258064,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00012758835654641548,
      "loss": 4.5439,
      "mean_token_accuracy": 0.24236177504062653,
      "num_tokens": 17133403.0,
      "step": 8275
    },
    {
      "entropy": 5.233997011184693,
      "epoch": 7.520672421626533,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000127321527133405,
      "loss": 4.493,
      "mean_token_accuracy": 0.2417943388223648,
      "num_tokens": 17143619.0,
      "step": 8280
    },
    {
      "entropy": 5.3361443996429445,
      "epoch": 7.525215810995002,
      "grad_norm": 1.328125,
      "learning_rate": 0.00012705506210234583,
      "loss": 4.5862,
      "mean_token_accuracy": 0.2364388585090637,
      "num_tokens": 17154811.0,
      "step": 8285
    },
    {
      "entropy": 5.272669744491577,
      "epoch": 7.529759200363471,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00012678896211071422,
      "loss": 4.5116,
      "mean_token_accuracy": 0.24047232717275618,
      "num_tokens": 17166039.0,
      "step": 8290
    },
    {
      "entropy": 5.3390312671661375,
      "epoch": 7.53430258973194,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001265232278150854,
      "loss": 4.5317,
      "mean_token_accuracy": 0.24058979153633117,
      "num_tokens": 17176555.0,
      "step": 8295
    },
    {
      "entropy": 5.26589994430542,
      "epoch": 7.538845979100409,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0001262578598711323,
      "loss": 4.5746,
      "mean_token_accuracy": 0.2331720173358917,
      "num_tokens": 17188220.0,
      "step": 8300
    },
    {
      "entropy": 5.305616283416748,
      "epoch": 7.543389368468878,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0001259928589336239,
      "loss": 4.5388,
      "mean_token_accuracy": 0.24974288642406464,
      "num_tokens": 17199528.0,
      "step": 8305
    },
    {
      "entropy": 5.319398403167725,
      "epoch": 7.547932757837347,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00012572822565642378,
      "loss": 4.5077,
      "mean_token_accuracy": 0.243218894302845,
      "num_tokens": 17210030.0,
      "step": 8310
    },
    {
      "entropy": 5.325741195678711,
      "epoch": 7.552476147205816,
      "grad_norm": 1.3125,
      "learning_rate": 0.00012546396069248813,
      "loss": 4.5915,
      "mean_token_accuracy": 0.23653500974178315,
      "num_tokens": 17220231.0,
      "step": 8315
    },
    {
      "entropy": 5.309752321243286,
      "epoch": 7.557019536574284,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00012520006469386458,
      "loss": 4.5399,
      "mean_token_accuracy": 0.24679558873176574,
      "num_tokens": 17229966.0,
      "step": 8320
    },
    {
      "entropy": 5.331660270690918,
      "epoch": 7.561562925942753,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00012493653831169014,
      "loss": 4.5721,
      "mean_token_accuracy": 0.23569018691778182,
      "num_tokens": 17240654.0,
      "step": 8325
    },
    {
      "entropy": 5.247731113433838,
      "epoch": 7.566106315311222,
      "grad_norm": 1.125,
      "learning_rate": 0.00012467338219619015,
      "loss": 4.4713,
      "mean_token_accuracy": 0.256223663687706,
      "num_tokens": 17250653.0,
      "step": 8330
    },
    {
      "entropy": 5.337149953842163,
      "epoch": 7.570649704679691,
      "grad_norm": 1.484375,
      "learning_rate": 0.000124410596996676,
      "loss": 4.5841,
      "mean_token_accuracy": 0.2381567656993866,
      "num_tokens": 17260223.0,
      "step": 8335
    },
    {
      "entropy": 5.306762933731079,
      "epoch": 7.57519309404816,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001241481833615442,
      "loss": 4.5361,
      "mean_token_accuracy": 0.24211796671152114,
      "num_tokens": 17269998.0,
      "step": 8340
    },
    {
      "entropy": 5.298351144790649,
      "epoch": 7.579736483416629,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00012388614193827407,
      "loss": 4.5043,
      "mean_token_accuracy": 0.2503449410200119,
      "num_tokens": 17279824.0,
      "step": 8345
    },
    {
      "entropy": 5.290048503875733,
      "epoch": 7.584279872785098,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000123624473373427,
      "loss": 4.5239,
      "mean_token_accuracy": 0.24153318107128144,
      "num_tokens": 17290242.0,
      "step": 8350
    },
    {
      "entropy": 5.245184898376465,
      "epoch": 7.588823262153566,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00012336317831264411,
      "loss": 4.4945,
      "mean_token_accuracy": 0.2511947125196457,
      "num_tokens": 17299515.0,
      "step": 8355
    },
    {
      "entropy": 5.290190315246582,
      "epoch": 7.593366651522036,
      "grad_norm": 1.328125,
      "learning_rate": 0.00012310225740064498,
      "loss": 4.4698,
      "mean_token_accuracy": 0.24569836854934693,
      "num_tokens": 17308627.0,
      "step": 8360
    },
    {
      "entropy": 5.245627498626709,
      "epoch": 7.597910040890504,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001228417112812259,
      "loss": 4.4574,
      "mean_token_accuracy": 0.24081072956323624,
      "num_tokens": 17319465.0,
      "step": 8365
    },
    {
      "entropy": 5.272551870346069,
      "epoch": 7.602453430258973,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00012258154059725877,
      "loss": 4.5082,
      "mean_token_accuracy": 0.24205060452222824,
      "num_tokens": 17329202.0,
      "step": 8370
    },
    {
      "entropy": 5.25915117263794,
      "epoch": 7.606996819627442,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001223217459906886,
      "loss": 4.4497,
      "mean_token_accuracy": 0.25542411804199217,
      "num_tokens": 17338956.0,
      "step": 8375
    },
    {
      "entropy": 5.343298816680909,
      "epoch": 7.611540208995911,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00012206232810253297,
      "loss": 4.6013,
      "mean_token_accuracy": 0.23706461489200592,
      "num_tokens": 17349380.0,
      "step": 8380
    },
    {
      "entropy": 5.344514513015747,
      "epoch": 7.61608359836438,
      "grad_norm": 1.359375,
      "learning_rate": 0.00012180328757287967,
      "loss": 4.6594,
      "mean_token_accuracy": 0.23120210617780684,
      "num_tokens": 17359770.0,
      "step": 8385
    },
    {
      "entropy": 5.288833665847778,
      "epoch": 7.620626987732849,
      "grad_norm": 1.34375,
      "learning_rate": 0.00012154462504088545,
      "loss": 4.5244,
      "mean_token_accuracy": 0.2430464208126068,
      "num_tokens": 17369209.0,
      "step": 8390
    },
    {
      "entropy": 5.284766054153442,
      "epoch": 7.625170377101318,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001212863411447744,
      "loss": 4.5373,
      "mean_token_accuracy": 0.2404411867260933,
      "num_tokens": 17379221.0,
      "step": 8395
    },
    {
      "entropy": 5.312490797042846,
      "epoch": 7.629713766469786,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00012102843652183628,
      "loss": 4.4704,
      "mean_token_accuracy": 0.24381019473075866,
      "num_tokens": 17389164.0,
      "step": 8400
    },
    {
      "entropy": 5.341345500946045,
      "epoch": 7.634257155838255,
      "grad_norm": 1.328125,
      "learning_rate": 0.00012077091180842523,
      "loss": 4.5654,
      "mean_token_accuracy": 0.2394126310944557,
      "num_tokens": 17399392.0,
      "step": 8405
    },
    {
      "entropy": 5.3454554080963135,
      "epoch": 7.638800545206724,
      "grad_norm": 1.296875,
      "learning_rate": 0.00012051376763995784,
      "loss": 4.5556,
      "mean_token_accuracy": 0.23798759132623673,
      "num_tokens": 17408933.0,
      "step": 8410
    },
    {
      "entropy": 5.319801568984985,
      "epoch": 7.643343934575193,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00012025700465091177,
      "loss": 4.5253,
      "mean_token_accuracy": 0.24044133573770524,
      "num_tokens": 17419391.0,
      "step": 8415
    },
    {
      "entropy": 5.319757461547852,
      "epoch": 7.647887323943662,
      "grad_norm": 1.421875,
      "learning_rate": 0.00012000062347482413,
      "loss": 4.5646,
      "mean_token_accuracy": 0.23679435402154922,
      "num_tokens": 17430562.0,
      "step": 8420
    },
    {
      "entropy": 5.2933997631073,
      "epoch": 7.652430713312131,
      "grad_norm": 1.25,
      "learning_rate": 0.00011974462474429015,
      "loss": 4.5176,
      "mean_token_accuracy": 0.2463404580950737,
      "num_tokens": 17441345.0,
      "step": 8425
    },
    {
      "entropy": 5.295692253112793,
      "epoch": 7.6569741026806,
      "grad_norm": 1.390625,
      "learning_rate": 0.00011948900909096101,
      "loss": 4.5362,
      "mean_token_accuracy": 0.24075821787118912,
      "num_tokens": 17450972.0,
      "step": 8430
    },
    {
      "entropy": 5.284755277633667,
      "epoch": 7.661517492049069,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001192337771455431,
      "loss": 4.5519,
      "mean_token_accuracy": 0.23616982400417327,
      "num_tokens": 17460930.0,
      "step": 8435
    },
    {
      "entropy": 5.345807409286499,
      "epoch": 7.666060881417538,
      "grad_norm": 1.234375,
      "learning_rate": 0.00011897892953779576,
      "loss": 4.5249,
      "mean_token_accuracy": 0.24177036434412003,
      "num_tokens": 17470865.0,
      "step": 8440
    },
    {
      "entropy": 5.280294752120971,
      "epoch": 7.670604270786006,
      "grad_norm": 1.296875,
      "learning_rate": 0.00011872446689653032,
      "loss": 4.4606,
      "mean_token_accuracy": 0.252616286277771,
      "num_tokens": 17481379.0,
      "step": 8445
    },
    {
      "entropy": 5.4127250671386715,
      "epoch": 7.675147660154475,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001184703898496078,
      "loss": 4.5754,
      "mean_token_accuracy": 0.24390935897827148,
      "num_tokens": 17492741.0,
      "step": 8450
    },
    {
      "entropy": 5.325567388534546,
      "epoch": 7.679691049522944,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00011821669902393823,
      "loss": 4.5752,
      "mean_token_accuracy": 0.2342488870024681,
      "num_tokens": 17504179.0,
      "step": 8455
    },
    {
      "entropy": 5.275407981872559,
      "epoch": 7.684234438891413,
      "grad_norm": 1.1875,
      "learning_rate": 0.00011796339504547846,
      "loss": 4.5338,
      "mean_token_accuracy": 0.24723386764526367,
      "num_tokens": 17514299.0,
      "step": 8460
    },
    {
      "entropy": 5.346652889251709,
      "epoch": 7.688777828259882,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00011771047853923086,
      "loss": 4.5269,
      "mean_token_accuracy": 0.2432613864541054,
      "num_tokens": 17524454.0,
      "step": 8465
    },
    {
      "entropy": 5.298489761352539,
      "epoch": 7.693321217628351,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00011745795012924169,
      "loss": 4.516,
      "mean_token_accuracy": 0.24704488962888718,
      "num_tokens": 17534887.0,
      "step": 8470
    },
    {
      "entropy": 5.298565244674682,
      "epoch": 7.69786460699682,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00011720581043859988,
      "loss": 4.5948,
      "mean_token_accuracy": 0.2392280876636505,
      "num_tokens": 17544944.0,
      "step": 8475
    },
    {
      "entropy": 5.311135005950928,
      "epoch": 7.702407996365288,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00011695406008943495,
      "loss": 4.5191,
      "mean_token_accuracy": 0.24975608736276628,
      "num_tokens": 17556481.0,
      "step": 8480
    },
    {
      "entropy": 5.377387428283692,
      "epoch": 7.706951385733757,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00011670269970291584,
      "loss": 4.6034,
      "mean_token_accuracy": 0.23583606779575347,
      "num_tokens": 17566776.0,
      "step": 8485
    },
    {
      "entropy": 5.2726915836334225,
      "epoch": 7.711494775102226,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001164517298992493,
      "loss": 4.4948,
      "mean_token_accuracy": 0.23745836019515992,
      "num_tokens": 17575441.0,
      "step": 8490
    },
    {
      "entropy": 5.326250696182251,
      "epoch": 7.716038164470695,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00011620115129767848,
      "loss": 4.6046,
      "mean_token_accuracy": 0.23760365396738053,
      "num_tokens": 17585792.0,
      "step": 8495
    },
    {
      "entropy": 5.293768072128296,
      "epoch": 7.720581553839164,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00011595096451648111,
      "loss": 4.5323,
      "mean_token_accuracy": 0.234491229057312,
      "num_tokens": 17595820.0,
      "step": 8500
    },
    {
      "epoch": 7.720581553839164,
      "eval_entropy": 5.191864113475001,
      "eval_loss": 5.658626079559326,
      "eval_mean_token_accuracy": 0.17632519177399403,
      "eval_num_tokens": 17595820.0,
      "eval_runtime": 2.0396,
      "eval_samples_per_second": 1684.172,
      "eval_steps_per_second": 210.828,
      "step": 8500
    },
    {
      "entropy": 5.287222623825073,
      "epoch": 7.725124943207633,
      "grad_norm": 1.4375,
      "learning_rate": 0.00011570117017296819,
      "loss": 4.4844,
      "mean_token_accuracy": 0.2444393217563629,
      "num_tokens": 17605791.0,
      "step": 8505
    },
    {
      "entropy": 5.3164466381072994,
      "epoch": 7.729668332576102,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00011545176888348237,
      "loss": 4.5522,
      "mean_token_accuracy": 0.24058474451303483,
      "num_tokens": 17616159.0,
      "step": 8510
    },
    {
      "entropy": 5.2832419872283936,
      "epoch": 7.734211721944571,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00011520276126339667,
      "loss": 4.508,
      "mean_token_accuracy": 0.24505597800016404,
      "num_tokens": 17626612.0,
      "step": 8515
    },
    {
      "entropy": 5.327975082397461,
      "epoch": 7.73875511131304,
      "grad_norm": 1.375,
      "learning_rate": 0.00011495414792711259,
      "loss": 4.5396,
      "mean_token_accuracy": 0.23940024971961976,
      "num_tokens": 17636254.0,
      "step": 8520
    },
    {
      "entropy": 5.393757581710815,
      "epoch": 7.743298500681508,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00011470592948805882,
      "loss": 4.6217,
      "mean_token_accuracy": 0.23228559345006944,
      "num_tokens": 17648097.0,
      "step": 8525
    },
    {
      "entropy": 5.292856502532959,
      "epoch": 7.747841890049977,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00011445810655868957,
      "loss": 4.5106,
      "mean_token_accuracy": 0.24486061781644822,
      "num_tokens": 17658361.0,
      "step": 8530
    },
    {
      "entropy": 5.248042106628418,
      "epoch": 7.752385279418446,
      "grad_norm": 1.3125,
      "learning_rate": 0.00011421067975048355,
      "loss": 4.4708,
      "mean_token_accuracy": 0.24903352558612823,
      "num_tokens": 17668640.0,
      "step": 8535
    },
    {
      "entropy": 5.235863733291626,
      "epoch": 7.756928668786915,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00011396364967394152,
      "loss": 4.4273,
      "mean_token_accuracy": 0.2518807530403137,
      "num_tokens": 17678257.0,
      "step": 8540
    },
    {
      "entropy": 5.233391904830933,
      "epoch": 7.761472058155384,
      "grad_norm": 1.234375,
      "learning_rate": 0.00011371701693858582,
      "loss": 4.5424,
      "mean_token_accuracy": 0.24088944643735885,
      "num_tokens": 17688635.0,
      "step": 8545
    },
    {
      "entropy": 5.2977296829223635,
      "epoch": 7.766015447523853,
      "grad_norm": 1.21875,
      "learning_rate": 0.00011347078215295817,
      "loss": 4.5776,
      "mean_token_accuracy": 0.2344253495335579,
      "num_tokens": 17699917.0,
      "step": 8550
    },
    {
      "entropy": 5.330959796905518,
      "epoch": 7.770558836892322,
      "grad_norm": 1.34375,
      "learning_rate": 0.00011322494592461839,
      "loss": 4.5329,
      "mean_token_accuracy": 0.24521564096212387,
      "num_tokens": 17710278.0,
      "step": 8555
    },
    {
      "entropy": 5.28265872001648,
      "epoch": 7.77510222626079,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00011297950886014292,
      "loss": 4.4986,
      "mean_token_accuracy": 0.24735629558563232,
      "num_tokens": 17720375.0,
      "step": 8560
    },
    {
      "entropy": 5.299135446548462,
      "epoch": 7.779645615629259,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0001127344715651234,
      "loss": 4.5552,
      "mean_token_accuracy": 0.23341630399227142,
      "num_tokens": 17730299.0,
      "step": 8565
    },
    {
      "entropy": 5.334642791748047,
      "epoch": 7.784189004997728,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00011248983464416497,
      "loss": 4.5677,
      "mean_token_accuracy": 0.24177423864603043,
      "num_tokens": 17741096.0,
      "step": 8570
    },
    {
      "entropy": 5.281172227859497,
      "epoch": 7.788732394366197,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00011224559870088488,
      "loss": 4.5263,
      "mean_token_accuracy": 0.2448069080710411,
      "num_tokens": 17751502.0,
      "step": 8575
    },
    {
      "entropy": 5.296409702301025,
      "epoch": 7.793275783734666,
      "grad_norm": 1.375,
      "learning_rate": 0.00011200176433791101,
      "loss": 4.5422,
      "mean_token_accuracy": 0.24706660807132722,
      "num_tokens": 17761955.0,
      "step": 8580
    },
    {
      "entropy": 5.251646852493286,
      "epoch": 7.797819173103135,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001117583321568805,
      "loss": 4.4885,
      "mean_token_accuracy": 0.24607499241828917,
      "num_tokens": 17772330.0,
      "step": 8585
    },
    {
      "entropy": 5.319650888442993,
      "epoch": 7.802362562471604,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00011151530275843805,
      "loss": 4.4654,
      "mean_token_accuracy": 0.24770622402429582,
      "num_tokens": 17781688.0,
      "step": 8590
    },
    {
      "entropy": 5.24897837638855,
      "epoch": 7.806905951840073,
      "grad_norm": 1.328125,
      "learning_rate": 0.00011127267674223449,
      "loss": 4.4977,
      "mean_token_accuracy": 0.2474204793572426,
      "num_tokens": 17792037.0,
      "step": 8595
    },
    {
      "entropy": 5.322576713562012,
      "epoch": 7.811449341208542,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00011103045470692532,
      "loss": 4.5813,
      "mean_token_accuracy": 0.24196469336748122,
      "num_tokens": 17802705.0,
      "step": 8600
    },
    {
      "entropy": 5.371616315841675,
      "epoch": 7.81599273057701,
      "grad_norm": 1.3125,
      "learning_rate": 0.00011078863725016955,
      "loss": 4.6323,
      "mean_token_accuracy": 0.22973187416791915,
      "num_tokens": 17813758.0,
      "step": 8605
    },
    {
      "entropy": 5.258491373062133,
      "epoch": 7.820536119945479,
      "grad_norm": 1.34375,
      "learning_rate": 0.00011054722496862759,
      "loss": 4.4991,
      "mean_token_accuracy": 0.242451074719429,
      "num_tokens": 17823465.0,
      "step": 8610
    },
    {
      "entropy": 5.305764150619507,
      "epoch": 7.825079509313948,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00011030621845796021,
      "loss": 4.5298,
      "mean_token_accuracy": 0.2479332134127617,
      "num_tokens": 17833719.0,
      "step": 8615
    },
    {
      "entropy": 5.286902141571045,
      "epoch": 7.829622898682417,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00011006561831282703,
      "loss": 4.4962,
      "mean_token_accuracy": 0.25299486368894575,
      "num_tokens": 17844519.0,
      "step": 8620
    },
    {
      "entropy": 5.303606843948364,
      "epoch": 7.834166288050886,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00010982542512688515,
      "loss": 4.5012,
      "mean_token_accuracy": 0.24302642792463303,
      "num_tokens": 17856463.0,
      "step": 8625
    },
    {
      "entropy": 5.25388240814209,
      "epoch": 7.838709677419355,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00010958563949278711,
      "loss": 4.4988,
      "mean_token_accuracy": 0.2417108178138733,
      "num_tokens": 17867449.0,
      "step": 8630
    },
    {
      "entropy": 5.234158277511597,
      "epoch": 7.843253066787824,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00010934626200218032,
      "loss": 4.4944,
      "mean_token_accuracy": 0.24299853891134263,
      "num_tokens": 17878522.0,
      "step": 8635
    },
    {
      "entropy": 5.325538110733032,
      "epoch": 7.847796456156293,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00010910729324570488,
      "loss": 4.5707,
      "mean_token_accuracy": 0.24308927059173585,
      "num_tokens": 17888544.0,
      "step": 8640
    },
    {
      "entropy": 5.32314658164978,
      "epoch": 7.852339845524762,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00010886873381299248,
      "loss": 4.6113,
      "mean_token_accuracy": 0.23674060851335527,
      "num_tokens": 17897432.0,
      "step": 8645
    },
    {
      "entropy": 5.318186855316162,
      "epoch": 7.85688323489323,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00010863058429266469,
      "loss": 4.5398,
      "mean_token_accuracy": 0.2446727603673935,
      "num_tokens": 17907868.0,
      "step": 8650
    },
    {
      "entropy": 5.34188814163208,
      "epoch": 7.861426624261699,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00010839284527233192,
      "loss": 4.4924,
      "mean_token_accuracy": 0.24667324721813202,
      "num_tokens": 17917887.0,
      "step": 8655
    },
    {
      "entropy": 5.266041612625122,
      "epoch": 7.865970013630168,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00010815551733859153,
      "loss": 4.4918,
      "mean_token_accuracy": 0.24261136949062348,
      "num_tokens": 17928030.0,
      "step": 8660
    },
    {
      "entropy": 5.311190938949585,
      "epoch": 7.870513402998637,
      "grad_norm": 1.484375,
      "learning_rate": 0.00010791860107702658,
      "loss": 4.6009,
      "mean_token_accuracy": 0.22790729850530625,
      "num_tokens": 17937990.0,
      "step": 8665
    },
    {
      "entropy": 5.28023591041565,
      "epoch": 7.875056792367106,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00010768209707220435,
      "loss": 4.5866,
      "mean_token_accuracy": 0.24163490980863572,
      "num_tokens": 17948262.0,
      "step": 8670
    },
    {
      "entropy": 5.315430450439453,
      "epoch": 7.879600181735575,
      "grad_norm": 1.390625,
      "learning_rate": 0.00010744600590767504,
      "loss": 4.5617,
      "mean_token_accuracy": 0.238010673224926,
      "num_tokens": 17958942.0,
      "step": 8675
    },
    {
      "entropy": 5.280965995788574,
      "epoch": 7.884143571104044,
      "grad_norm": 1.265625,
      "learning_rate": 0.00010721032816597012,
      "loss": 4.545,
      "mean_token_accuracy": 0.23891518861055375,
      "num_tokens": 17969362.0,
      "step": 8680
    },
    {
      "entropy": 5.315994739532471,
      "epoch": 7.888686960472512,
      "grad_norm": 1.25,
      "learning_rate": 0.000106975064428601,
      "loss": 4.514,
      "mean_token_accuracy": 0.24341868311166764,
      "num_tokens": 17979009.0,
      "step": 8685
    },
    {
      "entropy": 5.348607683181763,
      "epoch": 7.893230349840981,
      "grad_norm": 1.3125,
      "learning_rate": 0.00010674021527605742,
      "loss": 4.5774,
      "mean_token_accuracy": 0.24295536130666734,
      "num_tokens": 17989064.0,
      "step": 8690
    },
    {
      "entropy": 5.288234567642212,
      "epoch": 7.89777373920945,
      "grad_norm": 1.25,
      "learning_rate": 0.00010650578128780656,
      "loss": 4.5145,
      "mean_token_accuracy": 0.24598387330770494,
      "num_tokens": 17998574.0,
      "step": 8695
    },
    {
      "entropy": 5.380628442764282,
      "epoch": 7.902317128577919,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001062717630422908,
      "loss": 4.7053,
      "mean_token_accuracy": 0.2331725612282753,
      "num_tokens": 18010735.0,
      "step": 8700
    },
    {
      "entropy": 5.305588102340698,
      "epoch": 7.906860517946388,
      "grad_norm": 1.375,
      "learning_rate": 0.00010603816111692699,
      "loss": 4.5514,
      "mean_token_accuracy": 0.24253649711608888,
      "num_tokens": 18020316.0,
      "step": 8705
    },
    {
      "entropy": 5.317152070999145,
      "epoch": 7.911403907314857,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00010580497608810455,
      "loss": 4.5246,
      "mean_token_accuracy": 0.23880645632743835,
      "num_tokens": 18031058.0,
      "step": 8710
    },
    {
      "entropy": 5.304561948776245,
      "epoch": 7.915947296683326,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00010557220853118456,
      "loss": 4.5191,
      "mean_token_accuracy": 0.241433547437191,
      "num_tokens": 18040603.0,
      "step": 8715
    },
    {
      "entropy": 5.370265531539917,
      "epoch": 7.920490686051795,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00010533985902049755,
      "loss": 4.6083,
      "mean_token_accuracy": 0.23773934245109557,
      "num_tokens": 18051369.0,
      "step": 8720
    },
    {
      "entropy": 5.286510324478149,
      "epoch": 7.925034075420264,
      "grad_norm": 1.296875,
      "learning_rate": 0.00010510792812934307,
      "loss": 4.4759,
      "mean_token_accuracy": 0.24874655455350875,
      "num_tokens": 18062283.0,
      "step": 8725
    },
    {
      "entropy": 5.301109647750854,
      "epoch": 7.929577464788732,
      "grad_norm": 1.34375,
      "learning_rate": 0.00010487641642998746,
      "loss": 4.5611,
      "mean_token_accuracy": 0.24060118198394775,
      "num_tokens": 18072125.0,
      "step": 8730
    },
    {
      "entropy": 5.275309658050537,
      "epoch": 7.934120854157201,
      "grad_norm": 1.21875,
      "learning_rate": 0.00010464532449366281,
      "loss": 4.5991,
      "mean_token_accuracy": 0.2365257829427719,
      "num_tokens": 18083812.0,
      "step": 8735
    },
    {
      "entropy": 5.340100240707398,
      "epoch": 7.93866424352567,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00010441465289056548,
      "loss": 4.5531,
      "mean_token_accuracy": 0.24291856735944747,
      "num_tokens": 18095399.0,
      "step": 8740
    },
    {
      "entropy": 5.363664531707764,
      "epoch": 7.943207632894139,
      "grad_norm": 1.234375,
      "learning_rate": 0.00010418440218985482,
      "loss": 4.5579,
      "mean_token_accuracy": 0.24421486109495164,
      "num_tokens": 18106234.0,
      "step": 8745
    },
    {
      "entropy": 5.297828435897827,
      "epoch": 7.947751022262608,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00010395457295965143,
      "loss": 4.4618,
      "mean_token_accuracy": 0.25437427312135696,
      "num_tokens": 18116129.0,
      "step": 8750
    },
    {
      "entropy": 5.346349430084229,
      "epoch": 7.952294411631077,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00010372516576703617,
      "loss": 4.5872,
      "mean_token_accuracy": 0.2378690630197525,
      "num_tokens": 18126757.0,
      "step": 8755
    },
    {
      "entropy": 5.309607410430909,
      "epoch": 7.956837800999546,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001034961811780484,
      "loss": 4.5573,
      "mean_token_accuracy": 0.23789576143026353,
      "num_tokens": 18136645.0,
      "step": 8760
    },
    {
      "entropy": 5.362412452697754,
      "epoch": 7.961381190368014,
      "grad_norm": 1.4375,
      "learning_rate": 0.00010326761975768492,
      "loss": 4.6127,
      "mean_token_accuracy": 0.2343620777130127,
      "num_tokens": 18146458.0,
      "step": 8765
    },
    {
      "entropy": 5.331036996841431,
      "epoch": 7.965924579736483,
      "grad_norm": 1.40625,
      "learning_rate": 0.00010303948206989827,
      "loss": 4.5534,
      "mean_token_accuracy": 0.24569670259952545,
      "num_tokens": 18156151.0,
      "step": 8770
    },
    {
      "entropy": 5.302338600158691,
      "epoch": 7.970467969104952,
      "grad_norm": 1.296875,
      "learning_rate": 0.00010281176867759556,
      "loss": 4.5461,
      "mean_token_accuracy": 0.2418573647737503,
      "num_tokens": 18166695.0,
      "step": 8775
    },
    {
      "entropy": 5.297445058822632,
      "epoch": 7.975011358473421,
      "grad_norm": 1.34375,
      "learning_rate": 0.00010258448014263682,
      "loss": 4.5098,
      "mean_token_accuracy": 0.2445538729429245,
      "num_tokens": 18177011.0,
      "step": 8780
    },
    {
      "entropy": 5.337861013412476,
      "epoch": 7.97955474784189,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00010235761702583405,
      "loss": 4.6095,
      "mean_token_accuracy": 0.2405742034316063,
      "num_tokens": 18187179.0,
      "step": 8785
    },
    {
      "entropy": 5.301289129257202,
      "epoch": 7.984098137210359,
      "grad_norm": 1.28125,
      "learning_rate": 0.00010213117988694944,
      "loss": 4.5603,
      "mean_token_accuracy": 0.23668374717235566,
      "num_tokens": 18197106.0,
      "step": 8790
    },
    {
      "entropy": 5.2427229404449465,
      "epoch": 7.988641526578828,
      "grad_norm": 1.234375,
      "learning_rate": 0.00010190516928469407,
      "loss": 4.5407,
      "mean_token_accuracy": 0.2446129724383354,
      "num_tokens": 18207729.0,
      "step": 8795
    },
    {
      "entropy": 5.295536756515503,
      "epoch": 7.993184915947297,
      "grad_norm": 1.34375,
      "learning_rate": 0.00010167958577672654,
      "loss": 4.4852,
      "mean_token_accuracy": 0.2434520184993744,
      "num_tokens": 18217885.0,
      "step": 8800
    },
    {
      "entropy": 5.320849943161011,
      "epoch": 7.997728305315766,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00010145442991965198,
      "loss": 4.5766,
      "mean_token_accuracy": 0.23685490041971208,
      "num_tokens": 18229633.0,
      "step": 8805
    },
    {
      "entropy": 5.289232571919759,
      "epoch": 8.001817355747388,
      "grad_norm": 1.296875,
      "learning_rate": 0.00010122970226901981,
      "loss": 4.5556,
      "mean_token_accuracy": 0.2404453009366989,
      "num_tokens": 18239194.0,
      "step": 8810
    },
    {
      "entropy": 5.345217990875244,
      "epoch": 8.006360745115856,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00010100540337932339,
      "loss": 4.514,
      "mean_token_accuracy": 0.24221750795841218,
      "num_tokens": 18250061.0,
      "step": 8815
    },
    {
      "entropy": 5.308269071578979,
      "epoch": 8.010904134484326,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00010078153380399773,
      "loss": 4.4327,
      "mean_token_accuracy": 0.25739925801754,
      "num_tokens": 18260977.0,
      "step": 8820
    },
    {
      "entropy": 5.317370462417602,
      "epoch": 8.015447523852794,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00010055809409541902,
      "loss": 4.4736,
      "mean_token_accuracy": 0.24923319816589357,
      "num_tokens": 18271866.0,
      "step": 8825
    },
    {
      "entropy": 5.292099714279175,
      "epoch": 8.019990913221264,
      "grad_norm": 1.140625,
      "learning_rate": 0.00010033508480490222,
      "loss": 4.5419,
      "mean_token_accuracy": 0.24210608452558519,
      "num_tokens": 18283353.0,
      "step": 8830
    },
    {
      "entropy": 5.247559022903443,
      "epoch": 8.024534302589732,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00010011250648270081,
      "loss": 4.4778,
      "mean_token_accuracy": 0.24919706881046294,
      "num_tokens": 18293619.0,
      "step": 8835
    },
    {
      "entropy": 5.307931613922119,
      "epoch": 8.0290776919582,
      "grad_norm": 1.3203125,
      "learning_rate": 9.98903596780046e-05,
      "loss": 4.487,
      "mean_token_accuracy": 0.2579959526658058,
      "num_tokens": 18303411.0,
      "step": 8840
    },
    {
      "entropy": 5.305163717269897,
      "epoch": 8.03362108132667,
      "grad_norm": 1.328125,
      "learning_rate": 9.966864493893876e-05,
      "loss": 4.4291,
      "mean_token_accuracy": 0.25662620067596437,
      "num_tokens": 18312798.0,
      "step": 8845
    },
    {
      "entropy": 5.295398426055908,
      "epoch": 8.038164470695138,
      "grad_norm": 1.2734375,
      "learning_rate": 9.94473628125623e-05,
      "loss": 4.5011,
      "mean_token_accuracy": 0.24260002076625825,
      "num_tokens": 18324111.0,
      "step": 8850
    },
    {
      "entropy": 5.290332841873169,
      "epoch": 8.042707860063608,
      "grad_norm": 1.3125,
      "learning_rate": 9.922651384486695e-05,
      "loss": 4.4513,
      "mean_token_accuracy": 0.2458519458770752,
      "num_tokens": 18335197.0,
      "step": 8855
    },
    {
      "entropy": 5.325647497177124,
      "epoch": 8.047251249432076,
      "grad_norm": 1.265625,
      "learning_rate": 9.900609858077562e-05,
      "loss": 4.5022,
      "mean_token_accuracy": 0.2470851168036461,
      "num_tokens": 18345896.0,
      "step": 8860
    },
    {
      "entropy": 5.364380693435669,
      "epoch": 8.051794638800546,
      "grad_norm": 1.3359375,
      "learning_rate": 9.878611756414094e-05,
      "loss": 4.5306,
      "mean_token_accuracy": 0.24752559661865234,
      "num_tokens": 18356941.0,
      "step": 8865
    },
    {
      "entropy": 5.281443500518799,
      "epoch": 8.056338028169014,
      "grad_norm": 1.3515625,
      "learning_rate": 9.856657133774424e-05,
      "loss": 4.4426,
      "mean_token_accuracy": 0.25316920280456545,
      "num_tokens": 18366767.0,
      "step": 8870
    },
    {
      "entropy": 5.346744108200073,
      "epoch": 8.060881417537482,
      "grad_norm": 1.296875,
      "learning_rate": 9.834746044329411e-05,
      "loss": 4.5438,
      "mean_token_accuracy": 0.2444286584854126,
      "num_tokens": 18376698.0,
      "step": 8875
    },
    {
      "entropy": 5.297689008712768,
      "epoch": 8.065424806905952,
      "grad_norm": 1.2734375,
      "learning_rate": 9.81287854214248e-05,
      "loss": 4.4408,
      "mean_token_accuracy": 0.24842864871025086,
      "num_tokens": 18386678.0,
      "step": 8880
    },
    {
      "entropy": 5.3144937515258786,
      "epoch": 8.06996819627442,
      "grad_norm": 1.3046875,
      "learning_rate": 9.791054681169522e-05,
      "loss": 4.5015,
      "mean_token_accuracy": 0.2465794563293457,
      "num_tokens": 18397020.0,
      "step": 8885
    },
    {
      "entropy": 5.240437412261963,
      "epoch": 8.07451158564289,
      "grad_norm": 1.25,
      "learning_rate": 9.769274515258739e-05,
      "loss": 4.4643,
      "mean_token_accuracy": 0.24491700977087022,
      "num_tokens": 18407330.0,
      "step": 8890
    },
    {
      "entropy": 5.275600862503052,
      "epoch": 8.079054975011358,
      "grad_norm": 1.25,
      "learning_rate": 9.747538098150539e-05,
      "loss": 4.4707,
      "mean_token_accuracy": 0.24410786777734755,
      "num_tokens": 18417046.0,
      "step": 8895
    },
    {
      "entropy": 5.3151078701019285,
      "epoch": 8.083598364379828,
      "grad_norm": 1.25,
      "learning_rate": 9.725845483477352e-05,
      "loss": 4.5242,
      "mean_token_accuracy": 0.2438024416565895,
      "num_tokens": 18427907.0,
      "step": 8900
    },
    {
      "entropy": 5.285511493682861,
      "epoch": 8.088141753748296,
      "grad_norm": 1.4375,
      "learning_rate": 9.704196724763565e-05,
      "loss": 4.4823,
      "mean_token_accuracy": 0.24388936907052994,
      "num_tokens": 18437972.0,
      "step": 8905
    },
    {
      "entropy": 5.273269748687744,
      "epoch": 8.092685143116766,
      "grad_norm": 1.3203125,
      "learning_rate": 9.682591875425323e-05,
      "loss": 4.4896,
      "mean_token_accuracy": 0.24506725817918779,
      "num_tokens": 18448188.0,
      "step": 8910
    },
    {
      "entropy": 5.286931085586548,
      "epoch": 8.097228532485234,
      "grad_norm": 1.3671875,
      "learning_rate": 9.661030988770463e-05,
      "loss": 4.5134,
      "mean_token_accuracy": 0.24448368102312087,
      "num_tokens": 18458130.0,
      "step": 8915
    },
    {
      "entropy": 5.2800150394439695,
      "epoch": 8.101771921853702,
      "grad_norm": 1.265625,
      "learning_rate": 9.63951411799831e-05,
      "loss": 4.4694,
      "mean_token_accuracy": 0.2486296832561493,
      "num_tokens": 18468257.0,
      "step": 8920
    },
    {
      "entropy": 5.266165161132813,
      "epoch": 8.106315311222172,
      "grad_norm": 1.359375,
      "learning_rate": 9.618041316199617e-05,
      "loss": 4.4994,
      "mean_token_accuracy": 0.23944273144006728,
      "num_tokens": 18478720.0,
      "step": 8925
    },
    {
      "entropy": 5.32502589225769,
      "epoch": 8.11085870059064,
      "grad_norm": 1.4453125,
      "learning_rate": 9.596612636356383e-05,
      "loss": 4.4743,
      "mean_token_accuracy": 0.24662055373191832,
      "num_tokens": 18489543.0,
      "step": 8930
    },
    {
      "entropy": 5.301800966262817,
      "epoch": 8.11540208995911,
      "grad_norm": 1.3359375,
      "learning_rate": 9.575228131341745e-05,
      "loss": 4.5041,
      "mean_token_accuracy": 0.2465366691350937,
      "num_tokens": 18499505.0,
      "step": 8935
    },
    {
      "entropy": 5.340041732788086,
      "epoch": 8.119945479327578,
      "grad_norm": 1.3515625,
      "learning_rate": 9.55388785391984e-05,
      "loss": 4.5787,
      "mean_token_accuracy": 0.2395768642425537,
      "num_tokens": 18510044.0,
      "step": 8940
    },
    {
      "entropy": 5.303210067749023,
      "epoch": 8.124488868696048,
      "grad_norm": 1.4140625,
      "learning_rate": 9.532591856745689e-05,
      "loss": 4.5008,
      "mean_token_accuracy": 0.24818192273378373,
      "num_tokens": 18520537.0,
      "step": 8945
    },
    {
      "entropy": 5.32057375907898,
      "epoch": 8.129032258064516,
      "grad_norm": 1.2890625,
      "learning_rate": 9.511340192365044e-05,
      "loss": 4.549,
      "mean_token_accuracy": 0.23841260522603988,
      "num_tokens": 18531840.0,
      "step": 8950
    },
    {
      "entropy": 5.2923401355743405,
      "epoch": 8.133575647432984,
      "grad_norm": 1.40625,
      "learning_rate": 9.490132913214277e-05,
      "loss": 4.4484,
      "mean_token_accuracy": 0.2504141002893448,
      "num_tokens": 18542097.0,
      "step": 8955
    },
    {
      "entropy": 5.313136291503906,
      "epoch": 8.138119036801454,
      "grad_norm": 1.328125,
      "learning_rate": 9.468970071620231e-05,
      "loss": 4.5176,
      "mean_token_accuracy": 0.24735378473997116,
      "num_tokens": 18551629.0,
      "step": 8960
    },
    {
      "entropy": 5.284275341033935,
      "epoch": 8.142662426169922,
      "grad_norm": 1.296875,
      "learning_rate": 9.447851719800126e-05,
      "loss": 4.46,
      "mean_token_accuracy": 0.2426091566681862,
      "num_tokens": 18562248.0,
      "step": 8965
    },
    {
      "entropy": 5.284086084365844,
      "epoch": 8.147205815538392,
      "grad_norm": 1.3984375,
      "learning_rate": 9.426777909861395e-05,
      "loss": 4.4446,
      "mean_token_accuracy": 0.2495478793978691,
      "num_tokens": 18572107.0,
      "step": 8970
    },
    {
      "entropy": 5.29611759185791,
      "epoch": 8.15174920490686,
      "grad_norm": 1.265625,
      "learning_rate": 9.405748693801562e-05,
      "loss": 4.5173,
      "mean_token_accuracy": 0.24357538521289826,
      "num_tokens": 18582665.0,
      "step": 8975
    },
    {
      "entropy": 5.36781530380249,
      "epoch": 8.15629259427533,
      "grad_norm": 1.2890625,
      "learning_rate": 9.384764123508124e-05,
      "loss": 4.5449,
      "mean_token_accuracy": 0.24158413857221603,
      "num_tokens": 18593737.0,
      "step": 8980
    },
    {
      "entropy": 5.321993827819824,
      "epoch": 8.160835983643798,
      "grad_norm": 1.2890625,
      "learning_rate": 9.363824250758443e-05,
      "loss": 4.4886,
      "mean_token_accuracy": 0.2530485615134239,
      "num_tokens": 18604180.0,
      "step": 8985
    },
    {
      "entropy": 5.26632080078125,
      "epoch": 8.165379373012268,
      "grad_norm": 1.3984375,
      "learning_rate": 9.34292912721955e-05,
      "loss": 4.4204,
      "mean_token_accuracy": 0.250825272500515,
      "num_tokens": 18614328.0,
      "step": 8990
    },
    {
      "entropy": 5.306871747970581,
      "epoch": 8.169922762380736,
      "grad_norm": 1.2109375,
      "learning_rate": 9.322078804448095e-05,
      "loss": 4.4693,
      "mean_token_accuracy": 0.26003596633672715,
      "num_tokens": 18625357.0,
      "step": 8995
    },
    {
      "entropy": 5.235411024093628,
      "epoch": 8.174466151749204,
      "grad_norm": 1.3125,
      "learning_rate": 9.301273333890171e-05,
      "loss": 4.4344,
      "mean_token_accuracy": 0.2504829213023186,
      "num_tokens": 18636306.0,
      "step": 9000
    },
    {
      "epoch": 8.174466151749204,
      "eval_entropy": 5.1793446174887725,
      "eval_loss": 5.663380146026611,
      "eval_mean_token_accuracy": 0.17574178276713504,
      "eval_num_tokens": 18636306.0,
      "eval_runtime": 2.0275,
      "eval_samples_per_second": 1694.203,
      "eval_steps_per_second": 212.084,
      "step": 9000
    },
    {
      "entropy": 5.331611967086792,
      "epoch": 8.179009541117674,
      "grad_norm": 1.296875,
      "learning_rate": 9.280512766881227e-05,
      "loss": 4.5119,
      "mean_token_accuracy": 0.24494637548923492,
      "num_tokens": 18647134.0,
      "step": 9005
    },
    {
      "entropy": 5.274996376037597,
      "epoch": 8.183552930486142,
      "grad_norm": 1.421875,
      "learning_rate": 9.25979715464588e-05,
      "loss": 4.4276,
      "mean_token_accuracy": 0.2577662214636803,
      "num_tokens": 18656948.0,
      "step": 9010
    },
    {
      "entropy": 5.317999839782715,
      "epoch": 8.188096319854612,
      "grad_norm": 1.2421875,
      "learning_rate": 9.23912654829784e-05,
      "loss": 4.4771,
      "mean_token_accuracy": 0.24466777890920638,
      "num_tokens": 18667682.0,
      "step": 9015
    },
    {
      "entropy": 5.318257093429565,
      "epoch": 8.19263970922308,
      "grad_norm": 1.265625,
      "learning_rate": 9.218500998839794e-05,
      "loss": 4.5293,
      "mean_token_accuracy": 0.24993367046117781,
      "num_tokens": 18678124.0,
      "step": 9020
    },
    {
      "entropy": 5.293010139465332,
      "epoch": 8.19718309859155,
      "grad_norm": 1.375,
      "learning_rate": 9.197920557163215e-05,
      "loss": 4.483,
      "mean_token_accuracy": 0.2445767939090729,
      "num_tokens": 18689063.0,
      "step": 9025
    },
    {
      "entropy": 5.339461374282837,
      "epoch": 8.201726487960018,
      "grad_norm": 1.453125,
      "learning_rate": 9.177385274048312e-05,
      "loss": 4.5514,
      "mean_token_accuracy": 0.2438694179058075,
      "num_tokens": 18700118.0,
      "step": 9030
    },
    {
      "entropy": 5.311863231658935,
      "epoch": 8.206269877328488,
      "grad_norm": 1.234375,
      "learning_rate": 9.156895200163837e-05,
      "loss": 4.5208,
      "mean_token_accuracy": 0.24340611696243286,
      "num_tokens": 18711683.0,
      "step": 9035
    },
    {
      "entropy": 5.319491195678711,
      "epoch": 8.210813266696956,
      "grad_norm": 1.3359375,
      "learning_rate": 9.13645038606703e-05,
      "loss": 4.4945,
      "mean_token_accuracy": 0.24559812247753143,
      "num_tokens": 18722033.0,
      "step": 9040
    },
    {
      "entropy": 5.350789642333984,
      "epoch": 8.215356656065424,
      "grad_norm": 1.34375,
      "learning_rate": 9.116050882203427e-05,
      "loss": 4.5143,
      "mean_token_accuracy": 0.2404477998614311,
      "num_tokens": 18732387.0,
      "step": 9045
    },
    {
      "entropy": 5.332340383529663,
      "epoch": 8.219900045433894,
      "grad_norm": 1.265625,
      "learning_rate": 9.09569673890678e-05,
      "loss": 4.5214,
      "mean_token_accuracy": 0.24784988760948182,
      "num_tokens": 18743573.0,
      "step": 9050
    },
    {
      "entropy": 5.257116413116455,
      "epoch": 8.224443434802362,
      "grad_norm": 1.234375,
      "learning_rate": 9.075388006398907e-05,
      "loss": 4.4027,
      "mean_token_accuracy": 0.24500966668128968,
      "num_tokens": 18753897.0,
      "step": 9055
    },
    {
      "entropy": 5.2994238376617435,
      "epoch": 8.228986824170832,
      "grad_norm": 1.296875,
      "learning_rate": 9.0551247347896e-05,
      "loss": 4.5354,
      "mean_token_accuracy": 0.23927509635686875,
      "num_tokens": 18764849.0,
      "step": 9060
    },
    {
      "entropy": 5.251907777786255,
      "epoch": 8.2335302135393,
      "grad_norm": 1.2421875,
      "learning_rate": 9.034906974076449e-05,
      "loss": 4.4179,
      "mean_token_accuracy": 0.2500752732157707,
      "num_tokens": 18776404.0,
      "step": 9065
    },
    {
      "entropy": 5.319542407989502,
      "epoch": 8.23807360290777,
      "grad_norm": 1.3046875,
      "learning_rate": 9.014734774144784e-05,
      "loss": 4.5169,
      "mean_token_accuracy": 0.23979179561138153,
      "num_tokens": 18786803.0,
      "step": 9070
    },
    {
      "entropy": 5.32085952758789,
      "epoch": 8.242616992276238,
      "grad_norm": 1.2578125,
      "learning_rate": 8.994608184767491e-05,
      "loss": 4.532,
      "mean_token_accuracy": 0.24531794637441634,
      "num_tokens": 18797135.0,
      "step": 9075
    },
    {
      "entropy": 5.3029052257537845,
      "epoch": 8.247160381644706,
      "grad_norm": 1.4453125,
      "learning_rate": 8.97452725560495e-05,
      "loss": 4.4306,
      "mean_token_accuracy": 0.2566088870167732,
      "num_tokens": 18806495.0,
      "step": 9080
    },
    {
      "entropy": 5.309773683547974,
      "epoch": 8.251703771013176,
      "grad_norm": 1.359375,
      "learning_rate": 8.954492036204829e-05,
      "loss": 4.5022,
      "mean_token_accuracy": 0.24467596113681794,
      "num_tokens": 18816925.0,
      "step": 9085
    },
    {
      "entropy": 5.3029827117919925,
      "epoch": 8.256247160381644,
      "grad_norm": 1.3515625,
      "learning_rate": 8.934502576002059e-05,
      "loss": 4.4414,
      "mean_token_accuracy": 0.25423109233379365,
      "num_tokens": 18826604.0,
      "step": 9090
    },
    {
      "entropy": 5.334570503234863,
      "epoch": 8.260790549750114,
      "grad_norm": 1.234375,
      "learning_rate": 8.914558924318643e-05,
      "loss": 4.5225,
      "mean_token_accuracy": 0.2470988467335701,
      "num_tokens": 18837278.0,
      "step": 9095
    },
    {
      "entropy": 5.319367361068726,
      "epoch": 8.265333939118582,
      "grad_norm": 1.3828125,
      "learning_rate": 8.894661130363557e-05,
      "loss": 4.5464,
      "mean_token_accuracy": 0.24078957587480546,
      "num_tokens": 18846588.0,
      "step": 9100
    },
    {
      "entropy": 5.2655633926391605,
      "epoch": 8.269877328487052,
      "grad_norm": 1.3828125,
      "learning_rate": 8.874809243232624e-05,
      "loss": 4.4617,
      "mean_token_accuracy": 0.25947415083646774,
      "num_tokens": 18856345.0,
      "step": 9105
    },
    {
      "entropy": 5.307589864730835,
      "epoch": 8.27442071785552,
      "grad_norm": 1.3046875,
      "learning_rate": 8.85500331190841e-05,
      "loss": 4.5496,
      "mean_token_accuracy": 0.2374264895915985,
      "num_tokens": 18867000.0,
      "step": 9110
    },
    {
      "entropy": 5.266747236251831,
      "epoch": 8.27896410722399,
      "grad_norm": 1.4453125,
      "learning_rate": 8.835243385260083e-05,
      "loss": 4.5395,
      "mean_token_accuracy": 0.24101696461439132,
      "num_tokens": 18876925.0,
      "step": 9115
    },
    {
      "entropy": 5.308072805404663,
      "epoch": 8.283507496592458,
      "grad_norm": 1.5234375,
      "learning_rate": 8.815529512043293e-05,
      "loss": 4.4767,
      "mean_token_accuracy": 0.25313203781843185,
      "num_tokens": 18886376.0,
      "step": 9120
    },
    {
      "entropy": 5.291871023178101,
      "epoch": 8.288050885960926,
      "grad_norm": 1.2421875,
      "learning_rate": 8.795861740900061e-05,
      "loss": 4.4827,
      "mean_token_accuracy": 0.24776212126016617,
      "num_tokens": 18897258.0,
      "step": 9125
    },
    {
      "entropy": 5.3446691036224365,
      "epoch": 8.292594275329396,
      "grad_norm": 1.3125,
      "learning_rate": 8.77624012035866e-05,
      "loss": 4.465,
      "mean_token_accuracy": 0.25554633140563965,
      "num_tokens": 18906959.0,
      "step": 9130
    },
    {
      "entropy": 5.307412576675415,
      "epoch": 8.297137664697864,
      "grad_norm": 1.3125,
      "learning_rate": 8.756664698833492e-05,
      "loss": 4.5626,
      "mean_token_accuracy": 0.23735951334238053,
      "num_tokens": 18917930.0,
      "step": 9135
    },
    {
      "entropy": 5.360101270675659,
      "epoch": 8.301681054066334,
      "grad_norm": 1.203125,
      "learning_rate": 8.737135524624959e-05,
      "loss": 4.5825,
      "mean_token_accuracy": 0.23769719153642654,
      "num_tokens": 18928506.0,
      "step": 9140
    },
    {
      "entropy": 5.322899532318115,
      "epoch": 8.306224443434802,
      "grad_norm": 1.2265625,
      "learning_rate": 8.717652645919352e-05,
      "loss": 4.4725,
      "mean_token_accuracy": 0.2556913703680038,
      "num_tokens": 18938811.0,
      "step": 9145
    },
    {
      "entropy": 5.260947751998901,
      "epoch": 8.310767832803272,
      "grad_norm": 1.375,
      "learning_rate": 8.698216110788753e-05,
      "loss": 4.4622,
      "mean_token_accuracy": 0.2455285906791687,
      "num_tokens": 18948487.0,
      "step": 9150
    },
    {
      "entropy": 5.303894186019898,
      "epoch": 8.31531122217174,
      "grad_norm": 1.3359375,
      "learning_rate": 8.67882596719086e-05,
      "loss": 4.4831,
      "mean_token_accuracy": 0.2489192247390747,
      "num_tokens": 18958021.0,
      "step": 9155
    },
    {
      "entropy": 5.299885559082031,
      "epoch": 8.319854611540208,
      "grad_norm": 1.4140625,
      "learning_rate": 8.659482262968943e-05,
      "loss": 4.4841,
      "mean_token_accuracy": 0.24811865538358688,
      "num_tokens": 18967937.0,
      "step": 9160
    },
    {
      "entropy": 5.338036346435547,
      "epoch": 8.324398000908678,
      "grad_norm": 1.171875,
      "learning_rate": 8.640185045851653e-05,
      "loss": 4.501,
      "mean_token_accuracy": 0.24693352431058885,
      "num_tokens": 18978119.0,
      "step": 9165
    },
    {
      "entropy": 5.267579030990601,
      "epoch": 8.328941390277146,
      "grad_norm": 1.15625,
      "learning_rate": 8.620934363452973e-05,
      "loss": 4.4548,
      "mean_token_accuracy": 0.2458501785993576,
      "num_tokens": 18989036.0,
      "step": 9170
    },
    {
      "entropy": 5.328738737106323,
      "epoch": 8.333484779645616,
      "grad_norm": 1.3125,
      "learning_rate": 8.601730263272034e-05,
      "loss": 4.5453,
      "mean_token_accuracy": 0.24996053278446198,
      "num_tokens": 18999561.0,
      "step": 9175
    },
    {
      "entropy": 5.275832891464233,
      "epoch": 8.338028169014084,
      "grad_norm": 1.3359375,
      "learning_rate": 8.58257279269305e-05,
      "loss": 4.4459,
      "mean_token_accuracy": 0.24814351052045822,
      "num_tokens": 19010086.0,
      "step": 9180
    },
    {
      "entropy": 5.238356351852417,
      "epoch": 8.342571558382554,
      "grad_norm": 1.421875,
      "learning_rate": 8.563461998985172e-05,
      "loss": 4.3927,
      "mean_token_accuracy": 0.25193280577659605,
      "num_tokens": 19019697.0,
      "step": 9185
    },
    {
      "entropy": 5.331272983551026,
      "epoch": 8.347114947751022,
      "grad_norm": 1.46875,
      "learning_rate": 8.544397929302389e-05,
      "loss": 4.5285,
      "mean_token_accuracy": 0.24985535442829132,
      "num_tokens": 19030429.0,
      "step": 9190
    },
    {
      "entropy": 5.3380388736724855,
      "epoch": 8.351658337119492,
      "grad_norm": 1.2734375,
      "learning_rate": 8.525380630683389e-05,
      "loss": 4.5168,
      "mean_token_accuracy": 0.24382077753543854,
      "num_tokens": 19041052.0,
      "step": 9195
    },
    {
      "entropy": 5.247474479675293,
      "epoch": 8.35620172648796,
      "grad_norm": 1.4375,
      "learning_rate": 8.506410150051478e-05,
      "loss": 4.4539,
      "mean_token_accuracy": 0.25360618382692335,
      "num_tokens": 19050936.0,
      "step": 9200
    },
    {
      "entropy": 5.266754627227783,
      "epoch": 8.360745115856428,
      "grad_norm": 1.34375,
      "learning_rate": 8.487486534214423e-05,
      "loss": 4.4285,
      "mean_token_accuracy": 0.24814922660589217,
      "num_tokens": 19060884.0,
      "step": 9205
    },
    {
      "entropy": 5.287121534347534,
      "epoch": 8.365288505224898,
      "grad_norm": 1.375,
      "learning_rate": 8.468609829864366e-05,
      "loss": 4.5002,
      "mean_token_accuracy": 0.24660014361143112,
      "num_tokens": 19071405.0,
      "step": 9210
    },
    {
      "entropy": 5.283935022354126,
      "epoch": 8.369831894593366,
      "grad_norm": 1.359375,
      "learning_rate": 8.449780083577697e-05,
      "loss": 4.4689,
      "mean_token_accuracy": 0.24394806027412413,
      "num_tokens": 19081186.0,
      "step": 9215
    },
    {
      "entropy": 5.363908433914185,
      "epoch": 8.374375283961836,
      "grad_norm": 1.234375,
      "learning_rate": 8.430997341814947e-05,
      "loss": 4.5153,
      "mean_token_accuracy": 0.24663342982530595,
      "num_tokens": 19091042.0,
      "step": 9220
    },
    {
      "entropy": 5.295595455169678,
      "epoch": 8.378918673330304,
      "grad_norm": 1.4453125,
      "learning_rate": 8.41226165092066e-05,
      "loss": 4.4755,
      "mean_token_accuracy": 0.24762949645519255,
      "num_tokens": 19101408.0,
      "step": 9225
    },
    {
      "entropy": 5.281922245025635,
      "epoch": 8.383462062698774,
      "grad_norm": 1.2578125,
      "learning_rate": 8.393573057123299e-05,
      "loss": 4.4718,
      "mean_token_accuracy": 0.250074303150177,
      "num_tokens": 19111773.0,
      "step": 9230
    },
    {
      "entropy": 5.358229255676269,
      "epoch": 8.388005452067242,
      "grad_norm": 1.4609375,
      "learning_rate": 8.374931606535099e-05,
      "loss": 4.5247,
      "mean_token_accuracy": 0.24361408948898317,
      "num_tokens": 19121532.0,
      "step": 9235
    },
    {
      "entropy": 5.3060479164123535,
      "epoch": 8.392548841435712,
      "grad_norm": 1.3671875,
      "learning_rate": 8.356337345152002e-05,
      "loss": 4.524,
      "mean_token_accuracy": 0.2394486501812935,
      "num_tokens": 19131389.0,
      "step": 9240
    },
    {
      "entropy": 5.2396049976348875,
      "epoch": 8.39709223080418,
      "grad_norm": 1.5,
      "learning_rate": 8.33779031885348e-05,
      "loss": 4.4687,
      "mean_token_accuracy": 0.24704018980264664,
      "num_tokens": 19141304.0,
      "step": 9245
    },
    {
      "entropy": 5.250641298294068,
      "epoch": 8.401635620172648,
      "grad_norm": 1.515625,
      "learning_rate": 8.319290573402494e-05,
      "loss": 4.4605,
      "mean_token_accuracy": 0.24341040700674058,
      "num_tokens": 19151039.0,
      "step": 9250
    },
    {
      "entropy": 5.25805697441101,
      "epoch": 8.406179009541118,
      "grad_norm": 1.3515625,
      "learning_rate": 8.300838154445314e-05,
      "loss": 4.4451,
      "mean_token_accuracy": 0.2531593769788742,
      "num_tokens": 19160909.0,
      "step": 9255
    },
    {
      "entropy": 5.296714353561401,
      "epoch": 8.410722398909586,
      "grad_norm": 1.2578125,
      "learning_rate": 8.282433107511466e-05,
      "loss": 4.4782,
      "mean_token_accuracy": 0.2417339339852333,
      "num_tokens": 19171906.0,
      "step": 9260
    },
    {
      "entropy": 5.312888431549072,
      "epoch": 8.415265788278056,
      "grad_norm": 1.3515625,
      "learning_rate": 8.264075478013552e-05,
      "loss": 4.5222,
      "mean_token_accuracy": 0.2472866430878639,
      "num_tokens": 19182688.0,
      "step": 9265
    },
    {
      "entropy": 5.275963878631591,
      "epoch": 8.419809177646524,
      "grad_norm": 1.3125,
      "learning_rate": 8.245765311247216e-05,
      "loss": 4.4369,
      "mean_token_accuracy": 0.25104597508907317,
      "num_tokens": 19192429.0,
      "step": 9270
    },
    {
      "entropy": 5.334664964675904,
      "epoch": 8.424352567014994,
      "grad_norm": 1.3359375,
      "learning_rate": 8.227502652390968e-05,
      "loss": 4.5785,
      "mean_token_accuracy": 0.23479071110486985,
      "num_tokens": 19202454.0,
      "step": 9275
    },
    {
      "entropy": 5.370562744140625,
      "epoch": 8.428895956383462,
      "grad_norm": 1.3515625,
      "learning_rate": 8.2092875465061e-05,
      "loss": 4.5477,
      "mean_token_accuracy": 0.24561670273542405,
      "num_tokens": 19213057.0,
      "step": 9280
    },
    {
      "entropy": 5.251692295074463,
      "epoch": 8.43343934575193,
      "grad_norm": 1.359375,
      "learning_rate": 8.191120038536575e-05,
      "loss": 4.4018,
      "mean_token_accuracy": 0.251190721988678,
      "num_tokens": 19222417.0,
      "step": 9285
    },
    {
      "entropy": 5.32673225402832,
      "epoch": 8.4379827351204,
      "grad_norm": 1.2421875,
      "learning_rate": 8.173000173308921e-05,
      "loss": 4.5275,
      "mean_token_accuracy": 0.23937877118587494,
      "num_tokens": 19233743.0,
      "step": 9290
    },
    {
      "entropy": 5.3644757747650145,
      "epoch": 8.442526124488868,
      "grad_norm": 1.4375,
      "learning_rate": 8.154927995532098e-05,
      "loss": 4.6415,
      "mean_token_accuracy": 0.23598811477422715,
      "num_tokens": 19245000.0,
      "step": 9295
    },
    {
      "entropy": 5.356596422195435,
      "epoch": 8.447069513857338,
      "grad_norm": 1.375,
      "learning_rate": 8.136903549797408e-05,
      "loss": 4.506,
      "mean_token_accuracy": 0.2428935408592224,
      "num_tokens": 19255738.0,
      "step": 9300
    },
    {
      "entropy": 5.329964971542358,
      "epoch": 8.451612903225806,
      "grad_norm": 1.2734375,
      "learning_rate": 8.118926880578378e-05,
      "loss": 4.5367,
      "mean_token_accuracy": 0.24558568894863128,
      "num_tokens": 19266268.0,
      "step": 9305
    },
    {
      "entropy": 5.434853792190552,
      "epoch": 8.456156292594276,
      "grad_norm": 1.2578125,
      "learning_rate": 8.100998032230656e-05,
      "loss": 4.5829,
      "mean_token_accuracy": 0.24776215702295304,
      "num_tokens": 19276721.0,
      "step": 9310
    },
    {
      "entropy": 5.298271179199219,
      "epoch": 8.460699681962744,
      "grad_norm": 1.3125,
      "learning_rate": 8.083117048991895e-05,
      "loss": 4.45,
      "mean_token_accuracy": 0.24628928154706956,
      "num_tokens": 19287241.0,
      "step": 9315
    },
    {
      "entropy": 5.327994632720947,
      "epoch": 8.465243071331214,
      "grad_norm": 1.21875,
      "learning_rate": 8.065283974981638e-05,
      "loss": 4.5668,
      "mean_token_accuracy": 0.2341143324971199,
      "num_tokens": 19298065.0,
      "step": 9320
    },
    {
      "entropy": 5.369241666793823,
      "epoch": 8.469786460699682,
      "grad_norm": 1.34375,
      "learning_rate": 8.047498854201222e-05,
      "loss": 4.544,
      "mean_token_accuracy": 0.24229334890842438,
      "num_tokens": 19308108.0,
      "step": 9325
    },
    {
      "entropy": 5.317949867248535,
      "epoch": 8.47432985006815,
      "grad_norm": 1.4296875,
      "learning_rate": 8.029761730533678e-05,
      "loss": 4.524,
      "mean_token_accuracy": 0.2434487983584404,
      "num_tokens": 19317909.0,
      "step": 9330
    },
    {
      "entropy": 5.344450616836548,
      "epoch": 8.47887323943662,
      "grad_norm": 1.390625,
      "learning_rate": 8.012072647743569e-05,
      "loss": 4.573,
      "mean_token_accuracy": 0.24522460550069808,
      "num_tokens": 19328278.0,
      "step": 9335
    },
    {
      "entropy": 5.227662801742554,
      "epoch": 8.483416628805088,
      "grad_norm": 1.3203125,
      "learning_rate": 7.994431649476974e-05,
      "loss": 4.4721,
      "mean_token_accuracy": 0.25412004441022873,
      "num_tokens": 19338995.0,
      "step": 9340
    },
    {
      "entropy": 5.294491386413574,
      "epoch": 8.487960018173558,
      "grad_norm": 1.28125,
      "learning_rate": 7.976838779261285e-05,
      "loss": 4.5215,
      "mean_token_accuracy": 0.24315209686756134,
      "num_tokens": 19349177.0,
      "step": 9345
    },
    {
      "entropy": 5.419864130020142,
      "epoch": 8.492503407542026,
      "grad_norm": 1.375,
      "learning_rate": 7.959294080505179e-05,
      "loss": 4.5401,
      "mean_token_accuracy": 0.24278531670570375,
      "num_tokens": 19358871.0,
      "step": 9350
    },
    {
      "entropy": 5.343920707702637,
      "epoch": 8.497046796910496,
      "grad_norm": 1.34375,
      "learning_rate": 7.941797596498435e-05,
      "loss": 4.5609,
      "mean_token_accuracy": 0.24715256094932556,
      "num_tokens": 19368594.0,
      "step": 9355
    },
    {
      "entropy": 5.280983734130859,
      "epoch": 8.501590186278964,
      "grad_norm": 1.1875,
      "learning_rate": 7.924349370411904e-05,
      "loss": 4.4601,
      "mean_token_accuracy": 0.24279809445142747,
      "num_tokens": 19379973.0,
      "step": 9360
    },
    {
      "entropy": 5.299367666244507,
      "epoch": 8.506133575647432,
      "grad_norm": 1.421875,
      "learning_rate": 7.906949445297342e-05,
      "loss": 4.4726,
      "mean_token_accuracy": 0.25452430099248885,
      "num_tokens": 19390762.0,
      "step": 9365
    },
    {
      "entropy": 5.304686164855957,
      "epoch": 8.510676965015902,
      "grad_norm": 1.2890625,
      "learning_rate": 7.889597864087336e-05,
      "loss": 4.5305,
      "mean_token_accuracy": 0.24932739287614822,
      "num_tokens": 19401918.0,
      "step": 9370
    },
    {
      "entropy": 5.250623607635498,
      "epoch": 8.51522035438437,
      "grad_norm": 1.1640625,
      "learning_rate": 7.872294669595185e-05,
      "loss": 4.4678,
      "mean_token_accuracy": 0.2543991908431053,
      "num_tokens": 19411973.0,
      "step": 9375
    },
    {
      "entropy": 5.2999513149261475,
      "epoch": 8.51976374375284,
      "grad_norm": 1.265625,
      "learning_rate": 7.855039904514808e-05,
      "loss": 4.4533,
      "mean_token_accuracy": 0.24672039300203324,
      "num_tokens": 19422888.0,
      "step": 9380
    },
    {
      "entropy": 5.359530925750732,
      "epoch": 8.524307133121308,
      "grad_norm": 1.2734375,
      "learning_rate": 7.837833611420618e-05,
      "loss": 4.5442,
      "mean_token_accuracy": 0.24620097279548644,
      "num_tokens": 19433611.0,
      "step": 9385
    },
    {
      "entropy": 5.231460475921631,
      "epoch": 8.528850522489778,
      "grad_norm": 1.390625,
      "learning_rate": 7.820675832767435e-05,
      "loss": 4.4746,
      "mean_token_accuracy": 0.24608877599239348,
      "num_tokens": 19443712.0,
      "step": 9390
    },
    {
      "entropy": 5.262017488479614,
      "epoch": 8.533393911858246,
      "grad_norm": 1.4453125,
      "learning_rate": 7.803566610890365e-05,
      "loss": 4.3975,
      "mean_token_accuracy": 0.2519361674785614,
      "num_tokens": 19453410.0,
      "step": 9395
    },
    {
      "entropy": 5.272234392166138,
      "epoch": 8.537937301226716,
      "grad_norm": 1.2265625,
      "learning_rate": 7.786505988004722e-05,
      "loss": 4.4392,
      "mean_token_accuracy": 0.2544954165816307,
      "num_tokens": 19463306.0,
      "step": 9400
    },
    {
      "entropy": 5.342586517333984,
      "epoch": 8.542480690595184,
      "grad_norm": 1.28125,
      "learning_rate": 7.769494006205894e-05,
      "loss": 4.5495,
      "mean_token_accuracy": 0.24556162506341933,
      "num_tokens": 19474127.0,
      "step": 9405
    },
    {
      "entropy": 5.3682938575744625,
      "epoch": 8.547024079963652,
      "grad_norm": 1.0859375,
      "learning_rate": 7.752530707469252e-05,
      "loss": 4.5732,
      "mean_token_accuracy": 0.24459436535835266,
      "num_tokens": 19486472.0,
      "step": 9410
    },
    {
      "entropy": 5.30327730178833,
      "epoch": 8.551567469332122,
      "grad_norm": 1.3203125,
      "learning_rate": 7.73561613365005e-05,
      "loss": 4.517,
      "mean_token_accuracy": 0.2468592941761017,
      "num_tokens": 19497130.0,
      "step": 9415
    },
    {
      "entropy": 5.27040843963623,
      "epoch": 8.55611085870059,
      "grad_norm": 1.453125,
      "learning_rate": 7.718750326483321e-05,
      "loss": 4.4273,
      "mean_token_accuracy": 0.2541737169027328,
      "num_tokens": 19507321.0,
      "step": 9420
    },
    {
      "entropy": 5.352059602737427,
      "epoch": 8.56065424806906,
      "grad_norm": 1.5703125,
      "learning_rate": 7.701933327583766e-05,
      "loss": 4.5609,
      "mean_token_accuracy": 0.24057227969169617,
      "num_tokens": 19516244.0,
      "step": 9425
    },
    {
      "entropy": 5.266730690002442,
      "epoch": 8.565197637437528,
      "grad_norm": 1.2578125,
      "learning_rate": 7.685165178445658e-05,
      "loss": 4.4551,
      "mean_token_accuracy": 0.25301125943660735,
      "num_tokens": 19526594.0,
      "step": 9430
    },
    {
      "entropy": 5.2699432373046875,
      "epoch": 8.569741026805998,
      "grad_norm": 1.3359375,
      "learning_rate": 7.668445920442732e-05,
      "loss": 4.467,
      "mean_token_accuracy": 0.23839007765054704,
      "num_tokens": 19536835.0,
      "step": 9435
    },
    {
      "entropy": 5.268694829940796,
      "epoch": 8.574284416174466,
      "grad_norm": 1.375,
      "learning_rate": 7.651775594828112e-05,
      "loss": 4.5664,
      "mean_token_accuracy": 0.23348012417554856,
      "num_tokens": 19546474.0,
      "step": 9440
    },
    {
      "entropy": 5.3747905731201175,
      "epoch": 8.578827805542936,
      "grad_norm": 1.3125,
      "learning_rate": 7.635154242734146e-05,
      "loss": 4.5507,
      "mean_token_accuracy": 0.242520272731781,
      "num_tokens": 19556604.0,
      "step": 9445
    },
    {
      "entropy": 5.235417652130127,
      "epoch": 8.583371194911404,
      "grad_norm": 1.4375,
      "learning_rate": 7.618581905172385e-05,
      "loss": 4.4507,
      "mean_token_accuracy": 0.2493142768740654,
      "num_tokens": 19566824.0,
      "step": 9450
    },
    {
      "entropy": 5.301792764663697,
      "epoch": 8.587914584279872,
      "grad_norm": 1.2265625,
      "learning_rate": 7.60205862303342e-05,
      "loss": 4.4416,
      "mean_token_accuracy": 0.25014214366674425,
      "num_tokens": 19576917.0,
      "step": 9455
    },
    {
      "entropy": 5.305036830902099,
      "epoch": 8.592457973648342,
      "grad_norm": 1.2265625,
      "learning_rate": 7.585584437086806e-05,
      "loss": 4.5069,
      "mean_token_accuracy": 0.24072050899267197,
      "num_tokens": 19587651.0,
      "step": 9460
    },
    {
      "entropy": 5.301551580429077,
      "epoch": 8.59700136301681,
      "grad_norm": 1.2734375,
      "learning_rate": 7.569159387980958e-05,
      "loss": 4.482,
      "mean_token_accuracy": 0.25221242010593414,
      "num_tokens": 19598421.0,
      "step": 9465
    },
    {
      "entropy": 5.266011571884155,
      "epoch": 8.60154475238528,
      "grad_norm": 1.328125,
      "learning_rate": 7.552783516243056e-05,
      "loss": 4.4502,
      "mean_token_accuracy": 0.257013276219368,
      "num_tokens": 19608715.0,
      "step": 9470
    },
    {
      "entropy": 5.313950157165527,
      "epoch": 8.606088141753748,
      "grad_norm": 1.3515625,
      "learning_rate": 7.536456862278936e-05,
      "loss": 4.5065,
      "mean_token_accuracy": 0.24253215342760087,
      "num_tokens": 19619676.0,
      "step": 9475
    },
    {
      "entropy": 5.3228593349456785,
      "epoch": 8.610631531122218,
      "grad_norm": 1.1796875,
      "learning_rate": 7.520179466372988e-05,
      "loss": 4.5489,
      "mean_token_accuracy": 0.24673676490783691,
      "num_tokens": 19630837.0,
      "step": 9480
    },
    {
      "entropy": 5.257311773300171,
      "epoch": 8.615174920490686,
      "grad_norm": 1.4296875,
      "learning_rate": 7.503951368688068e-05,
      "loss": 4.3869,
      "mean_token_accuracy": 0.25306581407785417,
      "num_tokens": 19640607.0,
      "step": 9485
    },
    {
      "entropy": 5.32319598197937,
      "epoch": 8.619718309859154,
      "grad_norm": 1.3125,
      "learning_rate": 7.487772609265401e-05,
      "loss": 4.5759,
      "mean_token_accuracy": 0.24194207638502122,
      "num_tokens": 19651354.0,
      "step": 9490
    },
    {
      "entropy": 5.288431215286255,
      "epoch": 8.624261699227624,
      "grad_norm": 1.453125,
      "learning_rate": 7.471643228024466e-05,
      "loss": 4.4654,
      "mean_token_accuracy": 0.24967682212591172,
      "num_tokens": 19661266.0,
      "step": 9495
    },
    {
      "entropy": 5.285655355453491,
      "epoch": 8.628805088596092,
      "grad_norm": 1.3359375,
      "learning_rate": 7.455563264762903e-05,
      "loss": 4.506,
      "mean_token_accuracy": 0.24543285816907884,
      "num_tokens": 19671275.0,
      "step": 9500
    },
    {
      "epoch": 8.628805088596092,
      "eval_entropy": 5.186778477735298,
      "eval_loss": 5.662213325500488,
      "eval_mean_token_accuracy": 0.1760079992718475,
      "eval_num_tokens": 19671275.0,
      "eval_runtime": 2.0185,
      "eval_samples_per_second": 1701.732,
      "eval_steps_per_second": 213.026,
      "step": 9500
    }
  ],
  "logging_steps": 5,
  "max_steps": 11000,
  "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": 4232055032248320.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}