{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.11808466670602823,
  "eval_steps": 500,
  "global_step": 2000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 10.691952991485596,
      "epoch": 0.00029521166676507054,
      "grad_norm": 13.5,
      "learning_rate": 2e-06,
      "loss": 10.8063,
      "mean_token_accuracy": 0.00022093121660873293,
      "num_tokens": 9044.0,
      "step": 5
    },
    {
      "entropy": 10.691949939727783,
      "epoch": 0.0005904233335301411,
      "grad_norm": 12.3125,
      "learning_rate": 4.5e-06,
      "loss": 10.7746,
      "mean_token_accuracy": 0.00021052630618214607,
      "num_tokens": 20049.0,
      "step": 10
    },
    {
      "entropy": 10.691005229949951,
      "epoch": 0.0008856350002952116,
      "grad_norm": 10.5,
      "learning_rate": 7e-06,
      "loss": 10.5243,
      "mean_token_accuracy": 0.02576384658459574,
      "num_tokens": 30441.0,
      "step": 15
    },
    {
      "entropy": 10.681586647033692,
      "epoch": 0.0011808466670602822,
      "grad_norm": 6.03125,
      "learning_rate": 9.5e-06,
      "loss": 10.1811,
      "mean_token_accuracy": 0.04208611287176609,
      "num_tokens": 41245.0,
      "step": 20
    },
    {
      "entropy": 10.640601921081544,
      "epoch": 0.0014760583338253527,
      "grad_norm": 4.0625,
      "learning_rate": 1.2e-05,
      "loss": 9.9191,
      "mean_token_accuracy": 0.041578793339431284,
      "num_tokens": 51815.0,
      "step": 25
    },
    {
      "entropy": 10.59462308883667,
      "epoch": 0.0017712700005904232,
      "grad_norm": 3.203125,
      "learning_rate": 1.4500000000000002e-05,
      "loss": 9.7236,
      "mean_token_accuracy": 0.046854367852211,
      "num_tokens": 62291.0,
      "step": 30
    },
    {
      "entropy": 10.566697216033935,
      "epoch": 0.002066481667355494,
      "grad_norm": 2.765625,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 9.6479,
      "mean_token_accuracy": 0.04647400453686714,
      "num_tokens": 72859.0,
      "step": 35
    },
    {
      "entropy": 10.56708059310913,
      "epoch": 0.0023616933341205643,
      "grad_norm": 2.609375,
      "learning_rate": 1.95e-05,
      "loss": 9.6398,
      "mean_token_accuracy": 0.04794426709413528,
      "num_tokens": 83349.0,
      "step": 40
    },
    {
      "entropy": 10.572033309936524,
      "epoch": 0.002656905000885635,
      "grad_norm": 2.46875,
      "learning_rate": 2.2e-05,
      "loss": 9.6101,
      "mean_token_accuracy": 0.04311534278094768,
      "num_tokens": 93550.0,
      "step": 45
    },
    {
      "entropy": 10.580781936645508,
      "epoch": 0.0029521166676507054,
      "grad_norm": 2.375,
      "learning_rate": 2.4500000000000003e-05,
      "loss": 9.5518,
      "mean_token_accuracy": 0.0487774446606636,
      "num_tokens": 104354.0,
      "step": 50
    },
    {
      "entropy": 10.563411712646484,
      "epoch": 0.003247328334415776,
      "grad_norm": 2.296875,
      "learning_rate": 2.7e-05,
      "loss": 9.4688,
      "mean_token_accuracy": 0.05522352792322636,
      "num_tokens": 114901.0,
      "step": 55
    },
    {
      "entropy": 10.49093132019043,
      "epoch": 0.0035425400011808465,
      "grad_norm": 2.453125,
      "learning_rate": 2.95e-05,
      "loss": 9.3461,
      "mean_token_accuracy": 0.05991581827402115,
      "num_tokens": 125378.0,
      "step": 60
    },
    {
      "entropy": 10.3884859085083,
      "epoch": 0.0038377516679459172,
      "grad_norm": 2.28125,
      "learning_rate": 3.2e-05,
      "loss": 9.2938,
      "mean_token_accuracy": 0.05960136093199253,
      "num_tokens": 136074.0,
      "step": 65
    },
    {
      "entropy": 10.436979675292969,
      "epoch": 0.004132963334710988,
      "grad_norm": 2.21875,
      "learning_rate": 3.4500000000000005e-05,
      "loss": 9.1677,
      "mean_token_accuracy": 0.06536128893494605,
      "num_tokens": 147517.0,
      "step": 70
    },
    {
      "entropy": 10.378404426574708,
      "epoch": 0.004428175001476058,
      "grad_norm": 2.34375,
      "learning_rate": 3.7e-05,
      "loss": 9.0427,
      "mean_token_accuracy": 0.07047605030238628,
      "num_tokens": 158398.0,
      "step": 75
    },
    {
      "entropy": 10.407978820800782,
      "epoch": 0.004723386668241129,
      "grad_norm": 2.359375,
      "learning_rate": 3.95e-05,
      "loss": 8.9631,
      "mean_token_accuracy": 0.07082438990473747,
      "num_tokens": 168721.0,
      "step": 80
    },
    {
      "entropy": 10.277961730957031,
      "epoch": 0.0050185983350062,
      "grad_norm": 2.453125,
      "learning_rate": 4.2000000000000004e-05,
      "loss": 8.9505,
      "mean_token_accuracy": 0.06753718703985215,
      "num_tokens": 179053.0,
      "step": 85
    },
    {
      "entropy": 10.263121604919434,
      "epoch": 0.00531381000177127,
      "grad_norm": 2.1875,
      "learning_rate": 4.45e-05,
      "loss": 8.81,
      "mean_token_accuracy": 0.06845465488731861,
      "num_tokens": 189377.0,
      "step": 90
    },
    {
      "entropy": 10.137414455413818,
      "epoch": 0.0056090216685363405,
      "grad_norm": 2.078125,
      "learning_rate": 4.7000000000000004e-05,
      "loss": 8.6659,
      "mean_token_accuracy": 0.08077330142259598,
      "num_tokens": 199316.0,
      "step": 95
    },
    {
      "entropy": 10.143539524078369,
      "epoch": 0.005904233335301411,
      "grad_norm": 2.03125,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 8.5703,
      "mean_token_accuracy": 0.07332179993391037,
      "num_tokens": 209735.0,
      "step": 100
    },
    {
      "entropy": 10.063137626647949,
      "epoch": 0.006199445002066482,
      "grad_norm": 1.9296875,
      "learning_rate": 5.2e-05,
      "loss": 8.5061,
      "mean_token_accuracy": 0.07544220425188541,
      "num_tokens": 220153.0,
      "step": 105
    },
    {
      "entropy": 10.078357124328614,
      "epoch": 0.006494656668831552,
      "grad_norm": 1.8359375,
      "learning_rate": 5.45e-05,
      "loss": 8.4012,
      "mean_token_accuracy": 0.07546676695346832,
      "num_tokens": 231226.0,
      "step": 110
    },
    {
      "entropy": 9.911471176147462,
      "epoch": 0.006789868335596623,
      "grad_norm": 1.8515625,
      "learning_rate": 5.7e-05,
      "loss": 8.2365,
      "mean_token_accuracy": 0.07300828918814659,
      "num_tokens": 241715.0,
      "step": 115
    },
    {
      "entropy": 9.870896339416504,
      "epoch": 0.007085080002361693,
      "grad_norm": 1.5859375,
      "learning_rate": 5.9499999999999996e-05,
      "loss": 8.122,
      "mean_token_accuracy": 0.07934067547321319,
      "num_tokens": 253683.0,
      "step": 120
    },
    {
      "entropy": 9.708064746856689,
      "epoch": 0.007380291669126764,
      "grad_norm": 1.75,
      "learning_rate": 6.2e-05,
      "loss": 7.9982,
      "mean_token_accuracy": 0.07759345918893815,
      "num_tokens": 262863.0,
      "step": 125
    },
    {
      "entropy": 9.583840370178223,
      "epoch": 0.0076755033358918345,
      "grad_norm": 1.640625,
      "learning_rate": 6.450000000000001e-05,
      "loss": 7.9066,
      "mean_token_accuracy": 0.07621643394231796,
      "num_tokens": 273237.0,
      "step": 130
    },
    {
      "entropy": 9.455964183807373,
      "epoch": 0.007970715002656906,
      "grad_norm": 1.5390625,
      "learning_rate": 6.7e-05,
      "loss": 7.8192,
      "mean_token_accuracy": 0.0756739940494299,
      "num_tokens": 283772.0,
      "step": 135
    },
    {
      "entropy": 9.128956699371338,
      "epoch": 0.008265926669421976,
      "grad_norm": 1.5859375,
      "learning_rate": 6.950000000000001e-05,
      "loss": 7.686,
      "mean_token_accuracy": 0.08342597484588624,
      "num_tokens": 294776.0,
      "step": 140
    },
    {
      "entropy": 8.970587921142577,
      "epoch": 0.008561138336187046,
      "grad_norm": 1.453125,
      "learning_rate": 7.2e-05,
      "loss": 7.6361,
      "mean_token_accuracy": 0.07761476486921311,
      "num_tokens": 305244.0,
      "step": 145
    },
    {
      "entropy": 8.766981601715088,
      "epoch": 0.008856350002952117,
      "grad_norm": 1.328125,
      "learning_rate": 7.45e-05,
      "loss": 7.5218,
      "mean_token_accuracy": 0.08236418589949608,
      "num_tokens": 315175.0,
      "step": 150
    },
    {
      "entropy": 8.568089389801026,
      "epoch": 0.009151561669717187,
      "grad_norm": 1.328125,
      "learning_rate": 7.7e-05,
      "loss": 7.444,
      "mean_token_accuracy": 0.07824666872620582,
      "num_tokens": 326648.0,
      "step": 155
    },
    {
      "entropy": 8.377489757537841,
      "epoch": 0.009446773336482257,
      "grad_norm": 1.2421875,
      "learning_rate": 7.950000000000001e-05,
      "loss": 7.4133,
      "mean_token_accuracy": 0.07897001653909683,
      "num_tokens": 336905.0,
      "step": 160
    },
    {
      "entropy": 8.255996799468994,
      "epoch": 0.009741985003247328,
      "grad_norm": 1.3359375,
      "learning_rate": 8.2e-05,
      "loss": 7.3357,
      "mean_token_accuracy": 0.08415433838963508,
      "num_tokens": 348163.0,
      "step": 165
    },
    {
      "entropy": 8.156648921966553,
      "epoch": 0.0100371966700124,
      "grad_norm": 1.3359375,
      "learning_rate": 8.450000000000001e-05,
      "loss": 7.3681,
      "mean_token_accuracy": 0.08406007140874863,
      "num_tokens": 359420.0,
      "step": 170
    },
    {
      "entropy": 8.059180402755738,
      "epoch": 0.01033240833677747,
      "grad_norm": 1.328125,
      "learning_rate": 8.7e-05,
      "loss": 7.2938,
      "mean_token_accuracy": 0.08481760621070862,
      "num_tokens": 369712.0,
      "step": 175
    },
    {
      "entropy": 7.922717761993408,
      "epoch": 0.01062762000354254,
      "grad_norm": 1.234375,
      "learning_rate": 8.95e-05,
      "loss": 7.2317,
      "mean_token_accuracy": 0.08849364891648293,
      "num_tokens": 380413.0,
      "step": 180
    },
    {
      "entropy": 7.878174924850464,
      "epoch": 0.01092283167030761,
      "grad_norm": 1.3984375,
      "learning_rate": 9.2e-05,
      "loss": 7.2576,
      "mean_token_accuracy": 0.08616138696670532,
      "num_tokens": 391428.0,
      "step": 185
    },
    {
      "entropy": 7.907813787460327,
      "epoch": 0.011218043337072681,
      "grad_norm": 1.2109375,
      "learning_rate": 9.45e-05,
      "loss": 7.1902,
      "mean_token_accuracy": 0.08708868958055974,
      "num_tokens": 402383.0,
      "step": 190
    },
    {
      "entropy": 7.741433143615723,
      "epoch": 0.011513255003837751,
      "grad_norm": 1.4453125,
      "learning_rate": 9.7e-05,
      "loss": 7.2219,
      "mean_token_accuracy": 0.08469741642475129,
      "num_tokens": 412692.0,
      "step": 195
    },
    {
      "entropy": 7.7598718166351315,
      "epoch": 0.011808466670602822,
      "grad_norm": 1.640625,
      "learning_rate": 9.95e-05,
      "loss": 7.1954,
      "mean_token_accuracy": 0.08851397931575775,
      "num_tokens": 423175.0,
      "step": 200
    },
    {
      "entropy": 7.6599428176879885,
      "epoch": 0.012103678337367892,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000102,
      "loss": 7.164,
      "mean_token_accuracy": 0.08638672009110451,
      "num_tokens": 433076.0,
      "step": 205
    },
    {
      "entropy": 7.721544361114502,
      "epoch": 0.012398890004132964,
      "grad_norm": 1.296875,
      "learning_rate": 0.00010449999999999999,
      "loss": 7.1491,
      "mean_token_accuracy": 0.08772207424044609,
      "num_tokens": 443338.0,
      "step": 210
    },
    {
      "entropy": 7.719466781616211,
      "epoch": 0.012694101670898034,
      "grad_norm": 1.21875,
      "learning_rate": 0.000107,
      "loss": 7.0918,
      "mean_token_accuracy": 0.08484251573681831,
      "num_tokens": 453674.0,
      "step": 215
    },
    {
      "entropy": 7.642396736145019,
      "epoch": 0.012989313337663105,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001095,
      "loss": 7.112,
      "mean_token_accuracy": 0.0881990872323513,
      "num_tokens": 464660.0,
      "step": 220
    },
    {
      "entropy": 7.623027515411377,
      "epoch": 0.013284525004428175,
      "grad_norm": 1.109375,
      "learning_rate": 0.000112,
      "loss": 7.135,
      "mean_token_accuracy": 0.08962138071656227,
      "num_tokens": 475478.0,
      "step": 225
    },
    {
      "entropy": 7.573812961578369,
      "epoch": 0.013579736671193245,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001145,
      "loss": 7.0507,
      "mean_token_accuracy": 0.0912325844168663,
      "num_tokens": 486279.0,
      "step": 230
    },
    {
      "entropy": 7.559884262084961,
      "epoch": 0.013874948337958316,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00011700000000000001,
      "loss": 7.1265,
      "mean_token_accuracy": 0.08900480046868324,
      "num_tokens": 497947.0,
      "step": 235
    },
    {
      "entropy": 7.493877267837524,
      "epoch": 0.014170160004723386,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00011949999999999999,
      "loss": 6.948,
      "mean_token_accuracy": 0.09850702658295632,
      "num_tokens": 508492.0,
      "step": 240
    },
    {
      "entropy": 7.5359265327453615,
      "epoch": 0.014465371671488458,
      "grad_norm": 1.296875,
      "learning_rate": 0.000122,
      "loss": 7.034,
      "mean_token_accuracy": 0.09341262727975845,
      "num_tokens": 519231.0,
      "step": 245
    },
    {
      "entropy": 7.547327709197998,
      "epoch": 0.014760583338253528,
      "grad_norm": 1.5,
      "learning_rate": 0.0001245,
      "loss": 7.0388,
      "mean_token_accuracy": 0.08856186792254447,
      "num_tokens": 529936.0,
      "step": 250
    },
    {
      "entropy": 7.387069749832153,
      "epoch": 0.015055795005018599,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000127,
      "loss": 7.0339,
      "mean_token_accuracy": 0.09349972456693649,
      "num_tokens": 540315.0,
      "step": 255
    },
    {
      "entropy": 7.444060134887695,
      "epoch": 0.015351006671783669,
      "grad_norm": 1.203125,
      "learning_rate": 0.0001295,
      "loss": 6.9648,
      "mean_token_accuracy": 0.097850751131773,
      "num_tokens": 551587.0,
      "step": 260
    },
    {
      "entropy": 7.412183332443237,
      "epoch": 0.01564621833854874,
      "grad_norm": 1.234375,
      "learning_rate": 0.000132,
      "loss": 6.9316,
      "mean_token_accuracy": 0.09379192665219308,
      "num_tokens": 561857.0,
      "step": 265
    },
    {
      "entropy": 7.429265975952148,
      "epoch": 0.01594143000531381,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00013450000000000002,
      "loss": 6.9717,
      "mean_token_accuracy": 0.09063394144177436,
      "num_tokens": 571784.0,
      "step": 270
    },
    {
      "entropy": 7.434524583816528,
      "epoch": 0.01623664167207888,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00013700000000000002,
      "loss": 7.0056,
      "mean_token_accuracy": 0.09408731907606124,
      "num_tokens": 583475.0,
      "step": 275
    },
    {
      "entropy": 7.450810766220092,
      "epoch": 0.016531853338843952,
      "grad_norm": 1.203125,
      "learning_rate": 0.0001395,
      "loss": 6.9856,
      "mean_token_accuracy": 0.08944389000535011,
      "num_tokens": 594752.0,
      "step": 280
    },
    {
      "entropy": 7.2732994556427,
      "epoch": 0.016827065005609022,
      "grad_norm": 1.3125,
      "learning_rate": 0.00014199999999999998,
      "loss": 6.9268,
      "mean_token_accuracy": 0.09354007542133332,
      "num_tokens": 604459.0,
      "step": 285
    },
    {
      "entropy": 7.404242277145386,
      "epoch": 0.017122276672374093,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001445,
      "loss": 6.9605,
      "mean_token_accuracy": 0.09190556928515434,
      "num_tokens": 615477.0,
      "step": 290
    },
    {
      "entropy": 7.318359518051148,
      "epoch": 0.017417488339139163,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000147,
      "loss": 6.9085,
      "mean_token_accuracy": 0.09038040712475777,
      "num_tokens": 626527.0,
      "step": 295
    },
    {
      "entropy": 7.3704808235168455,
      "epoch": 0.017712700005904233,
      "grad_norm": 1.109375,
      "learning_rate": 0.0001495,
      "loss": 6.9326,
      "mean_token_accuracy": 0.08688042685389519,
      "num_tokens": 638236.0,
      "step": 300
    },
    {
      "entropy": 7.292983913421631,
      "epoch": 0.018007911672669304,
      "grad_norm": 1.0625,
      "learning_rate": 0.000152,
      "loss": 6.9029,
      "mean_token_accuracy": 0.0930778980255127,
      "num_tokens": 648836.0,
      "step": 305
    },
    {
      "entropy": 7.261309051513672,
      "epoch": 0.018303123339434374,
      "grad_norm": 1.359375,
      "learning_rate": 0.00015450000000000001,
      "loss": 6.8546,
      "mean_token_accuracy": 0.09702132865786553,
      "num_tokens": 659610.0,
      "step": 310
    },
    {
      "entropy": 7.29406886100769,
      "epoch": 0.018598335006199444,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000157,
      "loss": 6.9224,
      "mean_token_accuracy": 0.0894104540348053,
      "num_tokens": 671423.0,
      "step": 315
    },
    {
      "entropy": 7.343613338470459,
      "epoch": 0.018893546672964515,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0001595,
      "loss": 6.8996,
      "mean_token_accuracy": 0.08960375934839249,
      "num_tokens": 682645.0,
      "step": 320
    },
    {
      "entropy": 7.207725191116333,
      "epoch": 0.019188758339729585,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000162,
      "loss": 6.8131,
      "mean_token_accuracy": 0.09437809884548187,
      "num_tokens": 692315.0,
      "step": 325
    },
    {
      "entropy": 7.3380321025848385,
      "epoch": 0.019483970006494655,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00016450000000000001,
      "loss": 6.8965,
      "mean_token_accuracy": 0.09695785492658615,
      "num_tokens": 702973.0,
      "step": 330
    },
    {
      "entropy": 7.216346549987793,
      "epoch": 0.019779181673259726,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00016700000000000002,
      "loss": 6.8629,
      "mean_token_accuracy": 0.0920400932431221,
      "num_tokens": 713787.0,
      "step": 335
    },
    {
      "entropy": 7.180256462097168,
      "epoch": 0.0200743933400248,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00016950000000000003,
      "loss": 6.7804,
      "mean_token_accuracy": 0.09937777146697044,
      "num_tokens": 725357.0,
      "step": 340
    },
    {
      "entropy": 7.094199705123901,
      "epoch": 0.02036960500678987,
      "grad_norm": 1.1875,
      "learning_rate": 0.00017199999999999998,
      "loss": 6.7488,
      "mean_token_accuracy": 0.10237349718809127,
      "num_tokens": 735003.0,
      "step": 345
    },
    {
      "entropy": 7.314488554000855,
      "epoch": 0.02066481667355494,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00017449999999999999,
      "loss": 6.8222,
      "mean_token_accuracy": 0.09937268868088722,
      "num_tokens": 747278.0,
      "step": 350
    },
    {
      "entropy": 7.053475475311279,
      "epoch": 0.02096002834032001,
      "grad_norm": 1.3125,
      "learning_rate": 0.000177,
      "loss": 6.8199,
      "mean_token_accuracy": 0.0992422841489315,
      "num_tokens": 757974.0,
      "step": 355
    },
    {
      "entropy": 7.2850106716156,
      "epoch": 0.02125524000708508,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001795,
      "loss": 6.8103,
      "mean_token_accuracy": 0.09952985718846322,
      "num_tokens": 767584.0,
      "step": 360
    },
    {
      "entropy": 7.131100606918335,
      "epoch": 0.02155045167385015,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000182,
      "loss": 6.7109,
      "mean_token_accuracy": 0.09645930156111718,
      "num_tokens": 777397.0,
      "step": 365
    },
    {
      "entropy": 7.148231458663941,
      "epoch": 0.02184566334061522,
      "grad_norm": 1.171875,
      "learning_rate": 0.0001845,
      "loss": 6.7905,
      "mean_token_accuracy": 0.09331258609890938,
      "num_tokens": 788227.0,
      "step": 370
    },
    {
      "entropy": 7.114551830291748,
      "epoch": 0.02214087500738029,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000187,
      "loss": 6.7458,
      "mean_token_accuracy": 0.09811887294054031,
      "num_tokens": 799168.0,
      "step": 375
    },
    {
      "entropy": 7.097161912918091,
      "epoch": 0.022436086674145362,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0001895,
      "loss": 6.7355,
      "mean_token_accuracy": 0.10040534660220146,
      "num_tokens": 810199.0,
      "step": 380
    },
    {
      "entropy": 7.139192867279053,
      "epoch": 0.022731298340910432,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000192,
      "loss": 6.7409,
      "mean_token_accuracy": 0.09671718701720237,
      "num_tokens": 822391.0,
      "step": 385
    },
    {
      "entropy": 7.088096618652344,
      "epoch": 0.023026510007675503,
      "grad_norm": 1.15625,
      "learning_rate": 0.0001945,
      "loss": 6.8565,
      "mean_token_accuracy": 0.09625174552202224,
      "num_tokens": 834350.0,
      "step": 390
    },
    {
      "entropy": 7.0904669761657715,
      "epoch": 0.023321721674440573,
      "grad_norm": 1.0625,
      "learning_rate": 0.00019700000000000002,
      "loss": 6.7523,
      "mean_token_accuracy": 0.09790571630001069,
      "num_tokens": 845709.0,
      "step": 395
    },
    {
      "entropy": 6.981272554397583,
      "epoch": 0.023616933341205643,
      "grad_norm": 1.375,
      "learning_rate": 0.00019950000000000002,
      "loss": 6.7085,
      "mean_token_accuracy": 0.10158270671963691,
      "num_tokens": 855092.0,
      "step": 400
    },
    {
      "entropy": 7.02293906211853,
      "epoch": 0.023912145007970714,
      "grad_norm": 1.21875,
      "learning_rate": 0.000202,
      "loss": 6.7193,
      "mean_token_accuracy": 0.08962539583444595,
      "num_tokens": 865943.0,
      "step": 405
    },
    {
      "entropy": 7.03565149307251,
      "epoch": 0.024207356674735784,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00020449999999999998,
      "loss": 6.647,
      "mean_token_accuracy": 0.09619486778974533,
      "num_tokens": 876649.0,
      "step": 410
    },
    {
      "entropy": 6.996511030197143,
      "epoch": 0.024502568341500858,
      "grad_norm": 1.265625,
      "learning_rate": 0.000207,
      "loss": 6.6861,
      "mean_token_accuracy": 0.10710784569382667,
      "num_tokens": 886221.0,
      "step": 415
    },
    {
      "entropy": 7.017127513885498,
      "epoch": 0.024797780008265928,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002095,
      "loss": 6.7047,
      "mean_token_accuracy": 0.10007574260234833,
      "num_tokens": 896234.0,
      "step": 420
    },
    {
      "entropy": 6.983069896697998,
      "epoch": 0.025092991675031,
      "grad_norm": 1.375,
      "learning_rate": 0.000212,
      "loss": 6.6479,
      "mean_token_accuracy": 0.09976074621081352,
      "num_tokens": 907198.0,
      "step": 425
    },
    {
      "entropy": 6.8837158203125,
      "epoch": 0.02538820334179607,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002145,
      "loss": 6.702,
      "mean_token_accuracy": 0.10629369989037514,
      "num_tokens": 917579.0,
      "step": 430
    },
    {
      "entropy": 6.978274822235107,
      "epoch": 0.02568341500856114,
      "grad_norm": 1.09375,
      "learning_rate": 0.00021700000000000002,
      "loss": 6.6347,
      "mean_token_accuracy": 0.10065946727991104,
      "num_tokens": 928763.0,
      "step": 435
    },
    {
      "entropy": 6.867664194107055,
      "epoch": 0.02597862667532621,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002195,
      "loss": 6.6455,
      "mean_token_accuracy": 0.1076437845826149,
      "num_tokens": 939383.0,
      "step": 440
    },
    {
      "entropy": 7.063400411605835,
      "epoch": 0.02627383834209128,
      "grad_norm": 1.234375,
      "learning_rate": 0.000222,
      "loss": 6.7118,
      "mean_token_accuracy": 0.09779355004429817,
      "num_tokens": 949967.0,
      "step": 445
    },
    {
      "entropy": 6.977987289428711,
      "epoch": 0.02656905000885635,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002245,
      "loss": 6.7177,
      "mean_token_accuracy": 0.09856124073266984,
      "num_tokens": 960438.0,
      "step": 450
    },
    {
      "entropy": 6.9721705436706545,
      "epoch": 0.02686426167562142,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00022700000000000002,
      "loss": 6.678,
      "mean_token_accuracy": 0.1020229034125805,
      "num_tokens": 971771.0,
      "step": 455
    },
    {
      "entropy": 7.0487203121185305,
      "epoch": 0.02715947334238649,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00022950000000000002,
      "loss": 6.7437,
      "mean_token_accuracy": 0.09785925075411797,
      "num_tokens": 982431.0,
      "step": 460
    },
    {
      "entropy": 6.8306797504425045,
      "epoch": 0.02745468500915156,
      "grad_norm": 1.3125,
      "learning_rate": 0.00023200000000000003,
      "loss": 6.69,
      "mean_token_accuracy": 0.09700809046626091,
      "num_tokens": 992763.0,
      "step": 465
    },
    {
      "entropy": 6.909020566940308,
      "epoch": 0.02774989667591663,
      "grad_norm": 1.265625,
      "learning_rate": 0.00023449999999999998,
      "loss": 6.6022,
      "mean_token_accuracy": 0.107904601842165,
      "num_tokens": 1002977.0,
      "step": 470
    },
    {
      "entropy": 6.922635841369629,
      "epoch": 0.0280451083426817,
      "grad_norm": 1.21875,
      "learning_rate": 0.000237,
      "loss": 6.6653,
      "mean_token_accuracy": 0.10020340532064438,
      "num_tokens": 1014211.0,
      "step": 475
    },
    {
      "entropy": 6.923598337173462,
      "epoch": 0.028340320009446772,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0002395,
      "loss": 6.6715,
      "mean_token_accuracy": 0.10355789959430695,
      "num_tokens": 1025625.0,
      "step": 480
    },
    {
      "entropy": 6.909690856933594,
      "epoch": 0.028635531676211842,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000242,
      "loss": 6.686,
      "mean_token_accuracy": 0.10261352136731147,
      "num_tokens": 1036430.0,
      "step": 485
    },
    {
      "entropy": 6.986854982376099,
      "epoch": 0.028930743342976916,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0002445,
      "loss": 6.7083,
      "mean_token_accuracy": 0.10415773317217827,
      "num_tokens": 1047246.0,
      "step": 490
    },
    {
      "entropy": 6.8165546417236325,
      "epoch": 0.029225955009741986,
      "grad_norm": 1.0625,
      "learning_rate": 0.000247,
      "loss": 6.5998,
      "mean_token_accuracy": 0.10099067613482475,
      "num_tokens": 1057004.0,
      "step": 495
    },
    {
      "entropy": 6.876231527328491,
      "epoch": 0.029521166676507057,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002495,
      "loss": 6.6397,
      "mean_token_accuracy": 0.1102587789297104,
      "num_tokens": 1067224.0,
      "step": 500
    },
    {
      "entropy": 6.866609764099121,
      "epoch": 0.029816378343272127,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000252,
      "loss": 6.6158,
      "mean_token_accuracy": 0.1031361848115921,
      "num_tokens": 1077523.0,
      "step": 505
    },
    {
      "entropy": 6.851018238067627,
      "epoch": 0.030111590010037197,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002545,
      "loss": 6.5845,
      "mean_token_accuracy": 0.1057098388671875,
      "num_tokens": 1088591.0,
      "step": 510
    },
    {
      "entropy": 6.780634880065918,
      "epoch": 0.030406801676802268,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000257,
      "loss": 6.6068,
      "mean_token_accuracy": 0.10300912559032441,
      "num_tokens": 1100634.0,
      "step": 515
    },
    {
      "entropy": 6.772457027435303,
      "epoch": 0.030702013343567338,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002595,
      "loss": 6.5604,
      "mean_token_accuracy": 0.10456541404128075,
      "num_tokens": 1111231.0,
      "step": 520
    },
    {
      "entropy": 6.864635372161866,
      "epoch": 0.03099722501033241,
      "grad_norm": 1.40625,
      "learning_rate": 0.000262,
      "loss": 6.5368,
      "mean_token_accuracy": 0.11013480946421624,
      "num_tokens": 1121112.0,
      "step": 525
    },
    {
      "entropy": 6.817177820205688,
      "epoch": 0.03129243667709748,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00026450000000000003,
      "loss": 6.6807,
      "mean_token_accuracy": 0.10330734625458718,
      "num_tokens": 1130803.0,
      "step": 530
    },
    {
      "entropy": 6.852666234970092,
      "epoch": 0.03158764834386255,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00026700000000000004,
      "loss": 6.6568,
      "mean_token_accuracy": 0.09623496234416962,
      "num_tokens": 1141621.0,
      "step": 535
    },
    {
      "entropy": 6.851540040969849,
      "epoch": 0.03188286001062762,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00026950000000000005,
      "loss": 6.6752,
      "mean_token_accuracy": 0.10156160518527031,
      "num_tokens": 1152860.0,
      "step": 540
    },
    {
      "entropy": 6.910813045501709,
      "epoch": 0.03217807167739269,
      "grad_norm": 1.546875,
      "learning_rate": 0.00027200000000000005,
      "loss": 6.5423,
      "mean_token_accuracy": 0.10722377449274063,
      "num_tokens": 1162958.0,
      "step": 545
    },
    {
      "entropy": 6.7858928680419925,
      "epoch": 0.03247328334415776,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002745,
      "loss": 6.6395,
      "mean_token_accuracy": 0.10390800461173058,
      "num_tokens": 1174205.0,
      "step": 550
    },
    {
      "entropy": 6.813560676574707,
      "epoch": 0.032768495010922834,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000277,
      "loss": 6.5218,
      "mean_token_accuracy": 0.10657795593142509,
      "num_tokens": 1184336.0,
      "step": 555
    },
    {
      "entropy": 6.759924650192261,
      "epoch": 0.033063706677687904,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002795,
      "loss": 6.5285,
      "mean_token_accuracy": 0.11047871261835099,
      "num_tokens": 1194696.0,
      "step": 560
    },
    {
      "entropy": 6.675027894973755,
      "epoch": 0.033358918344452974,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00028199999999999997,
      "loss": 6.4689,
      "mean_token_accuracy": 0.10344288945198059,
      "num_tokens": 1206440.0,
      "step": 565
    },
    {
      "entropy": 6.843683671951294,
      "epoch": 0.033654130011218045,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002845,
      "loss": 6.545,
      "mean_token_accuracy": 0.10641451999545097,
      "num_tokens": 1216229.0,
      "step": 570
    },
    {
      "entropy": 6.739560270309449,
      "epoch": 0.033949341677983115,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000287,
      "loss": 6.6295,
      "mean_token_accuracy": 0.1049153484404087,
      "num_tokens": 1228077.0,
      "step": 575
    },
    {
      "entropy": 6.755360412597656,
      "epoch": 0.034244553344748185,
      "grad_norm": 1.125,
      "learning_rate": 0.0002895,
      "loss": 6.6024,
      "mean_token_accuracy": 0.10700471252202988,
      "num_tokens": 1239047.0,
      "step": 580
    },
    {
      "entropy": 6.835167455673218,
      "epoch": 0.034539765011513256,
      "grad_norm": 1.125,
      "learning_rate": 0.000292,
      "loss": 6.5143,
      "mean_token_accuracy": 0.11052370145916939,
      "num_tokens": 1249633.0,
      "step": 585
    },
    {
      "entropy": 6.736393117904663,
      "epoch": 0.034834976678278326,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002945,
      "loss": 6.5858,
      "mean_token_accuracy": 0.10385486260056495,
      "num_tokens": 1261085.0,
      "step": 590
    },
    {
      "entropy": 6.705540657043457,
      "epoch": 0.035130188345043396,
      "grad_norm": 1.3125,
      "learning_rate": 0.000297,
      "loss": 6.4767,
      "mean_token_accuracy": 0.11535756587982178,
      "num_tokens": 1271161.0,
      "step": 595
    },
    {
      "entropy": 6.711465167999267,
      "epoch": 0.03542540001180847,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002995,
      "loss": 6.5664,
      "mean_token_accuracy": 0.1055572934448719,
      "num_tokens": 1281776.0,
      "step": 600
    },
    {
      "entropy": 6.741065502166748,
      "epoch": 0.03572061167857354,
      "grad_norm": 1.109375,
      "learning_rate": 0.000302,
      "loss": 6.5441,
      "mean_token_accuracy": 0.10771351680159569,
      "num_tokens": 1291759.0,
      "step": 605
    },
    {
      "entropy": 6.709814834594726,
      "epoch": 0.03601582334533861,
      "grad_norm": 1.0,
      "learning_rate": 0.0003045,
      "loss": 6.4237,
      "mean_token_accuracy": 0.1066345103085041,
      "num_tokens": 1301783.0,
      "step": 610
    },
    {
      "entropy": 6.701594305038452,
      "epoch": 0.03631103501210368,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000307,
      "loss": 6.5523,
      "mean_token_accuracy": 0.10823088437318802,
      "num_tokens": 1312442.0,
      "step": 615
    },
    {
      "entropy": 6.710081911087036,
      "epoch": 0.03660624667886875,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003095,
      "loss": 6.5222,
      "mean_token_accuracy": 0.11155530214309692,
      "num_tokens": 1321787.0,
      "step": 620
    },
    {
      "entropy": 6.720113277435303,
      "epoch": 0.03690145834563382,
      "grad_norm": 1.171875,
      "learning_rate": 0.000312,
      "loss": 6.522,
      "mean_token_accuracy": 0.10557924285531044,
      "num_tokens": 1331724.0,
      "step": 625
    },
    {
      "entropy": 6.668059158325195,
      "epoch": 0.03719667001239889,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003145,
      "loss": 6.5261,
      "mean_token_accuracy": 0.10685517415404319,
      "num_tokens": 1342272.0,
      "step": 630
    },
    {
      "entropy": 6.737295341491699,
      "epoch": 0.03749188167916396,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000317,
      "loss": 6.5756,
      "mean_token_accuracy": 0.10473631396889686,
      "num_tokens": 1352806.0,
      "step": 635
    },
    {
      "entropy": 6.6583808898925785,
      "epoch": 0.03778709334592903,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003195,
      "loss": 6.4683,
      "mean_token_accuracy": 0.10982004255056381,
      "num_tokens": 1363668.0,
      "step": 640
    },
    {
      "entropy": 6.669088649749756,
      "epoch": 0.0380823050126941,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000322,
      "loss": 6.5768,
      "mean_token_accuracy": 0.10401294007897377,
      "num_tokens": 1376374.0,
      "step": 645
    },
    {
      "entropy": 6.727249479293823,
      "epoch": 0.03837751667945917,
      "grad_norm": 1.15625,
      "learning_rate": 0.00032450000000000003,
      "loss": 6.5341,
      "mean_token_accuracy": 0.10710064172744752,
      "num_tokens": 1387332.0,
      "step": 650
    },
    {
      "entropy": 6.656798219680786,
      "epoch": 0.03867272834622424,
      "grad_norm": 1.125,
      "learning_rate": 0.00032700000000000003,
      "loss": 6.4362,
      "mean_token_accuracy": 0.103745087236166,
      "num_tokens": 1398761.0,
      "step": 655
    },
    {
      "entropy": 6.517245817184448,
      "epoch": 0.03896794001298931,
      "grad_norm": 1.171875,
      "learning_rate": 0.00032950000000000004,
      "loss": 6.4851,
      "mean_token_accuracy": 0.11338173747062683,
      "num_tokens": 1409818.0,
      "step": 660
    },
    {
      "entropy": 6.730355882644654,
      "epoch": 0.03926315167975438,
      "grad_norm": 1.0625,
      "learning_rate": 0.00033200000000000005,
      "loss": 6.504,
      "mean_token_accuracy": 0.10808097869157791,
      "num_tokens": 1420839.0,
      "step": 665
    },
    {
      "entropy": 6.628807258605957,
      "epoch": 0.03955836334651945,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00033450000000000005,
      "loss": 6.5066,
      "mean_token_accuracy": 0.10702955350279808,
      "num_tokens": 1434314.0,
      "step": 670
    },
    {
      "entropy": 6.578580474853515,
      "epoch": 0.03985357501328453,
      "grad_norm": 1.265625,
      "learning_rate": 0.000337,
      "loss": 6.4204,
      "mean_token_accuracy": 0.11155755072832108,
      "num_tokens": 1445404.0,
      "step": 675
    },
    {
      "entropy": 6.606809616088867,
      "epoch": 0.0401487866800496,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003395,
      "loss": 6.555,
      "mean_token_accuracy": 0.11048336997628212,
      "num_tokens": 1456860.0,
      "step": 680
    },
    {
      "entropy": 6.589565467834473,
      "epoch": 0.04044399834681467,
      "grad_norm": 1.1875,
      "learning_rate": 0.000342,
      "loss": 6.4932,
      "mean_token_accuracy": 0.11120514348149299,
      "num_tokens": 1467597.0,
      "step": 685
    },
    {
      "entropy": 6.696408987045288,
      "epoch": 0.04073921001357974,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00034449999999999997,
      "loss": 6.4581,
      "mean_token_accuracy": 0.11563440561294555,
      "num_tokens": 1478093.0,
      "step": 690
    },
    {
      "entropy": 6.656173992156982,
      "epoch": 0.04103442168034481,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000347,
      "loss": 6.5137,
      "mean_token_accuracy": 0.11482275947928429,
      "num_tokens": 1488527.0,
      "step": 695
    },
    {
      "entropy": 6.550519704818726,
      "epoch": 0.04132963334710988,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003495,
      "loss": 6.3829,
      "mean_token_accuracy": 0.11443194523453712,
      "num_tokens": 1499370.0,
      "step": 700
    },
    {
      "entropy": 6.566233539581299,
      "epoch": 0.04162484501387495,
      "grad_norm": 1.140625,
      "learning_rate": 0.000352,
      "loss": 6.4064,
      "mean_token_accuracy": 0.11202551424503326,
      "num_tokens": 1510518.0,
      "step": 705
    },
    {
      "entropy": 6.535747432708741,
      "epoch": 0.04192005668064002,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003545,
      "loss": 6.4342,
      "mean_token_accuracy": 0.11363685354590417,
      "num_tokens": 1520791.0,
      "step": 710
    },
    {
      "entropy": 6.632448720932007,
      "epoch": 0.04221526834740509,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000357,
      "loss": 6.5126,
      "mean_token_accuracy": 0.10954305157065392,
      "num_tokens": 1532122.0,
      "step": 715
    },
    {
      "entropy": 6.5457751750946045,
      "epoch": 0.04251048001417016,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003595,
      "loss": 6.3958,
      "mean_token_accuracy": 0.11548751443624497,
      "num_tokens": 1542719.0,
      "step": 720
    },
    {
      "entropy": 6.563694810867309,
      "epoch": 0.04280569168093523,
      "grad_norm": 1.140625,
      "learning_rate": 0.000362,
      "loss": 6.4559,
      "mean_token_accuracy": 0.11078683212399483,
      "num_tokens": 1553397.0,
      "step": 725
    },
    {
      "entropy": 6.572269773483276,
      "epoch": 0.0431009033477003,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003645,
      "loss": 6.4017,
      "mean_token_accuracy": 0.11554159820079804,
      "num_tokens": 1563952.0,
      "step": 730
    },
    {
      "entropy": 6.650298690795898,
      "epoch": 0.04339611501446537,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000367,
      "loss": 6.4341,
      "mean_token_accuracy": 0.11468867510557175,
      "num_tokens": 1574248.0,
      "step": 735
    },
    {
      "entropy": 6.517534494400024,
      "epoch": 0.04369132668123044,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003695,
      "loss": 6.4001,
      "mean_token_accuracy": 0.11306158527731895,
      "num_tokens": 1584325.0,
      "step": 740
    },
    {
      "entropy": 6.510380697250366,
      "epoch": 0.04398653834799551,
      "grad_norm": 1.0625,
      "learning_rate": 0.000372,
      "loss": 6.3622,
      "mean_token_accuracy": 0.1204331710934639,
      "num_tokens": 1593737.0,
      "step": 745
    },
    {
      "entropy": 6.5366417407989506,
      "epoch": 0.04428175001476058,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003745,
      "loss": 6.42,
      "mean_token_accuracy": 0.11730596050620079,
      "num_tokens": 1604161.0,
      "step": 750
    },
    {
      "entropy": 6.487906742095947,
      "epoch": 0.044576961681525654,
      "grad_norm": 1.171875,
      "learning_rate": 0.000377,
      "loss": 6.3659,
      "mean_token_accuracy": 0.11893260702490807,
      "num_tokens": 1614719.0,
      "step": 755
    },
    {
      "entropy": 6.4263794898986815,
      "epoch": 0.044872173348290724,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003795,
      "loss": 6.3783,
      "mean_token_accuracy": 0.11692238673567772,
      "num_tokens": 1626450.0,
      "step": 760
    },
    {
      "entropy": 6.564861059188843,
      "epoch": 0.045167385015055794,
      "grad_norm": 1.03125,
      "learning_rate": 0.000382,
      "loss": 6.444,
      "mean_token_accuracy": 0.10918903425335884,
      "num_tokens": 1638297.0,
      "step": 765
    },
    {
      "entropy": 6.559746789932251,
      "epoch": 0.045462596681820865,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003845,
      "loss": 6.4014,
      "mean_token_accuracy": 0.11586136147379875,
      "num_tokens": 1648744.0,
      "step": 770
    },
    {
      "entropy": 6.450838375091553,
      "epoch": 0.045757808348585935,
      "grad_norm": 1.125,
      "learning_rate": 0.00038700000000000003,
      "loss": 6.3877,
      "mean_token_accuracy": 0.11436291560530662,
      "num_tokens": 1660291.0,
      "step": 775
    },
    {
      "entropy": 6.512089347839355,
      "epoch": 0.046053020015351005,
      "grad_norm": 1.25,
      "learning_rate": 0.00038950000000000003,
      "loss": 6.3968,
      "mean_token_accuracy": 0.12081583142280579,
      "num_tokens": 1670425.0,
      "step": 780
    },
    {
      "entropy": 6.560306644439697,
      "epoch": 0.046348231682116076,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039200000000000004,
      "loss": 6.3979,
      "mean_token_accuracy": 0.11076491773128509,
      "num_tokens": 1681390.0,
      "step": 785
    },
    {
      "entropy": 6.450212621688843,
      "epoch": 0.046643443348881146,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039450000000000005,
      "loss": 6.4569,
      "mean_token_accuracy": 0.11279196292161942,
      "num_tokens": 1690856.0,
      "step": 790
    },
    {
      "entropy": 6.431796646118164,
      "epoch": 0.046938655015646216,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039700000000000005,
      "loss": 6.3975,
      "mean_token_accuracy": 0.11749225705862046,
      "num_tokens": 1701985.0,
      "step": 795
    },
    {
      "entropy": 6.531089401245117,
      "epoch": 0.04723386668241129,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003995,
      "loss": 6.33,
      "mean_token_accuracy": 0.1177647665143013,
      "num_tokens": 1713125.0,
      "step": 800
    },
    {
      "entropy": 6.485761880874634,
      "epoch": 0.04752907834917636,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000402,
      "loss": 6.4297,
      "mean_token_accuracy": 0.11685430556535721,
      "num_tokens": 1725409.0,
      "step": 805
    },
    {
      "entropy": 6.4987870216369625,
      "epoch": 0.04782429001594143,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004045,
      "loss": 6.4797,
      "mean_token_accuracy": 0.11249323934316635,
      "num_tokens": 1736579.0,
      "step": 810
    },
    {
      "entropy": 6.412155485153198,
      "epoch": 0.0481195016827065,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00040699999999999997,
      "loss": 6.3084,
      "mean_token_accuracy": 0.11493954733014107,
      "num_tokens": 1747029.0,
      "step": 815
    },
    {
      "entropy": 6.421283435821533,
      "epoch": 0.04841471334947157,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004095,
      "loss": 6.2559,
      "mean_token_accuracy": 0.11367330402135849,
      "num_tokens": 1757444.0,
      "step": 820
    },
    {
      "entropy": 6.466654539108276,
      "epoch": 0.048709925016236645,
      "grad_norm": 1.203125,
      "learning_rate": 0.000412,
      "loss": 6.4014,
      "mean_token_accuracy": 0.11653372645378113,
      "num_tokens": 1767576.0,
      "step": 825
    },
    {
      "entropy": 6.4442650318145756,
      "epoch": 0.049005136683001715,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004145,
      "loss": 6.4257,
      "mean_token_accuracy": 0.11447887197136879,
      "num_tokens": 1777714.0,
      "step": 830
    },
    {
      "entropy": 6.418978452682495,
      "epoch": 0.049300348349766786,
      "grad_norm": 1.09375,
      "learning_rate": 0.000417,
      "loss": 6.4399,
      "mean_token_accuracy": 0.11199806556105614,
      "num_tokens": 1788857.0,
      "step": 835
    },
    {
      "entropy": 6.483746528625488,
      "epoch": 0.049595560016531856,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004195,
      "loss": 6.3862,
      "mean_token_accuracy": 0.11579035371541976,
      "num_tokens": 1800743.0,
      "step": 840
    },
    {
      "entropy": 6.482276964187622,
      "epoch": 0.049890771683296926,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000422,
      "loss": 6.3287,
      "mean_token_accuracy": 0.11087024435400963,
      "num_tokens": 1810477.0,
      "step": 845
    },
    {
      "entropy": 6.3517107486724855,
      "epoch": 0.050185983350062,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004245,
      "loss": 6.3575,
      "mean_token_accuracy": 0.11524266377091408,
      "num_tokens": 1821516.0,
      "step": 850
    },
    {
      "entropy": 6.587709379196167,
      "epoch": 0.05048119501682707,
      "grad_norm": 1.0,
      "learning_rate": 0.000427,
      "loss": 6.4653,
      "mean_token_accuracy": 0.11049215123057365,
      "num_tokens": 1832784.0,
      "step": 855
    },
    {
      "entropy": 6.34732494354248,
      "epoch": 0.05077640668359214,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004295,
      "loss": 6.2728,
      "mean_token_accuracy": 0.1179530993103981,
      "num_tokens": 1843737.0,
      "step": 860
    },
    {
      "entropy": 6.438384008407593,
      "epoch": 0.05107161835035721,
      "grad_norm": 1.0625,
      "learning_rate": 0.000432,
      "loss": 6.3462,
      "mean_token_accuracy": 0.12044489830732345,
      "num_tokens": 1854215.0,
      "step": 865
    },
    {
      "entropy": 6.3278967380523685,
      "epoch": 0.05136683001712228,
      "grad_norm": 1.125,
      "learning_rate": 0.0004345,
      "loss": 6.3059,
      "mean_token_accuracy": 0.1170185461640358,
      "num_tokens": 1864813.0,
      "step": 870
    },
    {
      "entropy": 6.467825937271118,
      "epoch": 0.05166204168388735,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000437,
      "loss": 6.364,
      "mean_token_accuracy": 0.11428075954318047,
      "num_tokens": 1874962.0,
      "step": 875
    },
    {
      "entropy": 6.507443571090699,
      "epoch": 0.05195725335065242,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004395,
      "loss": 6.4321,
      "mean_token_accuracy": 0.1131198450922966,
      "num_tokens": 1886476.0,
      "step": 880
    },
    {
      "entropy": 6.3558813571929935,
      "epoch": 0.05225246501741749,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000442,
      "loss": 6.3187,
      "mean_token_accuracy": 0.12074669077992439,
      "num_tokens": 1896860.0,
      "step": 885
    },
    {
      "entropy": 6.31374831199646,
      "epoch": 0.05254767668418256,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004445,
      "loss": 6.3,
      "mean_token_accuracy": 0.12395407184958458,
      "num_tokens": 1908727.0,
      "step": 890
    },
    {
      "entropy": 6.462383651733399,
      "epoch": 0.05284288835094763,
      "grad_norm": 1.015625,
      "learning_rate": 0.000447,
      "loss": 6.3968,
      "mean_token_accuracy": 0.11613713800907136,
      "num_tokens": 1920175.0,
      "step": 895
    },
    {
      "entropy": 6.438568925857544,
      "epoch": 0.0531381000177127,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044950000000000003,
      "loss": 6.3363,
      "mean_token_accuracy": 0.12365170270204544,
      "num_tokens": 1930267.0,
      "step": 900
    },
    {
      "entropy": 6.340644979476929,
      "epoch": 0.05343331168447777,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045200000000000004,
      "loss": 6.3596,
      "mean_token_accuracy": 0.12366607934236526,
      "num_tokens": 1940527.0,
      "step": 905
    },
    {
      "entropy": 6.441576862335205,
      "epoch": 0.05372852335124284,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045450000000000004,
      "loss": 6.4129,
      "mean_token_accuracy": 0.11123256385326385,
      "num_tokens": 1952557.0,
      "step": 910
    },
    {
      "entropy": 6.354656839370728,
      "epoch": 0.05402373501800791,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045700000000000005,
      "loss": 6.272,
      "mean_token_accuracy": 0.11737576723098755,
      "num_tokens": 1962858.0,
      "step": 915
    },
    {
      "entropy": 6.502988862991333,
      "epoch": 0.05431894668477298,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045950000000000006,
      "loss": 6.4562,
      "mean_token_accuracy": 0.10999115817248821,
      "num_tokens": 1974579.0,
      "step": 920
    },
    {
      "entropy": 6.265300655364991,
      "epoch": 0.05461415835153805,
      "grad_norm": 1.03125,
      "learning_rate": 0.000462,
      "loss": 6.2954,
      "mean_token_accuracy": 0.129633379727602,
      "num_tokens": 1984937.0,
      "step": 925
    },
    {
      "entropy": 6.484027338027954,
      "epoch": 0.05490937001830312,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004645,
      "loss": 6.3673,
      "mean_token_accuracy": 0.11584793254733086,
      "num_tokens": 1995046.0,
      "step": 930
    },
    {
      "entropy": 6.3297584533691404,
      "epoch": 0.05520458168506819,
      "grad_norm": 1.015625,
      "learning_rate": 0.000467,
      "loss": 6.3379,
      "mean_token_accuracy": 0.12105886042118072,
      "num_tokens": 2006261.0,
      "step": 935
    },
    {
      "entropy": 6.4803060531616214,
      "epoch": 0.05549979335183326,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004695,
      "loss": 6.3509,
      "mean_token_accuracy": 0.11742576733231544,
      "num_tokens": 2017470.0,
      "step": 940
    },
    {
      "entropy": 6.31043176651001,
      "epoch": 0.05579500501859833,
      "grad_norm": 1.125,
      "learning_rate": 0.000472,
      "loss": 6.3327,
      "mean_token_accuracy": 0.12294093891978264,
      "num_tokens": 2028104.0,
      "step": 945
    },
    {
      "entropy": 6.356010007858276,
      "epoch": 0.0560902166853634,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004745,
      "loss": 6.3971,
      "mean_token_accuracy": 0.11537264212965966,
      "num_tokens": 2039347.0,
      "step": 950
    },
    {
      "entropy": 6.297523021697998,
      "epoch": 0.056385428352128474,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000477,
      "loss": 6.2869,
      "mean_token_accuracy": 0.11124283596873283,
      "num_tokens": 2050464.0,
      "step": 955
    },
    {
      "entropy": 6.3561155796051025,
      "epoch": 0.056680640018893544,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004795,
      "loss": 6.2763,
      "mean_token_accuracy": 0.12250063344836234,
      "num_tokens": 2061554.0,
      "step": 960
    },
    {
      "entropy": 6.351288223266602,
      "epoch": 0.056975851685658614,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000482,
      "loss": 6.3241,
      "mean_token_accuracy": 0.11924508661031723,
      "num_tokens": 2072217.0,
      "step": 965
    },
    {
      "entropy": 6.317080402374268,
      "epoch": 0.057271063352423685,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004845,
      "loss": 6.2585,
      "mean_token_accuracy": 0.12346376478672028,
      "num_tokens": 2082088.0,
      "step": 970
    },
    {
      "entropy": 6.334501934051514,
      "epoch": 0.05756627501918876,
      "grad_norm": 1.296875,
      "learning_rate": 0.000487,
      "loss": 6.333,
      "mean_token_accuracy": 0.11622191295027733,
      "num_tokens": 2091959.0,
      "step": 975
    },
    {
      "entropy": 6.329196119308472,
      "epoch": 0.05786148668595383,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004895,
      "loss": 6.2826,
      "mean_token_accuracy": 0.12484293133020401,
      "num_tokens": 2102907.0,
      "step": 980
    },
    {
      "entropy": 6.299886989593506,
      "epoch": 0.0581566983527189,
      "grad_norm": 1.046875,
      "learning_rate": 0.000492,
      "loss": 6.3273,
      "mean_token_accuracy": 0.1229906715452671,
      "num_tokens": 2112818.0,
      "step": 985
    },
    {
      "entropy": 6.340411424636841,
      "epoch": 0.05845191001948397,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004945,
      "loss": 6.2827,
      "mean_token_accuracy": 0.12152156233787537,
      "num_tokens": 2122847.0,
      "step": 990
    },
    {
      "entropy": 6.255873918533325,
      "epoch": 0.05874712168624904,
      "grad_norm": 1.015625,
      "learning_rate": 0.000497,
      "loss": 6.2182,
      "mean_token_accuracy": 0.12108674049377441,
      "num_tokens": 2133317.0,
      "step": 995
    },
    {
      "entropy": 6.304410934448242,
      "epoch": 0.05904233335301411,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004995,
      "loss": 6.3561,
      "mean_token_accuracy": 0.11512539386749268,
      "num_tokens": 2144568.0,
      "step": 1000
    },
    {
      "entropy": 6.3178955078125,
      "epoch": 0.059337545019779184,
      "grad_norm": 1.09375,
      "learning_rate": 0.000499998026082006,
      "loss": 6.2824,
      "mean_token_accuracy": 0.123703832924366,
      "num_tokens": 2154869.0,
      "step": 1005
    },
    {
      "entropy": 6.275319528579712,
      "epoch": 0.059632756686544254,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999900070995136,
      "loss": 6.2407,
      "mean_token_accuracy": 0.12574376314878463,
      "num_tokens": 2165543.0,
      "step": 1010
    },
    {
      "entropy": 6.303437566757202,
      "epoch": 0.059927968353309324,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999758199023239,
      "loss": 6.2772,
      "mean_token_accuracy": 0.12176102846860885,
      "num_tokens": 2175873.0,
      "step": 1015
    },
    {
      "entropy": 6.239206266403198,
      "epoch": 0.060223180020074395,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999554648793858,
      "loss": 6.308,
      "mean_token_accuracy": 0.1178464025259018,
      "num_tokens": 2187968.0,
      "step": 1020
    },
    {
      "entropy": 6.357020711898803,
      "epoch": 0.060518391686839465,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999289425887425,
      "loss": 6.306,
      "mean_token_accuracy": 0.1196315847337246,
      "num_tokens": 2199153.0,
      "step": 1025
    },
    {
      "entropy": 6.238584899902344,
      "epoch": 0.060813603353604535,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998962537575161,
      "loss": 6.2466,
      "mean_token_accuracy": 0.12366516292095184,
      "num_tokens": 2210069.0,
      "step": 1030
    },
    {
      "entropy": 6.265579509735107,
      "epoch": 0.061108815020369606,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004998573992818874,
      "loss": 6.239,
      "mean_token_accuracy": 0.1157471477985382,
      "num_tokens": 2221635.0,
      "step": 1035
    },
    {
      "entropy": 6.212556266784668,
      "epoch": 0.061404026687134676,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998123802270715,
      "loss": 6.302,
      "mean_token_accuracy": 0.12124627754092217,
      "num_tokens": 2232272.0,
      "step": 1040
    },
    {
      "entropy": 6.408636903762817,
      "epoch": 0.061699238353899746,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004997611978272886,
      "loss": 6.3054,
      "mean_token_accuracy": 0.12193002700805664,
      "num_tokens": 2243504.0,
      "step": 1045
    },
    {
      "entropy": 6.175132989883423,
      "epoch": 0.06199445002066482,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004997038534857298,
      "loss": 6.227,
      "mean_token_accuracy": 0.11970143839716911,
      "num_tokens": 2253774.0,
      "step": 1050
    },
    {
      "entropy": 6.274414110183716,
      "epoch": 0.06228966168742989,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004996403487745194,
      "loss": 6.2259,
      "mean_token_accuracy": 0.11766009405255318,
      "num_tokens": 2264408.0,
      "step": 1055
    },
    {
      "entropy": 6.266839647293091,
      "epoch": 0.06258487335419496,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000499570685434671,
      "loss": 6.1665,
      "mean_token_accuracy": 0.12115408554673195,
      "num_tokens": 2274361.0,
      "step": 1060
    },
    {
      "entropy": 6.235118722915649,
      "epoch": 0.06288008502096003,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004994948653760405,
      "loss": 6.2513,
      "mean_token_accuracy": 0.12207862734794617,
      "num_tokens": 2284811.0,
      "step": 1065
    },
    {
      "entropy": 6.306531000137329,
      "epoch": 0.0631752966877251,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004994128906772729,
      "loss": 6.2779,
      "mean_token_accuracy": 0.11769364327192307,
      "num_tokens": 2295548.0,
      "step": 1070
    },
    {
      "entropy": 6.2641833305358885,
      "epoch": 0.06347050835449017,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000499324763585746,
      "loss": 6.2362,
      "mean_token_accuracy": 0.12173673734068871,
      "num_tokens": 2306185.0,
      "step": 1075
    },
    {
      "entropy": 6.290108299255371,
      "epoch": 0.06376572002125525,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004992304865175085,
      "loss": 6.2358,
      "mean_token_accuracy": 0.12764014080166816,
      "num_tokens": 2317057.0,
      "step": 1080
    },
    {
      "entropy": 6.204243850708008,
      "epoch": 0.06406093168802031,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004991300620572138,
      "loss": 6.2425,
      "mean_token_accuracy": 0.12284323722124099,
      "num_tokens": 2327066.0,
      "step": 1085
    },
    {
      "entropy": 6.2375068187713625,
      "epoch": 0.06435614335478539,
      "grad_norm": 1.0,
      "learning_rate": 0.0004990234929580494,
      "loss": 6.1871,
      "mean_token_accuracy": 0.1312653213739395,
      "num_tokens": 2337920.0,
      "step": 1090
    },
    {
      "entropy": 6.235481595993042,
      "epoch": 0.06465135502155045,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004989107821416609,
      "loss": 6.179,
      "mean_token_accuracy": 0.12998321577906607,
      "num_tokens": 2348234.0,
      "step": 1095
    },
    {
      "entropy": 6.154482316970825,
      "epoch": 0.06494656668831553,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004987919326980723,
      "loss": 6.2029,
      "mean_token_accuracy": 0.12362807467579842,
      "num_tokens": 2359112.0,
      "step": 1100
    },
    {
      "entropy": 6.2241229057312015,
      "epoch": 0.06524177835508059,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004986669478856011,
      "loss": 6.1887,
      "mean_token_accuracy": 0.1307649426162243,
      "num_tokens": 2370207.0,
      "step": 1105
    },
    {
      "entropy": 6.182603406906128,
      "epoch": 0.06553699002184567,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004985358311307688,
      "loss": 6.2972,
      "mean_token_accuracy": 0.1216890573501587,
      "num_tokens": 2381892.0,
      "step": 1110
    },
    {
      "entropy": 6.266233348846436,
      "epoch": 0.06583220168861073,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004983985860282081,
      "loss": 6.1947,
      "mean_token_accuracy": 0.12289397418498993,
      "num_tokens": 2391898.0,
      "step": 1115
    },
    {
      "entropy": 6.2191582202911375,
      "epoch": 0.06612741335537581,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004982552163405623,
      "loss": 6.225,
      "mean_token_accuracy": 0.12616347894072533,
      "num_tokens": 2402864.0,
      "step": 1120
    },
    {
      "entropy": 6.184364604949951,
      "epoch": 0.06642262502214087,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004981057259983839,
      "loss": 6.2268,
      "mean_token_accuracy": 0.12599567249417304,
      "num_tokens": 2413550.0,
      "step": 1125
    },
    {
      "entropy": 6.317190837860108,
      "epoch": 0.06671783668890595,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004979501191000262,
      "loss": 6.2299,
      "mean_token_accuracy": 0.12873752564191818,
      "num_tokens": 2423614.0,
      "step": 1130
    },
    {
      "entropy": 6.164299440383911,
      "epoch": 0.06701304835567101,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004977883999115311,
      "loss": 6.2196,
      "mean_token_accuracy": 0.12582850828766823,
      "num_tokens": 2434825.0,
      "step": 1135
    },
    {
      "entropy": 6.226464891433716,
      "epoch": 0.06730826002243609,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004976205728665113,
      "loss": 6.2104,
      "mean_token_accuracy": 0.1259925462305546,
      "num_tokens": 2447062.0,
      "step": 1140
    },
    {
      "entropy": 6.243313789367676,
      "epoch": 0.06760347168920115,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004974466425660307,
      "loss": 6.2953,
      "mean_token_accuracy": 0.11829956993460655,
      "num_tokens": 2458795.0,
      "step": 1145
    },
    {
      "entropy": 6.175207901000976,
      "epoch": 0.06789868335596623,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004972666137784759,
      "loss": 6.2034,
      "mean_token_accuracy": 0.11987875550985336,
      "num_tokens": 2468563.0,
      "step": 1150
    },
    {
      "entropy": 6.222134208679199,
      "epoch": 0.0681938950227313,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004970804914394271,
      "loss": 6.2674,
      "mean_token_accuracy": 0.12158862352371216,
      "num_tokens": 2479407.0,
      "step": 1155
    },
    {
      "entropy": 6.2589446067810055,
      "epoch": 0.06848910668949637,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004968882806515225,
      "loss": 6.1892,
      "mean_token_accuracy": 0.11877380311489105,
      "num_tokens": 2491790.0,
      "step": 1160
    },
    {
      "entropy": 6.106157493591309,
      "epoch": 0.06878431835626143,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004966899866843177,
      "loss": 6.1038,
      "mean_token_accuracy": 0.13081091716885568,
      "num_tokens": 2501006.0,
      "step": 1165
    },
    {
      "entropy": 6.112197923660278,
      "epoch": 0.06907953002302651,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000496485614974142,
      "loss": 6.0207,
      "mean_token_accuracy": 0.14026050567626952,
      "num_tokens": 2510831.0,
      "step": 1170
    },
    {
      "entropy": 6.15854549407959,
      "epoch": 0.06937474168979157,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004962751711239492,
      "loss": 6.2307,
      "mean_token_accuracy": 0.1229746699333191,
      "num_tokens": 2522270.0,
      "step": 1175
    },
    {
      "entropy": 6.273019552230835,
      "epoch": 0.06966995335655665,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004960586609031636,
      "loss": 6.2534,
      "mean_token_accuracy": 0.13002387955784797,
      "num_tokens": 2532997.0,
      "step": 1180
    },
    {
      "entropy": 6.171327447891235,
      "epoch": 0.06996516502332172,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004958360902475224,
      "loss": 6.2381,
      "mean_token_accuracy": 0.12486061081290245,
      "num_tokens": 2543717.0,
      "step": 1185
    },
    {
      "entropy": 6.239405155181885,
      "epoch": 0.07026037669008679,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004956074652589125,
      "loss": 6.1521,
      "mean_token_accuracy": 0.12982202172279358,
      "num_tokens": 2554538.0,
      "step": 1190
    },
    {
      "entropy": 6.090667629241944,
      "epoch": 0.07055558835685186,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004953727922052035,
      "loss": 6.1565,
      "mean_token_accuracy": 0.12723496705293655,
      "num_tokens": 2564891.0,
      "step": 1195
    },
    {
      "entropy": 6.208614540100098,
      "epoch": 0.07085080002361693,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004951320775200756,
      "loss": 6.1242,
      "mean_token_accuracy": 0.1281391814351082,
      "num_tokens": 2575883.0,
      "step": 1200
    },
    {
      "entropy": 6.09900484085083,
      "epoch": 0.07114601169038201,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004948853278028436,
      "loss": 6.1945,
      "mean_token_accuracy": 0.12702925950288774,
      "num_tokens": 2588125.0,
      "step": 1205
    },
    {
      "entropy": 6.176623964309693,
      "epoch": 0.07144122335714707,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004946325498182755,
      "loss": 6.1666,
      "mean_token_accuracy": 0.12884261831641197,
      "num_tokens": 2598389.0,
      "step": 1210
    },
    {
      "entropy": 6.238792133331299,
      "epoch": 0.07173643502391215,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004943737504964076,
      "loss": 6.2851,
      "mean_token_accuracy": 0.12441871240735054,
      "num_tokens": 2609797.0,
      "step": 1215
    },
    {
      "entropy": 6.220934152603149,
      "epoch": 0.07203164669067721,
      "grad_norm": 1.0625,
      "learning_rate": 0.000494108936932354,
      "loss": 6.2374,
      "mean_token_accuracy": 0.1285350039601326,
      "num_tokens": 2622106.0,
      "step": 1220
    },
    {
      "entropy": 6.068195247650147,
      "epoch": 0.07232685835744229,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004938381163861124,
      "loss": 5.9866,
      "mean_token_accuracy": 0.13788015022873878,
      "num_tokens": 2632278.0,
      "step": 1225
    },
    {
      "entropy": 6.202727603912353,
      "epoch": 0.07262207002420736,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004935612962823645,
      "loss": 6.1787,
      "mean_token_accuracy": 0.12711300998926162,
      "num_tokens": 2642145.0,
      "step": 1230
    },
    {
      "entropy": 6.107152271270752,
      "epoch": 0.07291728169097243,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004932784842102739,
      "loss": 6.1096,
      "mean_token_accuracy": 0.13203658685088157,
      "num_tokens": 2653066.0,
      "step": 1235
    },
    {
      "entropy": 6.20755410194397,
      "epoch": 0.0732124933577375,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004929896879232758,
      "loss": 6.2229,
      "mean_token_accuracy": 0.13365396112203598,
      "num_tokens": 2662617.0,
      "step": 1240
    },
    {
      "entropy": 6.1612598419189455,
      "epoch": 0.07350770502450257,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004926949153388668,
      "loss": 6.0693,
      "mean_token_accuracy": 0.13576316982507705,
      "num_tokens": 2672968.0,
      "step": 1245
    },
    {
      "entropy": 6.090233898162841,
      "epoch": 0.07380291669126764,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004923941745383859,
      "loss": 6.1326,
      "mean_token_accuracy": 0.12741539999842644,
      "num_tokens": 2683021.0,
      "step": 1250
    },
    {
      "entropy": 6.147735118865967,
      "epoch": 0.07409812835803271,
      "grad_norm": 1.015625,
      "learning_rate": 0.000492087473766794,
      "loss": 6.1772,
      "mean_token_accuracy": 0.12292596101760864,
      "num_tokens": 2692942.0,
      "step": 1255
    },
    {
      "entropy": 6.074502658843994,
      "epoch": 0.07439334002479778,
      "grad_norm": 1.109375,
      "learning_rate": 0.000491774821432448,
      "loss": 6.0631,
      "mean_token_accuracy": 0.1280621863901615,
      "num_tokens": 2703208.0,
      "step": 1260
    },
    {
      "entropy": 6.130913639068604,
      "epoch": 0.07468855169156285,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004914562261068693,
      "loss": 6.1887,
      "mean_token_accuracy": 0.1308736138045788,
      "num_tokens": 2714535.0,
      "step": 1265
    },
    {
      "entropy": 6.153866815567016,
      "epoch": 0.07498376335832792,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004911316965245098,
      "loss": 6.1732,
      "mean_token_accuracy": 0.12706683427095414,
      "num_tokens": 2725646.0,
      "step": 1270
    },
    {
      "entropy": 6.082538604736328,
      "epoch": 0.075278975025093,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000490801241582512,
      "loss": 6.0147,
      "mean_token_accuracy": 0.14108175337314605,
      "num_tokens": 2734884.0,
      "step": 1275
    },
    {
      "entropy": 6.047293281555175,
      "epoch": 0.07557418669185806,
      "grad_norm": 0.984375,
      "learning_rate": 0.000490464870340465,
      "loss": 6.1083,
      "mean_token_accuracy": 0.12774249389767647,
      "num_tokens": 2745254.0,
      "step": 1280
    },
    {
      "entropy": 6.279677200317383,
      "epoch": 0.07586939835862314,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004901225920201563,
      "loss": 6.2694,
      "mean_token_accuracy": 0.1227074109017849,
      "num_tokens": 2756299.0,
      "step": 1285
    },
    {
      "entropy": 6.141456747055054,
      "epoch": 0.0761646100253882,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000489774416005319,
      "loss": 6.092,
      "mean_token_accuracy": 0.1252087853848934,
      "num_tokens": 2766796.0,
      "step": 1290
    },
    {
      "entropy": 6.08054838180542,
      "epoch": 0.07645982169215328,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004894203518413742,
      "loss": 6.1348,
      "mean_token_accuracy": 0.1298709914088249,
      "num_tokens": 2778542.0,
      "step": 1295
    },
    {
      "entropy": 6.206808567047119,
      "epoch": 0.07675503335891834,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004890604092351701,
      "loss": 6.1068,
      "mean_token_accuracy": 0.1335411436855793,
      "num_tokens": 2789065.0,
      "step": 1300
    },
    {
      "entropy": 6.104327201843262,
      "epoch": 0.07705024502568342,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000488694598054715,
      "loss": 6.1255,
      "mean_token_accuracy": 0.1259346254169941,
      "num_tokens": 2800217.0,
      "step": 1305
    },
    {
      "entropy": 6.1008704662322994,
      "epoch": 0.07734545669244848,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004883229283289071,
      "loss": 6.1793,
      "mean_token_accuracy": 0.1286480575799942,
      "num_tokens": 2810770.0,
      "step": 1310
    },
    {
      "entropy": 6.196523761749267,
      "epoch": 0.07764066835921356,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048794541024725993,
      "loss": 6.1347,
      "mean_token_accuracy": 0.13080908134579658,
      "num_tokens": 2821912.0,
      "step": 1315
    },
    {
      "entropy": 6.083820390701294,
      "epoch": 0.07793588002597862,
      "grad_norm": 1.125,
      "learning_rate": 0.0004875620541596221,
      "loss": 6.1511,
      "mean_token_accuracy": 0.13275840133428574,
      "num_tokens": 2832023.0,
      "step": 1320
    },
    {
      "entropy": 6.151384973526001,
      "epoch": 0.0782310916927437,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00048717287057589454,
      "loss": 6.0937,
      "mean_token_accuracy": 0.1350390985608101,
      "num_tokens": 2843568.0,
      "step": 1325
    },
    {
      "entropy": 6.052801275253296,
      "epoch": 0.07852630335950876,
      "grad_norm": 1.0,
      "learning_rate": 0.0004867778701657417,
      "loss": 6.0963,
      "mean_token_accuracy": 0.12285801768302917,
      "num_tokens": 2854135.0,
      "step": 1330
    },
    {
      "entropy": 6.16489896774292,
      "epoch": 0.07882151502627384,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048637706375829955,
      "loss": 6.1121,
      "mean_token_accuracy": 0.1272979386150837,
      "num_tokens": 2865359.0,
      "step": 1335
    },
    {
      "entropy": 6.000171232223511,
      "epoch": 0.0791167266930389,
      "grad_norm": 1.03125,
      "learning_rate": 0.000485970462341878,
      "loss": 6.0949,
      "mean_token_accuracy": 0.12831125780940056,
      "num_tokens": 2876417.0,
      "step": 1340
    },
    {
      "entropy": 6.137427186965942,
      "epoch": 0.07941193835980398,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048555807706366044,
      "loss": 6.1143,
      "mean_token_accuracy": 0.13167017474770545,
      "num_tokens": 2886893.0,
      "step": 1345
    },
    {
      "entropy": 6.134990930557251,
      "epoch": 0.07970715002656906,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048513991922939756,
      "loss": 6.1247,
      "mean_token_accuracy": 0.12959954217076303,
      "num_tokens": 2897020.0,
      "step": 1350
    },
    {
      "entropy": 6.113159084320069,
      "epoch": 0.08000236169333412,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00048471600030309744,
      "loss": 6.0958,
      "mean_token_accuracy": 0.12387151196599007,
      "num_tokens": 2907587.0,
      "step": 1355
    },
    {
      "entropy": 6.174454307556152,
      "epoch": 0.0802975733600992,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048428633190671186,
      "loss": 6.1923,
      "mean_token_accuracy": 0.12944753468036652,
      "num_tokens": 2918571.0,
      "step": 1360
    },
    {
      "entropy": 6.268069934844971,
      "epoch": 0.08059278502686426,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004838509258198167,
      "loss": 6.1705,
      "mean_token_accuracy": 0.12942629233002662,
      "num_tokens": 2930114.0,
      "step": 1365
    },
    {
      "entropy": 5.925813770294189,
      "epoch": 0.08088799669362934,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048340979397929,
      "loss": 6.0308,
      "mean_token_accuracy": 0.1300771601498127,
      "num_tokens": 2941553.0,
      "step": 1370
    },
    {
      "entropy": 6.137379360198975,
      "epoch": 0.0811832083603944,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048296294847898386,
      "loss": 6.1542,
      "mean_token_accuracy": 0.12957529947161675,
      "num_tokens": 2952973.0,
      "step": 1375
    },
    {
      "entropy": 6.083988285064697,
      "epoch": 0.08147842002715948,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004825104015693934,
      "loss": 6.0894,
      "mean_token_accuracy": 0.12802857980132104,
      "num_tokens": 2963351.0,
      "step": 1380
    },
    {
      "entropy": 6.117639541625977,
      "epoch": 0.08177363169392454,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004820521656573208,
      "loss": 6.1069,
      "mean_token_accuracy": 0.12783970609307288,
      "num_tokens": 2975456.0,
      "step": 1385
    },
    {
      "entropy": 6.201921701431274,
      "epoch": 0.08206884336068962,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048158825330553505,
      "loss": 6.1909,
      "mean_token_accuracy": 0.12405910715460777,
      "num_tokens": 2986454.0,
      "step": 1390
    },
    {
      "entropy": 6.006986284255982,
      "epoch": 0.08236405502745468,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048111867723242763,
      "loss": 6.0643,
      "mean_token_accuracy": 0.12654304653406143,
      "num_tokens": 2997893.0,
      "step": 1395
    },
    {
      "entropy": 6.1468892097473145,
      "epoch": 0.08265926669421976,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004806434503116637,
      "loss": 5.9938,
      "mean_token_accuracy": 0.13587316051125525,
      "num_tokens": 3008866.0,
      "step": 1400
    },
    {
      "entropy": 6.054237937927246,
      "epoch": 0.08295447836098482,
      "grad_norm": 1.0,
      "learning_rate": 0.0004801625855718296,
      "loss": 6.1694,
      "mean_token_accuracy": 0.1326708622276783,
      "num_tokens": 3020017.0,
      "step": 1405
    },
    {
      "entropy": 6.010798788070678,
      "epoch": 0.0832496900277499,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047967609619607477,
      "loss": 5.9746,
      "mean_token_accuracy": 0.13155399784445762,
      "num_tokens": 3030707.0,
      "step": 1410
    },
    {
      "entropy": 6.1784576892852785,
      "epoch": 0.08354490169451496,
      "grad_norm": 1.125,
      "learning_rate": 0.0004791839955217513,
      "loss": 6.0924,
      "mean_token_accuracy": 0.1290546603500843,
      "num_tokens": 3041305.0,
      "step": 1415
    },
    {
      "entropy": 6.029217767715454,
      "epoch": 0.08384011336128004,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047868629704004786,
      "loss": 6.1692,
      "mean_token_accuracy": 0.1270539216697216,
      "num_tokens": 3052952.0,
      "step": 1420
    },
    {
      "entropy": 6.17234206199646,
      "epoch": 0.0841353250280451,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047818301439561965,
      "loss": 6.0294,
      "mean_token_accuracy": 0.13541628271341324,
      "num_tokens": 3063604.0,
      "step": 1425
    },
    {
      "entropy": 5.994857454299927,
      "epoch": 0.08443053669481018,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047767416138621454,
      "loss": 6.106,
      "mean_token_accuracy": 0.13280996158719063,
      "num_tokens": 3074199.0,
      "step": 1430
    },
    {
      "entropy": 6.197816705703735,
      "epoch": 0.08472574836157525,
      "grad_norm": 1.09375,
      "learning_rate": 0.000477159751962295,
      "loss": 6.0778,
      "mean_token_accuracy": 0.1278144374489784,
      "num_tokens": 3084811.0,
      "step": 1435
    },
    {
      "entropy": 6.100957155227661,
      "epoch": 0.08502096002834032,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047663980022665507,
      "loss": 6.1508,
      "mean_token_accuracy": 0.13362270593643188,
      "num_tokens": 3095934.0,
      "step": 1440
    },
    {
      "entropy": 6.144081687927246,
      "epoch": 0.08531617169510539,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047611432043403437,
      "loss": 6.0093,
      "mean_token_accuracy": 0.1304073341190815,
      "num_tokens": 3106469.0,
      "step": 1445
    },
    {
      "entropy": 6.062932109832763,
      "epoch": 0.08561138336187046,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004755833269907267,
      "loss": 6.2009,
      "mean_token_accuracy": 0.1269908919930458,
      "num_tokens": 3117860.0,
      "step": 1450
    },
    {
      "entropy": 6.200025177001953,
      "epoch": 0.08590659502863553,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004750468344541857,
      "loss": 6.1303,
      "mean_token_accuracy": 0.1287964791059494,
      "num_tokens": 3127983.0,
      "step": 1455
    },
    {
      "entropy": 6.026769542694092,
      "epoch": 0.0862018066954006,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047450485753262525,
      "loss": 6.0796,
      "mean_token_accuracy": 0.13652199730277062,
      "num_tokens": 3138705.0,
      "step": 1460
    },
    {
      "entropy": 6.097449016571045,
      "epoch": 0.08649701836216567,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00047395741108461633,
      "loss": 6.0427,
      "mean_token_accuracy": 0.13442591875791549,
      "num_tokens": 3150079.0,
      "step": 1465
    },
    {
      "entropy": 5.980276203155517,
      "epoch": 0.08679223002893074,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00047340451011867985,
      "loss": 6.0421,
      "mean_token_accuracy": 0.13306428194046022,
      "num_tokens": 3160261.0,
      "step": 1470
    },
    {
      "entropy": 5.978704452514648,
      "epoch": 0.08708744169569581,
      "grad_norm": 1.0,
      "learning_rate": 0.00047284616979287515,
      "loss": 5.9897,
      "mean_token_accuracy": 0.1374903328716755,
      "num_tokens": 3169779.0,
      "step": 1475
    },
    {
      "entropy": 6.020472669601441,
      "epoch": 0.08738265336246089,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047228240541438433,
      "loss": 6.0162,
      "mean_token_accuracy": 0.13789537474513053,
      "num_tokens": 3180400.0,
      "step": 1480
    },
    {
      "entropy": 6.140784788131714,
      "epoch": 0.08767786502922595,
      "grad_norm": 0.984375,
      "learning_rate": 0.00047171323243909257,
      "loss": 6.0684,
      "mean_token_accuracy": 0.13254758790135385,
      "num_tokens": 3191546.0,
      "step": 1485
    },
    {
      "entropy": 6.07828917503357,
      "epoch": 0.08797307669599103,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047113866647116457,
      "loss": 6.0377,
      "mean_token_accuracy": 0.13490098640322684,
      "num_tokens": 3202222.0,
      "step": 1490
    },
    {
      "entropy": 5.974679851531983,
      "epoch": 0.08826828836275609,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004705587232626164,
      "loss": 6.0031,
      "mean_token_accuracy": 0.13679205402731895,
      "num_tokens": 3212973.0,
      "step": 1495
    },
    {
      "entropy": 6.074326848983764,
      "epoch": 0.08856350002952117,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046997341871288424,
      "loss": 6.0159,
      "mean_token_accuracy": 0.13410410583019255,
      "num_tokens": 3224661.0,
      "step": 1500
    },
    {
      "entropy": 6.04751615524292,
      "epoch": 0.08885871169628624,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004693827688683879,
      "loss": 6.1216,
      "mean_token_accuracy": 0.13102777674794197,
      "num_tokens": 3235629.0,
      "step": 1505
    },
    {
      "entropy": 6.056356430053711,
      "epoch": 0.08915392336305131,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004687867899220914,
      "loss": 6.0095,
      "mean_token_accuracy": 0.1373201921582222,
      "num_tokens": 3246016.0,
      "step": 1510
    },
    {
      "entropy": 5.955541944503784,
      "epoch": 0.08944913502981638,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00046818549821305846,
      "loss": 5.9698,
      "mean_token_accuracy": 0.1393904373049736,
      "num_tokens": 3255875.0,
      "step": 1515
    },
    {
      "entropy": 6.11029281616211,
      "epoch": 0.08974434669658145,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046757891022600494,
      "loss": 6.0744,
      "mean_token_accuracy": 0.13373544216156005,
      "num_tokens": 3266783.0,
      "step": 1520
    },
    {
      "entropy": 6.076243829727173,
      "epoch": 0.09003955836334653,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004669670425908471,
      "loss": 6.0769,
      "mean_token_accuracy": 0.13891345784068107,
      "num_tokens": 3277949.0,
      "step": 1525
    },
    {
      "entropy": 6.02083101272583,
      "epoch": 0.09033477003011159,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004663499120822451,
      "loss": 6.0448,
      "mean_token_accuracy": 0.13696586415171624,
      "num_tokens": 3288225.0,
      "step": 1530
    },
    {
      "entropy": 5.9968249797821045,
      "epoch": 0.09062998169687667,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004657275356191437,
      "loss": 5.9192,
      "mean_token_accuracy": 0.1334097608923912,
      "num_tokens": 3298877.0,
      "step": 1535
    },
    {
      "entropy": 5.955824184417724,
      "epoch": 0.09092519336364173,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046509993026430804,
      "loss": 5.9731,
      "mean_token_accuracy": 0.13175629526376725,
      "num_tokens": 3308094.0,
      "step": 1540
    },
    {
      "entropy": 6.100501203536988,
      "epoch": 0.0912204050304068,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004644671132238558,
      "loss": 6.0481,
      "mean_token_accuracy": 0.1411367565393448,
      "num_tokens": 3319367.0,
      "step": 1545
    },
    {
      "entropy": 6.0393892288208,
      "epoch": 0.09151561669717187,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046382910184678585,
      "loss": 5.951,
      "mean_token_accuracy": 0.14005291908979417,
      "num_tokens": 3328807.0,
      "step": 1550
    },
    {
      "entropy": 5.946415185928345,
      "epoch": 0.09181082836393695,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004631859136245025,
      "loss": 6.0224,
      "mean_token_accuracy": 0.13421571403741836,
      "num_tokens": 3339377.0,
      "step": 1555
    },
    {
      "entropy": 6.116823434829712,
      "epoch": 0.09210604003070201,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004625375661903357,
      "loss": 6.068,
      "mean_token_accuracy": 0.13405861109495162,
      "num_tokens": 3349555.0,
      "step": 1560
    },
    {
      "entropy": 6.04711422920227,
      "epoch": 0.09240125169746709,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046188407731905787,
      "loss": 6.0327,
      "mean_token_accuracy": 0.1323117010295391,
      "num_tokens": 3360187.0,
      "step": 1565
    },
    {
      "entropy": 6.006165599822998,
      "epoch": 0.09269646336423215,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046122546492639643,
      "loss": 5.9497,
      "mean_token_accuracy": 0.1417209431529045,
      "num_tokens": 3370702.0,
      "step": 1570
    },
    {
      "entropy": 6.042462253570557,
      "epoch": 0.09299167503099723,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000460561747068543,
      "loss": 6.0761,
      "mean_token_accuracy": 0.12965757995843888,
      "num_tokens": 3381944.0,
      "step": 1575
    },
    {
      "entropy": 6.071602487564087,
      "epoch": 0.09328688669776229,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004598929419416578,
      "loss": 5.9718,
      "mean_token_accuracy": 0.13460363522171975,
      "num_tokens": 3392376.0,
      "step": 1580
    },
    {
      "entropy": 5.9583460807800295,
      "epoch": 0.09358209836452737,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045921906788137123,
      "loss": 5.9224,
      "mean_token_accuracy": 0.1380501091480255,
      "num_tokens": 3403048.0,
      "step": 1585
    },
    {
      "entropy": 5.977456855773926,
      "epoch": 0.09387731003129243,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00045854014336228115,
      "loss": 6.0307,
      "mean_token_accuracy": 0.12758268117904664,
      "num_tokens": 3413553.0,
      "step": 1590
    },
    {
      "entropy": 6.032625532150268,
      "epoch": 0.09417252169805751,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00045785618699744615,
      "loss": 5.9238,
      "mean_token_accuracy": 0.13623402193188666,
      "num_tokens": 3424295.0,
      "step": 1595
    },
    {
      "entropy": 5.985020351409912,
      "epoch": 0.09446773336482257,
      "grad_norm": 1.0,
      "learning_rate": 0.00045716721753787543,
      "loss": 5.9476,
      "mean_token_accuracy": 0.13396428748965264,
      "num_tokens": 3434449.0,
      "step": 1600
    },
    {
      "entropy": 5.971049022674561,
      "epoch": 0.09476294503158765,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004564732538720148,
      "loss": 5.9766,
      "mean_token_accuracy": 0.14037723690271378,
      "num_tokens": 3444601.0,
      "step": 1605
    },
    {
      "entropy": 6.03154559135437,
      "epoch": 0.09505815669835271,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045577431502522877,
      "loss": 5.956,
      "mean_token_accuracy": 0.13928258866071702,
      "num_tokens": 3455052.0,
      "step": 1610
    },
    {
      "entropy": 6.047151565551758,
      "epoch": 0.09535336836511779,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004550704201592787,
      "loss": 5.9694,
      "mean_token_accuracy": 0.13797362372279168,
      "num_tokens": 3466129.0,
      "step": 1615
    },
    {
      "entropy": 6.001207876205444,
      "epoch": 0.09564858003188285,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004543615885717981,
      "loss": 6.0408,
      "mean_token_accuracy": 0.13441915214061737,
      "num_tokens": 3476225.0,
      "step": 1620
    },
    {
      "entropy": 5.993210792541504,
      "epoch": 0.09594379169864793,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045364783969576296,
      "loss": 5.9967,
      "mean_token_accuracy": 0.13255249112844467,
      "num_tokens": 3487140.0,
      "step": 1625
    },
    {
      "entropy": 5.97896409034729,
      "epoch": 0.096239003365413,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004529291930989592,
      "loss": 5.9137,
      "mean_token_accuracy": 0.13355278670787812,
      "num_tokens": 3497725.0,
      "step": 1630
    },
    {
      "entropy": 5.922145891189575,
      "epoch": 0.09653421503217807,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004522056684834464,
      "loss": 6.0266,
      "mean_token_accuracy": 0.12917877361178398,
      "num_tokens": 3509360.0,
      "step": 1635
    },
    {
      "entropy": 6.056468343734741,
      "epoch": 0.09682942669894314,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004514772856850173,
      "loss": 5.96,
      "mean_token_accuracy": 0.1380475141108036,
      "num_tokens": 3519919.0,
      "step": 1640
    },
    {
      "entropy": 6.010870742797851,
      "epoch": 0.09712463836570821,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004507440646726542,
      "loss": 6.0126,
      "mean_token_accuracy": 0.13415253460407256,
      "num_tokens": 3530277.0,
      "step": 1645
    },
    {
      "entropy": 5.949351453781128,
      "epoch": 0.09741985003247329,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004500060255479818,
      "loss": 5.9733,
      "mean_token_accuracy": 0.13921516239643097,
      "num_tokens": 3540641.0,
      "step": 1650
    },
    {
      "entropy": 6.150185585021973,
      "epoch": 0.09771506169923835,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004492631885447151,
      "loss": 6.0799,
      "mean_token_accuracy": 0.13410310819745064,
      "num_tokens": 3550069.0,
      "step": 1655
    },
    {
      "entropy": 5.998241806030274,
      "epoch": 0.09801027336600343,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044851557402810616,
      "loss": 5.9941,
      "mean_token_accuracy": 0.13666213899850846,
      "num_tokens": 3560980.0,
      "step": 1660
    },
    {
      "entropy": 6.000314521789551,
      "epoch": 0.0983054850327685,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044776320249438444,
      "loss": 5.9014,
      "mean_token_accuracy": 0.1379874236881733,
      "num_tokens": 3572352.0,
      "step": 1665
    },
    {
      "entropy": 5.864667844772339,
      "epoch": 0.09860069669953357,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00044700609457019565,
      "loss": 5.8057,
      "mean_token_accuracy": 0.14547686278820038,
      "num_tokens": 3582414.0,
      "step": 1670
    },
    {
      "entropy": 5.913028717041016,
      "epoch": 0.09889590836629863,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004462442710120359,
      "loss": 5.9267,
      "mean_token_accuracy": 0.1374290555715561,
      "num_tokens": 3593655.0,
      "step": 1675
    },
    {
      "entropy": 6.0694255352020265,
      "epoch": 0.09919112003306371,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000445477752705683,
      "loss": 5.976,
      "mean_token_accuracy": 0.14330636113882064,
      "num_tokens": 3603784.0,
      "step": 1680
    },
    {
      "entropy": 5.914968252182007,
      "epoch": 0.09948633169982878,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00044470656066562336,
      "loss": 5.9023,
      "mean_token_accuracy": 0.1387188106775284,
      "num_tokens": 3613955.0,
      "step": 1685
    },
    {
      "entropy": 5.958784723281861,
      "epoch": 0.09978154336659385,
      "grad_norm": 1.0,
      "learning_rate": 0.0004439307160344765,
      "loss": 5.9596,
      "mean_token_accuracy": 0.13545051887631415,
      "num_tokens": 3624421.0,
      "step": 1690
    },
    {
      "entropy": 5.906978559494019,
      "epoch": 0.10007675503335892,
      "grad_norm": 1.125,
      "learning_rate": 0.00044315024008241473,
      "loss": 5.9132,
      "mean_token_accuracy": 0.14482396841049194,
      "num_tokens": 3633782.0,
      "step": 1695
    },
    {
      "entropy": 5.976182460784912,
      "epoch": 0.100371966700124,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004423651542065806,
      "loss": 5.9933,
      "mean_token_accuracy": 0.13364353850483895,
      "num_tokens": 3644560.0,
      "step": 1700
    },
    {
      "entropy": 5.938481664657592,
      "epoch": 0.10066717836688906,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00044157547993050006,
      "loss": 5.9719,
      "mean_token_accuracy": 0.1391169995069504,
      "num_tokens": 3655434.0,
      "step": 1705
    },
    {
      "entropy": 6.016226863861084,
      "epoch": 0.10096239003365413,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00044078123890349227,
      "loss": 5.8969,
      "mean_token_accuracy": 0.14098571836948395,
      "num_tokens": 3665628.0,
      "step": 1710
    },
    {
      "entropy": 5.949463367462158,
      "epoch": 0.1012576017004192,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043998245290007606,
      "loss": 5.9951,
      "mean_token_accuracy": 0.13202026411890982,
      "num_tokens": 3676483.0,
      "step": 1715
    },
    {
      "entropy": 6.087437105178833,
      "epoch": 0.10155281336718427,
      "grad_norm": 1.015625,
      "learning_rate": 0.00043917914381937323,
      "loss": 5.9868,
      "mean_token_accuracy": 0.13543638065457345,
      "num_tokens": 3687299.0,
      "step": 1720
    },
    {
      "entropy": 5.952702236175537,
      "epoch": 0.10184802503394934,
      "grad_norm": 1.015625,
      "learning_rate": 0.00043837133368450815,
      "loss": 5.9212,
      "mean_token_accuracy": 0.1348453126847744,
      "num_tokens": 3698068.0,
      "step": 1725
    },
    {
      "entropy": 6.064343881607056,
      "epoch": 0.10214323670071442,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004375590446420037,
      "loss": 5.976,
      "mean_token_accuracy": 0.1393336459994316,
      "num_tokens": 3708004.0,
      "step": 1730
    },
    {
      "entropy": 5.993683958053589,
      "epoch": 0.10243844836747948,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004367422989611743,
      "loss": 6.0219,
      "mean_token_accuracy": 0.13565473407506942,
      "num_tokens": 3718845.0,
      "step": 1735
    },
    {
      "entropy": 5.993088483810425,
      "epoch": 0.10273366003424456,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004359211190335153,
      "loss": 5.8671,
      "mean_token_accuracy": 0.143757364153862,
      "num_tokens": 3729532.0,
      "step": 1740
    },
    {
      "entropy": 5.85788049697876,
      "epoch": 0.10302887170100962,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043509552737208923,
      "loss": 5.9378,
      "mean_token_accuracy": 0.13778024390339852,
      "num_tokens": 3740596.0,
      "step": 1745
    },
    {
      "entropy": 6.01426477432251,
      "epoch": 0.1033240833677747,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043426554661090853,
      "loss": 5.9339,
      "mean_token_accuracy": 0.1473975345492363,
      "num_tokens": 3751144.0,
      "step": 1750
    },
    {
      "entropy": 5.970757246017456,
      "epoch": 0.10361929503453976,
      "grad_norm": 0.984375,
      "learning_rate": 0.00043343119950431516,
      "loss": 5.9217,
      "mean_token_accuracy": 0.14652103036642075,
      "num_tokens": 3762243.0,
      "step": 1755
    },
    {
      "entropy": 5.988554239273071,
      "epoch": 0.10391450670130484,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043259250892635644,
      "loss": 5.9527,
      "mean_token_accuracy": 0.13955445364117622,
      "num_tokens": 3772099.0,
      "step": 1760
    },
    {
      "entropy": 5.956176042556763,
      "epoch": 0.1042097183680699,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004317494978701582,
      "loss": 5.9312,
      "mean_token_accuracy": 0.1413627102971077,
      "num_tokens": 3782670.0,
      "step": 1765
    },
    {
      "entropy": 5.960332298278809,
      "epoch": 0.10450493003483498,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004309021894472943,
      "loss": 5.9411,
      "mean_token_accuracy": 0.14020357206463813,
      "num_tokens": 3792569.0,
      "step": 1770
    },
    {
      "entropy": 5.945288133621216,
      "epoch": 0.10480014170160004,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004300506068871534,
      "loss": 5.8895,
      "mean_token_accuracy": 0.14070242494344712,
      "num_tokens": 3802954.0,
      "step": 1775
    },
    {
      "entropy": 5.972039985656738,
      "epoch": 0.10509535336836512,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042919477353630135,
      "loss": 5.9133,
      "mean_token_accuracy": 0.13573219552636145,
      "num_tokens": 3813529.0,
      "step": 1780
    },
    {
      "entropy": 5.870369863510132,
      "epoch": 0.10539056503513018,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000428334712857842,
      "loss": 5.9186,
      "mean_token_accuracy": 0.1356649711728096,
      "num_tokens": 3823603.0,
      "step": 1785
    },
    {
      "entropy": 5.912766456604004,
      "epoch": 0.10568577670189526,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042747044843077304,
      "loss": 5.8746,
      "mean_token_accuracy": 0.14842769429087638,
      "num_tokens": 3833831.0,
      "step": 1790
    },
    {
      "entropy": 5.970458984375,
      "epoch": 0.10598098836866032,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042660200394934047,
      "loss": 5.8865,
      "mean_token_accuracy": 0.1426880329847336,
      "num_tokens": 3844282.0,
      "step": 1795
    },
    {
      "entropy": 5.995832252502441,
      "epoch": 0.1062762000354254,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042572940322238844,
      "loss": 5.9336,
      "mean_token_accuracy": 0.13592784032225608,
      "num_tokens": 3856018.0,
      "step": 1800
    },
    {
      "entropy": 6.016539716720581,
      "epoch": 0.10657141170219048,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042485267017270664,
      "loss": 5.9671,
      "mean_token_accuracy": 0.13218472227454187,
      "num_tokens": 3867545.0,
      "step": 1805
    },
    {
      "entropy": 5.93374195098877,
      "epoch": 0.10686662336895554,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004239718288363745,
      "loss": 5.8434,
      "mean_token_accuracy": 0.1460932120680809,
      "num_tokens": 3878215.0,
      "step": 1810
    },
    {
      "entropy": 5.9325737953186035,
      "epoch": 0.10716183503572062,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004230869033621023,
      "loss": 5.9142,
      "mean_token_accuracy": 0.13958162665367127,
      "num_tokens": 3888172.0,
      "step": 1815
    },
    {
      "entropy": 5.992139291763306,
      "epoch": 0.10745704670248568,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004221979180105688,
      "loss": 5.991,
      "mean_token_accuracy": 0.13233617097139358,
      "num_tokens": 3899482.0,
      "step": 1820
    },
    {
      "entropy": 6.027761936187744,
      "epoch": 0.10775225836925076,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00042130489715375645,
      "loss": 5.9591,
      "mean_token_accuracy": 0.13765794709324836,
      "num_tokens": 3909464.0,
      "step": 1825
    },
    {
      "entropy": 5.881091165542602,
      "epoch": 0.10804747003601582,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00042040786527428335,
      "loss": 5.8755,
      "mean_token_accuracy": 0.1399849072098732,
      "num_tokens": 3919656.0,
      "step": 1830
    },
    {
      "entropy": 5.987238645553589,
      "epoch": 0.1083426817027809,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004195068469647315,
      "loss": 5.935,
      "mean_token_accuracy": 0.1402907982468605,
      "num_tokens": 3930420.0,
      "step": 1835
    },
    {
      "entropy": 5.9825396060943605,
      "epoch": 0.10863789336954596,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041860186692697297,
      "loss": 5.9339,
      "mean_token_accuracy": 0.1349881947040558,
      "num_tokens": 3941670.0,
      "step": 1840
    },
    {
      "entropy": 5.956534337997437,
      "epoch": 0.10893310503631104,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041769294997149264,
      "loss": 5.9098,
      "mean_token_accuracy": 0.1411547601222992,
      "num_tokens": 3952550.0,
      "step": 1845
    },
    {
      "entropy": 5.965771055221557,
      "epoch": 0.1092283167030761,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004167801210167081,
      "loss": 5.9172,
      "mean_token_accuracy": 0.1462905302643776,
      "num_tokens": 3963316.0,
      "step": 1850
    },
    {
      "entropy": 5.914458608627319,
      "epoch": 0.10952352836984118,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004158634050882861,
      "loss": 5.9136,
      "mean_token_accuracy": 0.13789980560541154,
      "num_tokens": 3974213.0,
      "step": 1855
    },
    {
      "entropy": 5.990167331695557,
      "epoch": 0.10981874003660624,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004149428273184569,
      "loss": 5.8843,
      "mean_token_accuracy": 0.14071305990219116,
      "num_tokens": 3985490.0,
      "step": 1860
    },
    {
      "entropy": 5.925612258911133,
      "epoch": 0.11011395170337132,
      "grad_norm": 1.125,
      "learning_rate": 0.0004140184129453253,
      "loss": 5.9623,
      "mean_token_accuracy": 0.13739006742835044,
      "num_tokens": 3996859.0,
      "step": 1865
    },
    {
      "entropy": 5.9599072456359865,
      "epoch": 0.11040916337013638,
      "grad_norm": 1.171875,
      "learning_rate": 0.000413090187312178,
      "loss": 5.9094,
      "mean_token_accuracy": 0.1436771422624588,
      "num_tokens": 4007177.0,
      "step": 1870
    },
    {
      "entropy": 5.916435241699219,
      "epoch": 0.11070437503690146,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004121581758667898,
      "loss": 5.841,
      "mean_token_accuracy": 0.13843855261802673,
      "num_tokens": 4017976.0,
      "step": 1875
    },
    {
      "entropy": 5.848360681533814,
      "epoch": 0.11099958670366653,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00041122240416072533,
      "loss": 5.8013,
      "mean_token_accuracy": 0.1479976147413254,
      "num_tokens": 4028925.0,
      "step": 1880
    },
    {
      "entropy": 5.91417384147644,
      "epoch": 0.1112947983704316,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004102828978486385,
      "loss": 5.8587,
      "mean_token_accuracy": 0.14406538307666777,
      "num_tokens": 4040458.0,
      "step": 1885
    },
    {
      "entropy": 5.935881662368774,
      "epoch": 0.11159001003719667,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004093396826875695,
      "loss": 5.9524,
      "mean_token_accuracy": 0.14116467982530595,
      "num_tokens": 4051787.0,
      "step": 1890
    },
    {
      "entropy": 5.996190929412842,
      "epoch": 0.11188522170396174,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00040839278453623837,
      "loss": 5.9463,
      "mean_token_accuracy": 0.14074928015470506,
      "num_tokens": 4062476.0,
      "step": 1895
    },
    {
      "entropy": 6.024869632720947,
      "epoch": 0.1121804333707268,
      "grad_norm": 1.0,
      "learning_rate": 0.0004074422293543363,
      "loss": 6.0117,
      "mean_token_accuracy": 0.13238218873739244,
      "num_tokens": 4073843.0,
      "step": 1900
    },
    {
      "entropy": 5.916808605194092,
      "epoch": 0.11247564503749188,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004064880432018137,
      "loss": 5.8341,
      "mean_token_accuracy": 0.14415794909000396,
      "num_tokens": 4084102.0,
      "step": 1905
    },
    {
      "entropy": 5.880737257003784,
      "epoch": 0.11277085670425695,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040553025223816615,
      "loss": 5.9033,
      "mean_token_accuracy": 0.14525620341300965,
      "num_tokens": 4094418.0,
      "step": 1910
    },
    {
      "entropy": 5.964789915084839,
      "epoch": 0.11306606837102202,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00040456888272171653,
      "loss": 5.885,
      "mean_token_accuracy": 0.14050159752368926,
      "num_tokens": 4105189.0,
      "step": 1915
    },
    {
      "entropy": 5.889213705062867,
      "epoch": 0.11336128003778709,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040360396100889577,
      "loss": 5.871,
      "mean_token_accuracy": 0.149219286441803,
      "num_tokens": 4116368.0,
      "step": 1920
    },
    {
      "entropy": 5.976749038696289,
      "epoch": 0.11365649170455216,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004026355135535202,
      "loss": 5.8981,
      "mean_token_accuracy": 0.13695148453116418,
      "num_tokens": 4126472.0,
      "step": 1925
    },
    {
      "entropy": 5.97646541595459,
      "epoch": 0.11395170337131723,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000401663566906066,
      "loss": 5.8212,
      "mean_token_accuracy": 0.14775698632001877,
      "num_tokens": 4137875.0,
      "step": 1930
    },
    {
      "entropy": 5.871037244796753,
      "epoch": 0.1142469150380823,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040068814771294134,
      "loss": 5.872,
      "mean_token_accuracy": 0.14493257254362107,
      "num_tokens": 4147555.0,
      "step": 1935
    },
    {
      "entropy": 5.897066640853882,
      "epoch": 0.11454212670484737,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0003997092827157562,
      "loss": 5.8789,
      "mean_token_accuracy": 0.14204722344875337,
      "num_tokens": 4159469.0,
      "step": 1940
    },
    {
      "entropy": 5.976615333557129,
      "epoch": 0.11483733837161245,
      "grad_norm": 1.125,
      "learning_rate": 0.000398726998750589,
      "loss": 5.8292,
      "mean_token_accuracy": 0.1409693591296673,
      "num_tokens": 4170049.0,
      "step": 1945
    },
    {
      "entropy": 5.792118501663208,
      "epoch": 0.11513255003837752,
      "grad_norm": 1.21875,
      "learning_rate": 0.00039774132274725076,
      "loss": 5.8178,
      "mean_token_accuracy": 0.14296735152602197,
      "num_tokens": 4180933.0,
      "step": 1950
    },
    {
      "entropy": 5.834021520614624,
      "epoch": 0.11542776170514259,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039675228172854707,
      "loss": 5.8922,
      "mean_token_accuracy": 0.1392677716910839,
      "num_tokens": 4191883.0,
      "step": 1955
    },
    {
      "entropy": 6.045830583572387,
      "epoch": 0.11572297337190766,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003957599028095371,
      "loss": 5.9612,
      "mean_token_accuracy": 0.1446747064590454,
      "num_tokens": 4202265.0,
      "step": 1960
    },
    {
      "entropy": 5.915384101867676,
      "epoch": 0.11601818503867273,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039476421319679017,
      "loss": 5.8879,
      "mean_token_accuracy": 0.13635574281215668,
      "num_tokens": 4212760.0,
      "step": 1965
    },
    {
      "entropy": 6.001104831695557,
      "epoch": 0.1163133967054378,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00039376524018764,
      "loss": 5.9018,
      "mean_token_accuracy": 0.14595609679818153,
      "num_tokens": 4222739.0,
      "step": 1970
    },
    {
      "entropy": 5.93096809387207,
      "epoch": 0.11660860837220287,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039276301116943616,
      "loss": 5.8194,
      "mean_token_accuracy": 0.1551318235695362,
      "num_tokens": 4232886.0,
      "step": 1975
    },
    {
      "entropy": 6.026076221466065,
      "epoch": 0.11690382003896795,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003917575536187936,
      "loss": 6.0242,
      "mean_token_accuracy": 0.13150568827986717,
      "num_tokens": 4243196.0,
      "step": 1980
    },
    {
      "entropy": 5.901855039596557,
      "epoch": 0.11719903170573301,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039074889510083894,
      "loss": 5.8645,
      "mean_token_accuracy": 0.14477338045835494,
      "num_tokens": 4254301.0,
      "step": 1985
    },
    {
      "entropy": 5.965902423858642,
      "epoch": 0.11749424337249809,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00038973706326845495,
      "loss": 5.9498,
      "mean_token_accuracy": 0.13250884115695954,
      "num_tokens": 4264233.0,
      "step": 1990
    },
    {
      "entropy": 5.929144954681396,
      "epoch": 0.11778945503926315,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003887220858615225,
      "loss": 5.8727,
      "mean_token_accuracy": 0.14213800951838493,
      "num_tokens": 4274865.0,
      "step": 1995
    },
    {
      "entropy": 5.892509651184082,
      "epoch": 0.11808466670602823,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003877039907061597,
      "loss": 5.833,
      "mean_token_accuracy": 0.14505684971809388,
      "num_tokens": 4285138.0,
      "step": 2000
    }
  ],
  "logging_steps": 5,
  "max_steps": 4000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "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": 6294018797568000.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}