{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.5442670537010159,
  "eval_steps": 3000,
  "global_step": 6000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 10.69201955795288,
      "epoch": 0.00045355587808418,
      "grad_norm": 14.125,
      "learning_rate": 2e-06,
      "loss": 10.8421,
      "mean_token_accuracy": 0.0,
      "num_tokens": 10827.0,
      "step": 5
    },
    {
      "entropy": 10.691895008087158,
      "epoch": 0.00090711175616836,
      "grad_norm": 13.0,
      "learning_rate": 4.5e-06,
      "loss": 10.7626,
      "mean_token_accuracy": 0.0,
      "num_tokens": 21025.0,
      "step": 10
    },
    {
      "entropy": 10.690971279144287,
      "epoch": 0.00136066763425254,
      "grad_norm": 9.875,
      "learning_rate": 7e-06,
      "loss": 10.5141,
      "mean_token_accuracy": 0.02440520448144525,
      "num_tokens": 31678.0,
      "step": 15
    },
    {
      "entropy": 10.678609657287598,
      "epoch": 0.00181422351233672,
      "grad_norm": 6.3125,
      "learning_rate": 9.5e-06,
      "loss": 10.1495,
      "mean_token_accuracy": 0.04602932706475258,
      "num_tokens": 41025.0,
      "step": 20
    },
    {
      "entropy": 10.615153217315674,
      "epoch": 0.0022677793904208998,
      "grad_norm": 4.25,
      "learning_rate": 1.2e-05,
      "loss": 9.9137,
      "mean_token_accuracy": 0.043584797717630866,
      "num_tokens": 51270.0,
      "step": 25
    },
    {
      "entropy": 10.595486927032471,
      "epoch": 0.00272133526850508,
      "grad_norm": 3.3125,
      "learning_rate": 1.4500000000000002e-05,
      "loss": 9.7488,
      "mean_token_accuracy": 0.04369584359228611,
      "num_tokens": 62142.0,
      "step": 30
    },
    {
      "entropy": 10.605365085601807,
      "epoch": 0.00317489114658926,
      "grad_norm": 2.890625,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 9.6708,
      "mean_token_accuracy": 0.0410692248493433,
      "num_tokens": 72709.0,
      "step": 35
    },
    {
      "entropy": 10.59789743423462,
      "epoch": 0.00362844702467344,
      "grad_norm": 2.609375,
      "learning_rate": 1.95e-05,
      "loss": 9.6037,
      "mean_token_accuracy": 0.042245643213391304,
      "num_tokens": 82313.0,
      "step": 40
    },
    {
      "entropy": 10.562069129943847,
      "epoch": 0.00408200290275762,
      "grad_norm": 2.859375,
      "learning_rate": 2.2e-05,
      "loss": 9.5375,
      "mean_token_accuracy": 0.047857620567083356,
      "num_tokens": 92450.0,
      "step": 45
    },
    {
      "entropy": 10.572824478149414,
      "epoch": 0.0045355587808417995,
      "grad_norm": 2.5,
      "learning_rate": 2.4500000000000003e-05,
      "loss": 9.5533,
      "mean_token_accuracy": 0.04238316640257835,
      "num_tokens": 103978.0,
      "step": 50
    },
    {
      "entropy": 10.59686508178711,
      "epoch": 0.00498911465892598,
      "grad_norm": 2.28125,
      "learning_rate": 2.7e-05,
      "loss": 9.451,
      "mean_token_accuracy": 0.04611069448292256,
      "num_tokens": 114134.0,
      "step": 55
    },
    {
      "entropy": 10.546150016784669,
      "epoch": 0.00544267053701016,
      "grad_norm": 2.40625,
      "learning_rate": 2.95e-05,
      "loss": 9.4197,
      "mean_token_accuracy": 0.049464210495352746,
      "num_tokens": 125037.0,
      "step": 60
    },
    {
      "entropy": 10.460656833648681,
      "epoch": 0.00589622641509434,
      "grad_norm": 2.40625,
      "learning_rate": 3.2e-05,
      "loss": 9.2345,
      "mean_token_accuracy": 0.06379429697990417,
      "num_tokens": 134721.0,
      "step": 65
    },
    {
      "entropy": 10.382928371429443,
      "epoch": 0.00634978229317852,
      "grad_norm": 2.546875,
      "learning_rate": 3.4500000000000005e-05,
      "loss": 9.143,
      "mean_token_accuracy": 0.06433153599500656,
      "num_tokens": 144448.0,
      "step": 70
    },
    {
      "entropy": 10.468272972106934,
      "epoch": 0.006803338171262699,
      "grad_norm": 2.40625,
      "learning_rate": 3.7e-05,
      "loss": 9.0501,
      "mean_token_accuracy": 0.0702673528343439,
      "num_tokens": 154343.0,
      "step": 75
    },
    {
      "entropy": 10.42321710586548,
      "epoch": 0.00725689404934688,
      "grad_norm": 2.359375,
      "learning_rate": 3.95e-05,
      "loss": 9.0122,
      "mean_token_accuracy": 0.07085253149271012,
      "num_tokens": 165650.0,
      "step": 80
    },
    {
      "entropy": 10.375396060943604,
      "epoch": 0.007710449927431059,
      "grad_norm": 2.21875,
      "learning_rate": 4.2000000000000004e-05,
      "loss": 9.0095,
      "mean_token_accuracy": 0.06769064106047154,
      "num_tokens": 177098.0,
      "step": 85
    },
    {
      "entropy": 10.209734153747558,
      "epoch": 0.00816400580551524,
      "grad_norm": 2.46875,
      "learning_rate": 4.45e-05,
      "loss": 8.7346,
      "mean_token_accuracy": 0.08102216497063637,
      "num_tokens": 187903.0,
      "step": 90
    },
    {
      "entropy": 10.184309768676759,
      "epoch": 0.00861756168359942,
      "grad_norm": 2.9375,
      "learning_rate": 4.7000000000000004e-05,
      "loss": 8.6625,
      "mean_token_accuracy": 0.07625751569867134,
      "num_tokens": 198077.0,
      "step": 95
    },
    {
      "entropy": 10.164143085479736,
      "epoch": 0.009071117561683599,
      "grad_norm": 2.0625,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 8.6488,
      "mean_token_accuracy": 0.0765529602766037,
      "num_tokens": 208641.0,
      "step": 100
    },
    {
      "entropy": 10.11404504776001,
      "epoch": 0.009524673439767779,
      "grad_norm": 2.21875,
      "learning_rate": 5.2e-05,
      "loss": 8.4459,
      "mean_token_accuracy": 0.07904929965734482,
      "num_tokens": 218398.0,
      "step": 105
    },
    {
      "entropy": 10.020095443725586,
      "epoch": 0.00997822931785196,
      "grad_norm": 1.796875,
      "learning_rate": 5.45e-05,
      "loss": 8.3096,
      "mean_token_accuracy": 0.08072945550084114,
      "num_tokens": 229059.0,
      "step": 110
    },
    {
      "entropy": 9.92689781188965,
      "epoch": 0.01043178519593614,
      "grad_norm": 1.6796875,
      "learning_rate": 5.7e-05,
      "loss": 8.2562,
      "mean_token_accuracy": 0.07845577970147133,
      "num_tokens": 239775.0,
      "step": 115
    },
    {
      "entropy": 9.88659610748291,
      "epoch": 0.01088534107402032,
      "grad_norm": 1.90625,
      "learning_rate": 5.9499999999999996e-05,
      "loss": 8.0783,
      "mean_token_accuracy": 0.08167890310287476,
      "num_tokens": 249954.0,
      "step": 120
    },
    {
      "entropy": 9.705071830749512,
      "epoch": 0.011338896952104499,
      "grad_norm": 1.6875,
      "learning_rate": 6.2e-05,
      "loss": 7.9982,
      "mean_token_accuracy": 0.08171927258372307,
      "num_tokens": 260056.0,
      "step": 125
    },
    {
      "entropy": 9.503845977783204,
      "epoch": 0.01179245283018868,
      "grad_norm": 1.7421875,
      "learning_rate": 6.450000000000001e-05,
      "loss": 7.8979,
      "mean_token_accuracy": 0.08274013251066208,
      "num_tokens": 269904.0,
      "step": 130
    },
    {
      "entropy": 9.399583435058593,
      "epoch": 0.01224600870827286,
      "grad_norm": 1.640625,
      "learning_rate": 6.7e-05,
      "loss": 7.8002,
      "mean_token_accuracy": 0.08209368959069252,
      "num_tokens": 280900.0,
      "step": 135
    },
    {
      "entropy": 9.210491371154784,
      "epoch": 0.01269956458635704,
      "grad_norm": 1.84375,
      "learning_rate": 6.950000000000001e-05,
      "loss": 7.7037,
      "mean_token_accuracy": 0.08104285597801208,
      "num_tokens": 290844.0,
      "step": 140
    },
    {
      "entropy": 8.886974906921386,
      "epoch": 0.013153120464441219,
      "grad_norm": 1.203125,
      "learning_rate": 7.2e-05,
      "loss": 7.6102,
      "mean_token_accuracy": 0.08451902270317077,
      "num_tokens": 301345.0,
      "step": 145
    },
    {
      "entropy": 8.810495853424072,
      "epoch": 0.013606676342525399,
      "grad_norm": 1.2109375,
      "learning_rate": 7.45e-05,
      "loss": 7.5207,
      "mean_token_accuracy": 0.08492689803242684,
      "num_tokens": 312681.0,
      "step": 150
    },
    {
      "entropy": 8.537842845916748,
      "epoch": 0.01406023222060958,
      "grad_norm": 1.0546875,
      "learning_rate": 7.7e-05,
      "loss": 7.4392,
      "mean_token_accuracy": 0.08759672790765763,
      "num_tokens": 323492.0,
      "step": 155
    },
    {
      "entropy": 8.368613147735596,
      "epoch": 0.01451378809869376,
      "grad_norm": 1.4296875,
      "learning_rate": 7.950000000000001e-05,
      "loss": 7.4411,
      "mean_token_accuracy": 0.08510910198092461,
      "num_tokens": 334521.0,
      "step": 160
    },
    {
      "entropy": 8.248584079742432,
      "epoch": 0.014967343976777939,
      "grad_norm": 1.1953125,
      "learning_rate": 8.2e-05,
      "loss": 7.4246,
      "mean_token_accuracy": 0.08887577280402184,
      "num_tokens": 345468.0,
      "step": 165
    },
    {
      "entropy": 8.11111536026001,
      "epoch": 0.015420899854862119,
      "grad_norm": 1.4375,
      "learning_rate": 8.450000000000001e-05,
      "loss": 7.3297,
      "mean_token_accuracy": 0.0877467267215252,
      "num_tokens": 355911.0,
      "step": 170
    },
    {
      "entropy": 8.027539014816284,
      "epoch": 0.0158744557329463,
      "grad_norm": 1.25,
      "learning_rate": 8.7e-05,
      "loss": 7.3504,
      "mean_token_accuracy": 0.08467918261885643,
      "num_tokens": 367537.0,
      "step": 175
    },
    {
      "entropy": 8.06540780067444,
      "epoch": 0.01632801161103048,
      "grad_norm": 1.3984375,
      "learning_rate": 8.95e-05,
      "loss": 7.4771,
      "mean_token_accuracy": 0.0847482830286026,
      "num_tokens": 377918.0,
      "step": 180
    },
    {
      "entropy": 7.908505392074585,
      "epoch": 0.01678156748911466,
      "grad_norm": 1.3125,
      "learning_rate": 9.2e-05,
      "loss": 7.3165,
      "mean_token_accuracy": 0.08902316763997078,
      "num_tokens": 388461.0,
      "step": 185
    },
    {
      "entropy": 8.000589084625243,
      "epoch": 0.01723512336719884,
      "grad_norm": 1.3125,
      "learning_rate": 9.45e-05,
      "loss": 7.2895,
      "mean_token_accuracy": 0.08787789791822434,
      "num_tokens": 399771.0,
      "step": 190
    },
    {
      "entropy": 7.731726121902466,
      "epoch": 0.01768867924528302,
      "grad_norm": 1.234375,
      "learning_rate": 9.7e-05,
      "loss": 7.1673,
      "mean_token_accuracy": 0.08963026329874993,
      "num_tokens": 410254.0,
      "step": 195
    },
    {
      "entropy": 7.811504650115967,
      "epoch": 0.018142235123367198,
      "grad_norm": 1.390625,
      "learning_rate": 9.95e-05,
      "loss": 7.1894,
      "mean_token_accuracy": 0.08560735136270523,
      "num_tokens": 419792.0,
      "step": 200
    },
    {
      "entropy": 7.708149337768555,
      "epoch": 0.018595791001451378,
      "grad_norm": 1.46875,
      "learning_rate": 0.000102,
      "loss": 7.1962,
      "mean_token_accuracy": 0.09234726577997207,
      "num_tokens": 429264.0,
      "step": 205
    },
    {
      "entropy": 7.65235538482666,
      "epoch": 0.019049346879535557,
      "grad_norm": 1.4375,
      "learning_rate": 0.00010449999999999999,
      "loss": 7.1416,
      "mean_token_accuracy": 0.09288287535309792,
      "num_tokens": 439624.0,
      "step": 210
    },
    {
      "entropy": 7.664721393585205,
      "epoch": 0.01950290275761974,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000107,
      "loss": 7.1224,
      "mean_token_accuracy": 0.0913820892572403,
      "num_tokens": 449556.0,
      "step": 215
    },
    {
      "entropy": 7.666817283630371,
      "epoch": 0.01995645863570392,
      "grad_norm": 1.53125,
      "learning_rate": 0.0001095,
      "loss": 7.1527,
      "mean_token_accuracy": 0.09391891658306122,
      "num_tokens": 460243.0,
      "step": 220
    },
    {
      "entropy": 7.6871044635772705,
      "epoch": 0.0204100145137881,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000112,
      "loss": 7.193,
      "mean_token_accuracy": 0.08998195976018905,
      "num_tokens": 469767.0,
      "step": 225
    },
    {
      "entropy": 7.621262836456299,
      "epoch": 0.02086357039187228,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001145,
      "loss": 7.1106,
      "mean_token_accuracy": 0.09542692303657532,
      "num_tokens": 480329.0,
      "step": 230
    },
    {
      "entropy": 7.601617050170899,
      "epoch": 0.02131712626995646,
      "grad_norm": 1.28125,
      "learning_rate": 0.00011700000000000001,
      "loss": 7.0818,
      "mean_token_accuracy": 0.08891230374574662,
      "num_tokens": 490096.0,
      "step": 235
    },
    {
      "entropy": 7.60865626335144,
      "epoch": 0.02177068214804064,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00011949999999999999,
      "loss": 7.1143,
      "mean_token_accuracy": 0.08756190165877342,
      "num_tokens": 501231.0,
      "step": 240
    },
    {
      "entropy": 7.543486261367798,
      "epoch": 0.022224238026124818,
      "grad_norm": 1.6640625,
      "learning_rate": 0.000122,
      "loss": 7.0258,
      "mean_token_accuracy": 0.09758798629045487,
      "num_tokens": 510722.0,
      "step": 245
    },
    {
      "entropy": 7.489023685455322,
      "epoch": 0.022677793904208998,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0001245,
      "loss": 7.0792,
      "mean_token_accuracy": 0.09397116005420685,
      "num_tokens": 521452.0,
      "step": 250
    },
    {
      "entropy": 7.515778923034668,
      "epoch": 0.023131349782293177,
      "grad_norm": 1.34375,
      "learning_rate": 0.000127,
      "loss": 7.0911,
      "mean_token_accuracy": 0.09408443942666053,
      "num_tokens": 532397.0,
      "step": 255
    },
    {
      "entropy": 7.523461151123047,
      "epoch": 0.02358490566037736,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0001295,
      "loss": 7.0524,
      "mean_token_accuracy": 0.09337728172540664,
      "num_tokens": 542271.0,
      "step": 260
    },
    {
      "entropy": 7.498798656463623,
      "epoch": 0.02403846153846154,
      "grad_norm": 1.296875,
      "learning_rate": 0.000132,
      "loss": 7.0403,
      "mean_token_accuracy": 0.09828322902321815,
      "num_tokens": 552411.0,
      "step": 265
    },
    {
      "entropy": 7.433657121658325,
      "epoch": 0.02449201741654572,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00013450000000000002,
      "loss": 7.0246,
      "mean_token_accuracy": 0.09725753068923951,
      "num_tokens": 561518.0,
      "step": 270
    },
    {
      "entropy": 7.546285247802734,
      "epoch": 0.0249455732946299,
      "grad_norm": 1.265625,
      "learning_rate": 0.00013700000000000002,
      "loss": 7.0652,
      "mean_token_accuracy": 0.09419438466429711,
      "num_tokens": 571361.0,
      "step": 275
    },
    {
      "entropy": 7.3379639148712155,
      "epoch": 0.02539912917271408,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001395,
      "loss": 6.9636,
      "mean_token_accuracy": 0.10313116908073425,
      "num_tokens": 580926.0,
      "step": 280
    },
    {
      "entropy": 7.439048814773559,
      "epoch": 0.025852685050798258,
      "grad_norm": 1.1875,
      "learning_rate": 0.00014199999999999998,
      "loss": 7.0476,
      "mean_token_accuracy": 0.09056098833680153,
      "num_tokens": 592282.0,
      "step": 285
    },
    {
      "entropy": 7.535349416732788,
      "epoch": 0.026306240928882438,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001445,
      "loss": 7.0665,
      "mean_token_accuracy": 0.09440166875720024,
      "num_tokens": 602767.0,
      "step": 290
    },
    {
      "entropy": 7.4240789890289305,
      "epoch": 0.026759796806966617,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000147,
      "loss": 6.9896,
      "mean_token_accuracy": 0.0924597330391407,
      "num_tokens": 612763.0,
      "step": 295
    },
    {
      "entropy": 7.324904584884644,
      "epoch": 0.027213352685050797,
      "grad_norm": 1.046875,
      "learning_rate": 0.0001495,
      "loss": 6.9363,
      "mean_token_accuracy": 0.09375968724489211,
      "num_tokens": 622796.0,
      "step": 300
    },
    {
      "entropy": 7.385919713973999,
      "epoch": 0.027666908563134977,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000152,
      "loss": 6.9192,
      "mean_token_accuracy": 0.09953035265207291,
      "num_tokens": 633136.0,
      "step": 305
    },
    {
      "entropy": 7.27361741065979,
      "epoch": 0.02812046444121916,
      "grad_norm": 1.1875,
      "learning_rate": 0.00015450000000000001,
      "loss": 6.9231,
      "mean_token_accuracy": 0.10048090890049935,
      "num_tokens": 643347.0,
      "step": 310
    },
    {
      "entropy": 7.3609613418579105,
      "epoch": 0.02857402031930334,
      "grad_norm": 1.28125,
      "learning_rate": 0.000157,
      "loss": 6.9037,
      "mean_token_accuracy": 0.10072512328624725,
      "num_tokens": 653096.0,
      "step": 315
    },
    {
      "entropy": 7.219988059997559,
      "epoch": 0.02902757619738752,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0001595,
      "loss": 6.9318,
      "mean_token_accuracy": 0.09513700902462005,
      "num_tokens": 663557.0,
      "step": 320
    },
    {
      "entropy": 7.375028610229492,
      "epoch": 0.0294811320754717,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000162,
      "loss": 7.0052,
      "mean_token_accuracy": 0.0983717530965805,
      "num_tokens": 674395.0,
      "step": 325
    },
    {
      "entropy": 7.270230674743653,
      "epoch": 0.029934687953555878,
      "grad_norm": 1.265625,
      "learning_rate": 0.00016450000000000001,
      "loss": 6.9129,
      "mean_token_accuracy": 0.10161509737372398,
      "num_tokens": 684776.0,
      "step": 330
    },
    {
      "entropy": 7.269373607635498,
      "epoch": 0.030388243831640058,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00016700000000000002,
      "loss": 6.8875,
      "mean_token_accuracy": 0.0997242234647274,
      "num_tokens": 694886.0,
      "step": 335
    },
    {
      "entropy": 7.338870811462402,
      "epoch": 0.030841799709724237,
      "grad_norm": 1.1875,
      "learning_rate": 0.00016950000000000003,
      "loss": 6.9575,
      "mean_token_accuracy": 0.09762076139450074,
      "num_tokens": 705426.0,
      "step": 340
    },
    {
      "entropy": 7.253270244598388,
      "epoch": 0.03129535558780842,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00017199999999999998,
      "loss": 6.9498,
      "mean_token_accuracy": 0.09457632303237914,
      "num_tokens": 716271.0,
      "step": 345
    },
    {
      "entropy": 7.2550302028656,
      "epoch": 0.0317489114658926,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00017449999999999999,
      "loss": 6.9409,
      "mean_token_accuracy": 0.09868695959448814,
      "num_tokens": 726875.0,
      "step": 350
    },
    {
      "entropy": 7.2600914478302006,
      "epoch": 0.032202467343976776,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000177,
      "loss": 6.8738,
      "mean_token_accuracy": 0.09764649495482444,
      "num_tokens": 737825.0,
      "step": 355
    },
    {
      "entropy": 7.240714120864868,
      "epoch": 0.03265602322206096,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0001795,
      "loss": 6.8413,
      "mean_token_accuracy": 0.10239819139242172,
      "num_tokens": 747424.0,
      "step": 360
    },
    {
      "entropy": 7.195548105239868,
      "epoch": 0.033109579100145135,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000182,
      "loss": 6.854,
      "mean_token_accuracy": 0.10043720304965972,
      "num_tokens": 757649.0,
      "step": 365
    },
    {
      "entropy": 7.267161417007446,
      "epoch": 0.03356313497822932,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0001845,
      "loss": 6.8436,
      "mean_token_accuracy": 0.09980886504054069,
      "num_tokens": 768556.0,
      "step": 370
    },
    {
      "entropy": 7.0987019538879395,
      "epoch": 0.034016690856313495,
      "grad_norm": 1.234375,
      "learning_rate": 0.000187,
      "loss": 6.8373,
      "mean_token_accuracy": 0.10219831764698029,
      "num_tokens": 778331.0,
      "step": 375
    },
    {
      "entropy": 7.19783296585083,
      "epoch": 0.03447024673439768,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001895,
      "loss": 6.8352,
      "mean_token_accuracy": 0.09996062368154526,
      "num_tokens": 788065.0,
      "step": 380
    },
    {
      "entropy": 7.185236835479737,
      "epoch": 0.03492380261248186,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000192,
      "loss": 6.8695,
      "mean_token_accuracy": 0.10359352231025695,
      "num_tokens": 798912.0,
      "step": 385
    },
    {
      "entropy": 7.1850183486938475,
      "epoch": 0.03537735849056604,
      "grad_norm": 1.15625,
      "learning_rate": 0.0001945,
      "loss": 6.8211,
      "mean_token_accuracy": 0.10563454255461693,
      "num_tokens": 809580.0,
      "step": 390
    },
    {
      "entropy": 7.208040618896485,
      "epoch": 0.03583091436865022,
      "grad_norm": 1.15625,
      "learning_rate": 0.00019700000000000002,
      "loss": 6.8958,
      "mean_token_accuracy": 0.10004983395338059,
      "num_tokens": 820207.0,
      "step": 395
    },
    {
      "entropy": 7.000587606430054,
      "epoch": 0.036284470246734396,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00019950000000000002,
      "loss": 6.7196,
      "mean_token_accuracy": 0.09804629981517791,
      "num_tokens": 831038.0,
      "step": 400
    },
    {
      "entropy": 7.166787433624267,
      "epoch": 0.03673802612481858,
      "grad_norm": 1.03125,
      "learning_rate": 0.000202,
      "loss": 6.8141,
      "mean_token_accuracy": 0.10281971544027328,
      "num_tokens": 842144.0,
      "step": 405
    },
    {
      "entropy": 7.101940298080445,
      "epoch": 0.037191582002902755,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00020449999999999998,
      "loss": 6.8197,
      "mean_token_accuracy": 0.10505499988794327,
      "num_tokens": 851315.0,
      "step": 410
    },
    {
      "entropy": 7.090035247802734,
      "epoch": 0.03764513788098694,
      "grad_norm": 1.265625,
      "learning_rate": 0.000207,
      "loss": 6.8306,
      "mean_token_accuracy": 0.10175477415323257,
      "num_tokens": 861233.0,
      "step": 415
    },
    {
      "entropy": 7.051157379150391,
      "epoch": 0.038098693759071114,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002095,
      "loss": 6.8096,
      "mean_token_accuracy": 0.0993909016251564,
      "num_tokens": 871131.0,
      "step": 420
    },
    {
      "entropy": 7.143013191223145,
      "epoch": 0.0385522496371553,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000212,
      "loss": 6.7764,
      "mean_token_accuracy": 0.09792252779006957,
      "num_tokens": 882195.0,
      "step": 425
    },
    {
      "entropy": 7.111673784255982,
      "epoch": 0.03900580551523948,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002145,
      "loss": 6.761,
      "mean_token_accuracy": 0.10355701595544815,
      "num_tokens": 891806.0,
      "step": 430
    },
    {
      "entropy": 7.103079462051392,
      "epoch": 0.03945936139332366,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00021700000000000002,
      "loss": 6.8234,
      "mean_token_accuracy": 0.09959669709205628,
      "num_tokens": 901350.0,
      "step": 435
    },
    {
      "entropy": 7.085621690750122,
      "epoch": 0.03991291727140784,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002195,
      "loss": 6.7362,
      "mean_token_accuracy": 0.09833786115050316,
      "num_tokens": 912061.0,
      "step": 440
    },
    {
      "entropy": 6.99485411643982,
      "epoch": 0.040366473149492016,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000222,
      "loss": 6.6882,
      "mean_token_accuracy": 0.1011364847421646,
      "num_tokens": 922220.0,
      "step": 445
    },
    {
      "entropy": 7.05831937789917,
      "epoch": 0.0408200290275762,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002245,
      "loss": 6.7282,
      "mean_token_accuracy": 0.10195261687040329,
      "num_tokens": 932319.0,
      "step": 450
    },
    {
      "entropy": 6.9827405452728275,
      "epoch": 0.041273584905660375,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00022700000000000002,
      "loss": 6.7504,
      "mean_token_accuracy": 0.10525996536016465,
      "num_tokens": 943254.0,
      "step": 455
    },
    {
      "entropy": 7.01410927772522,
      "epoch": 0.04172714078374456,
      "grad_norm": 1.03125,
      "learning_rate": 0.00022950000000000002,
      "loss": 6.7792,
      "mean_token_accuracy": 0.0980103112757206,
      "num_tokens": 953921.0,
      "step": 460
    },
    {
      "entropy": 6.935913562774658,
      "epoch": 0.042180696661828734,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00023200000000000003,
      "loss": 6.6898,
      "mean_token_accuracy": 0.10519141629338265,
      "num_tokens": 963832.0,
      "step": 465
    },
    {
      "entropy": 6.9887090682983395,
      "epoch": 0.04263425253991292,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00023449999999999998,
      "loss": 6.7863,
      "mean_token_accuracy": 0.0979742519557476,
      "num_tokens": 974066.0,
      "step": 470
    },
    {
      "entropy": 7.043638467788696,
      "epoch": 0.0430878084179971,
      "grad_norm": 1.15625,
      "learning_rate": 0.000237,
      "loss": 6.7626,
      "mean_token_accuracy": 0.10633939728140832,
      "num_tokens": 984402.0,
      "step": 475
    },
    {
      "entropy": 6.972577047348023,
      "epoch": 0.04354136429608128,
      "grad_norm": 1.125,
      "learning_rate": 0.0002395,
      "loss": 6.6771,
      "mean_token_accuracy": 0.10603873878717422,
      "num_tokens": 993555.0,
      "step": 480
    },
    {
      "entropy": 6.969974994659424,
      "epoch": 0.04399492017416546,
      "grad_norm": 1.234375,
      "learning_rate": 0.000242,
      "loss": 6.6629,
      "mean_token_accuracy": 0.11064062714576721,
      "num_tokens": 1002920.0,
      "step": 485
    },
    {
      "entropy": 6.993032169342041,
      "epoch": 0.044448476052249636,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002445,
      "loss": 6.7527,
      "mean_token_accuracy": 0.09811311662197113,
      "num_tokens": 1013611.0,
      "step": 490
    },
    {
      "entropy": 6.870787668228149,
      "epoch": 0.04490203193033382,
      "grad_norm": 1.078125,
      "learning_rate": 0.000247,
      "loss": 6.6381,
      "mean_token_accuracy": 0.102886413782835,
      "num_tokens": 1024121.0,
      "step": 495
    },
    {
      "entropy": 7.048051881790161,
      "epoch": 0.045355587808417995,
      "grad_norm": 1.125,
      "learning_rate": 0.0002495,
      "loss": 6.8934,
      "mean_token_accuracy": 0.1061844177544117,
      "num_tokens": 1033845.0,
      "step": 500
    },
    {
      "entropy": 6.991887521743775,
      "epoch": 0.04580914368650218,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000252,
      "loss": 6.6783,
      "mean_token_accuracy": 0.1054874338209629,
      "num_tokens": 1044642.0,
      "step": 505
    },
    {
      "entropy": 6.883953094482422,
      "epoch": 0.046262699564586354,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002545,
      "loss": 6.7359,
      "mean_token_accuracy": 0.10576739087700844,
      "num_tokens": 1054728.0,
      "step": 510
    },
    {
      "entropy": 6.952514743804931,
      "epoch": 0.04671625544267054,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000257,
      "loss": 6.6919,
      "mean_token_accuracy": 0.09888017326593398,
      "num_tokens": 1065743.0,
      "step": 515
    },
    {
      "entropy": 6.847628164291382,
      "epoch": 0.04716981132075472,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002595,
      "loss": 6.6143,
      "mean_token_accuracy": 0.10563460439443588,
      "num_tokens": 1076798.0,
      "step": 520
    },
    {
      "entropy": 6.913150882720947,
      "epoch": 0.047623367198838897,
      "grad_norm": 1.15625,
      "learning_rate": 0.000262,
      "loss": 6.7491,
      "mean_token_accuracy": 0.09963728263974189,
      "num_tokens": 1087445.0,
      "step": 525
    },
    {
      "entropy": 6.850391578674317,
      "epoch": 0.04807692307692308,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00026450000000000003,
      "loss": 6.6457,
      "mean_token_accuracy": 0.10645358338952064,
      "num_tokens": 1098379.0,
      "step": 530
    },
    {
      "entropy": 6.9510291576385494,
      "epoch": 0.048530478955007256,
      "grad_norm": 1.21875,
      "learning_rate": 0.00026700000000000004,
      "loss": 6.7504,
      "mean_token_accuracy": 0.10485646203160286,
      "num_tokens": 1108016.0,
      "step": 535
    },
    {
      "entropy": 6.889407110214234,
      "epoch": 0.04898403483309144,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00026950000000000005,
      "loss": 6.6183,
      "mean_token_accuracy": 0.10373573526740074,
      "num_tokens": 1117887.0,
      "step": 540
    },
    {
      "entropy": 6.825150203704834,
      "epoch": 0.049437590711175615,
      "grad_norm": 1.171875,
      "learning_rate": 0.00027200000000000005,
      "loss": 6.6096,
      "mean_token_accuracy": 0.10104240402579308,
      "num_tokens": 1128497.0,
      "step": 545
    },
    {
      "entropy": 6.80326771736145,
      "epoch": 0.0498911465892598,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002745,
      "loss": 6.6592,
      "mean_token_accuracy": 0.10856908708810806,
      "num_tokens": 1138168.0,
      "step": 550
    },
    {
      "entropy": 6.879168224334717,
      "epoch": 0.050344702467343974,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000277,
      "loss": 6.6407,
      "mean_token_accuracy": 0.10094325393438339,
      "num_tokens": 1148357.0,
      "step": 555
    },
    {
      "entropy": 6.832116413116455,
      "epoch": 0.05079825834542816,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002795,
      "loss": 6.6591,
      "mean_token_accuracy": 0.10421488359570504,
      "num_tokens": 1159964.0,
      "step": 560
    },
    {
      "entropy": 6.893824195861816,
      "epoch": 0.05125181422351233,
      "grad_norm": 1.1875,
      "learning_rate": 0.00028199999999999997,
      "loss": 6.7137,
      "mean_token_accuracy": 0.1055032454431057,
      "num_tokens": 1170683.0,
      "step": 565
    },
    {
      "entropy": 6.914015007019043,
      "epoch": 0.051705370101596516,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002845,
      "loss": 6.6792,
      "mean_token_accuracy": 0.10148620307445526,
      "num_tokens": 1181680.0,
      "step": 570
    },
    {
      "entropy": 6.74883451461792,
      "epoch": 0.0521589259796807,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000287,
      "loss": 6.5654,
      "mean_token_accuracy": 0.10732218623161316,
      "num_tokens": 1191619.0,
      "step": 575
    },
    {
      "entropy": 6.685351276397705,
      "epoch": 0.052612481857764876,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002895,
      "loss": 6.5629,
      "mean_token_accuracy": 0.10640141144394874,
      "num_tokens": 1202264.0,
      "step": 580
    },
    {
      "entropy": 6.805472993850708,
      "epoch": 0.05306603773584906,
      "grad_norm": 1.1875,
      "learning_rate": 0.000292,
      "loss": 6.5815,
      "mean_token_accuracy": 0.1027822844684124,
      "num_tokens": 1212119.0,
      "step": 585
    },
    {
      "entropy": 6.801087665557861,
      "epoch": 0.053519593613933235,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002945,
      "loss": 6.7049,
      "mean_token_accuracy": 0.10856684222817421,
      "num_tokens": 1222317.0,
      "step": 590
    },
    {
      "entropy": 6.853984594345093,
      "epoch": 0.05397314949201742,
      "grad_norm": 1.21875,
      "learning_rate": 0.000297,
      "loss": 6.6443,
      "mean_token_accuracy": 0.1087227538228035,
      "num_tokens": 1232914.0,
      "step": 595
    },
    {
      "entropy": 6.680654525756836,
      "epoch": 0.054426705370101594,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0002995,
      "loss": 6.487,
      "mean_token_accuracy": 0.11319171860814095,
      "num_tokens": 1244214.0,
      "step": 600
    },
    {
      "entropy": 6.830823040008545,
      "epoch": 0.05488026124818578,
      "grad_norm": 1.265625,
      "learning_rate": 0.000302,
      "loss": 6.5837,
      "mean_token_accuracy": 0.10821337848901749,
      "num_tokens": 1254626.0,
      "step": 605
    },
    {
      "entropy": 6.7139935970306395,
      "epoch": 0.05533381712626995,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003045,
      "loss": 6.6937,
      "mean_token_accuracy": 0.10711480379104614,
      "num_tokens": 1265461.0,
      "step": 610
    },
    {
      "entropy": 6.8157871723175045,
      "epoch": 0.055787373004354136,
      "grad_norm": 1.21875,
      "learning_rate": 0.000307,
      "loss": 6.5595,
      "mean_token_accuracy": 0.1111778050661087,
      "num_tokens": 1275810.0,
      "step": 615
    },
    {
      "entropy": 6.6582403659820555,
      "epoch": 0.05624092888243832,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003095,
      "loss": 6.5723,
      "mean_token_accuracy": 0.11056382060050965,
      "num_tokens": 1285170.0,
      "step": 620
    },
    {
      "entropy": 6.752780246734619,
      "epoch": 0.056694484760522496,
      "grad_norm": 1.03125,
      "learning_rate": 0.000312,
      "loss": 6.5104,
      "mean_token_accuracy": 0.11242734342813492,
      "num_tokens": 1295460.0,
      "step": 625
    },
    {
      "entropy": 6.783557653427124,
      "epoch": 0.05714804063860668,
      "grad_norm": 1.375,
      "learning_rate": 0.0003145,
      "loss": 6.647,
      "mean_token_accuracy": 0.10393900722265244,
      "num_tokens": 1306656.0,
      "step": 630
    },
    {
      "entropy": 6.771673727035522,
      "epoch": 0.057601596516690855,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000317,
      "loss": 6.6099,
      "mean_token_accuracy": 0.10909517779946327,
      "num_tokens": 1318019.0,
      "step": 635
    },
    {
      "entropy": 6.627294826507568,
      "epoch": 0.05805515239477504,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003195,
      "loss": 6.5314,
      "mean_token_accuracy": 0.114180638641119,
      "num_tokens": 1328120.0,
      "step": 640
    },
    {
      "entropy": 6.828495216369629,
      "epoch": 0.058508708272859214,
      "grad_norm": 1.15625,
      "learning_rate": 0.000322,
      "loss": 6.6283,
      "mean_token_accuracy": 0.1077908605337143,
      "num_tokens": 1338354.0,
      "step": 645
    },
    {
      "entropy": 6.780737257003784,
      "epoch": 0.0589622641509434,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00032450000000000003,
      "loss": 6.6005,
      "mean_token_accuracy": 0.10598717033863067,
      "num_tokens": 1348666.0,
      "step": 650
    },
    {
      "entropy": 6.688767719268799,
      "epoch": 0.05941582002902757,
      "grad_norm": 1.09375,
      "learning_rate": 0.00032700000000000003,
      "loss": 6.6481,
      "mean_token_accuracy": 0.11127779558300972,
      "num_tokens": 1359713.0,
      "step": 655
    },
    {
      "entropy": 6.6816778659820555,
      "epoch": 0.059869375907111756,
      "grad_norm": 1.125,
      "learning_rate": 0.00032950000000000004,
      "loss": 6.4972,
      "mean_token_accuracy": 0.11233635172247887,
      "num_tokens": 1369507.0,
      "step": 660
    },
    {
      "entropy": 6.753318500518799,
      "epoch": 0.06032293178519594,
      "grad_norm": 1.15625,
      "learning_rate": 0.00033200000000000005,
      "loss": 6.6613,
      "mean_token_accuracy": 0.10458081513643265,
      "num_tokens": 1380726.0,
      "step": 665
    },
    {
      "entropy": 6.791505956649781,
      "epoch": 0.060776487663280115,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00033450000000000005,
      "loss": 6.5884,
      "mean_token_accuracy": 0.1085951142013073,
      "num_tokens": 1391423.0,
      "step": 670
    },
    {
      "entropy": 6.758508014678955,
      "epoch": 0.0612300435413643,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000337,
      "loss": 6.5566,
      "mean_token_accuracy": 0.1129811353981495,
      "num_tokens": 1402492.0,
      "step": 675
    },
    {
      "entropy": 6.64742956161499,
      "epoch": 0.061683599419448475,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003395,
      "loss": 6.6151,
      "mean_token_accuracy": 0.11265368685126305,
      "num_tokens": 1412497.0,
      "step": 680
    },
    {
      "entropy": 6.80935697555542,
      "epoch": 0.06213715529753266,
      "grad_norm": 1.203125,
      "learning_rate": 0.000342,
      "loss": 6.5751,
      "mean_token_accuracy": 0.11512991040945053,
      "num_tokens": 1423214.0,
      "step": 685
    },
    {
      "entropy": 6.651473712921143,
      "epoch": 0.06259071117561683,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00034449999999999997,
      "loss": 6.5673,
      "mean_token_accuracy": 0.1103017933666706,
      "num_tokens": 1434044.0,
      "step": 690
    },
    {
      "entropy": 6.6841058254241945,
      "epoch": 0.06304426705370102,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000347,
      "loss": 6.5303,
      "mean_token_accuracy": 0.1138860672712326,
      "num_tokens": 1444387.0,
      "step": 695
    },
    {
      "entropy": 6.592344141006469,
      "epoch": 0.0634978229317852,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003495,
      "loss": 6.4809,
      "mean_token_accuracy": 0.10669072344899178,
      "num_tokens": 1455157.0,
      "step": 700
    },
    {
      "entropy": 6.729357099533081,
      "epoch": 0.06395137880986937,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000352,
      "loss": 6.595,
      "mean_token_accuracy": 0.11270184963941574,
      "num_tokens": 1466145.0,
      "step": 705
    },
    {
      "entropy": 6.65928430557251,
      "epoch": 0.06440493468795355,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003545,
      "loss": 6.5818,
      "mean_token_accuracy": 0.11218819990754128,
      "num_tokens": 1476713.0,
      "step": 710
    },
    {
      "entropy": 6.686679744720459,
      "epoch": 0.06485849056603774,
      "grad_norm": 1.09375,
      "learning_rate": 0.000357,
      "loss": 6.581,
      "mean_token_accuracy": 0.11432325467467308,
      "num_tokens": 1486908.0,
      "step": 715
    },
    {
      "entropy": 6.704012012481689,
      "epoch": 0.06531204644412192,
      "grad_norm": 0.96875,
      "learning_rate": 0.0003595,
      "loss": 6.5049,
      "mean_token_accuracy": 0.1139966145157814,
      "num_tokens": 1498185.0,
      "step": 720
    },
    {
      "entropy": 6.605288028717041,
      "epoch": 0.0657656023222061,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000362,
      "loss": 6.5477,
      "mean_token_accuracy": 0.11151361390948296,
      "num_tokens": 1508376.0,
      "step": 725
    },
    {
      "entropy": 6.708197355270386,
      "epoch": 0.06621915820029027,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003645,
      "loss": 6.6343,
      "mean_token_accuracy": 0.1107229582965374,
      "num_tokens": 1519953.0,
      "step": 730
    },
    {
      "entropy": 6.5456921577453615,
      "epoch": 0.06667271407837445,
      "grad_norm": 1.046875,
      "learning_rate": 0.000367,
      "loss": 6.4996,
      "mean_token_accuracy": 0.11313785165548325,
      "num_tokens": 1529793.0,
      "step": 735
    },
    {
      "entropy": 6.6532182693481445,
      "epoch": 0.06712626995645864,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003695,
      "loss": 6.4993,
      "mean_token_accuracy": 0.11373248919844628,
      "num_tokens": 1540859.0,
      "step": 740
    },
    {
      "entropy": 6.61553168296814,
      "epoch": 0.06757982583454282,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000372,
      "loss": 6.4324,
      "mean_token_accuracy": 0.1168427474796772,
      "num_tokens": 1549783.0,
      "step": 745
    },
    {
      "entropy": 6.514708757400513,
      "epoch": 0.06803338171262699,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003745,
      "loss": 6.4901,
      "mean_token_accuracy": 0.11619689613580704,
      "num_tokens": 1560522.0,
      "step": 750
    },
    {
      "entropy": 6.587539339065552,
      "epoch": 0.06848693759071117,
      "grad_norm": 1.125,
      "learning_rate": 0.000377,
      "loss": 6.4759,
      "mean_token_accuracy": 0.11305172964930535,
      "num_tokens": 1570668.0,
      "step": 755
    },
    {
      "entropy": 6.66793270111084,
      "epoch": 0.06894049346879536,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003795,
      "loss": 6.5919,
      "mean_token_accuracy": 0.11405429989099503,
      "num_tokens": 1581523.0,
      "step": 760
    },
    {
      "entropy": 6.56304898262024,
      "epoch": 0.06939404934687954,
      "grad_norm": 1.171875,
      "learning_rate": 0.000382,
      "loss": 6.4933,
      "mean_token_accuracy": 0.11518137082457543,
      "num_tokens": 1591356.0,
      "step": 765
    },
    {
      "entropy": 6.5586266040802,
      "epoch": 0.06984760522496372,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003845,
      "loss": 6.4862,
      "mean_token_accuracy": 0.11520140692591667,
      "num_tokens": 1601779.0,
      "step": 770
    },
    {
      "entropy": 6.645415878295898,
      "epoch": 0.07030116110304789,
      "grad_norm": 1.125,
      "learning_rate": 0.00038700000000000003,
      "loss": 6.4799,
      "mean_token_accuracy": 0.11675522029399872,
      "num_tokens": 1611995.0,
      "step": 775
    },
    {
      "entropy": 6.643508338928223,
      "epoch": 0.07075471698113207,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00038950000000000003,
      "loss": 6.51,
      "mean_token_accuracy": 0.11030993685126304,
      "num_tokens": 1621834.0,
      "step": 780
    },
    {
      "entropy": 6.5690505027771,
      "epoch": 0.07120827285921626,
      "grad_norm": 1.28125,
      "learning_rate": 0.00039200000000000004,
      "loss": 6.5502,
      "mean_token_accuracy": 0.10731691345572472,
      "num_tokens": 1632881.0,
      "step": 785
    },
    {
      "entropy": 6.497133684158325,
      "epoch": 0.07166182873730044,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039450000000000005,
      "loss": 6.5383,
      "mean_token_accuracy": 0.11029602736234664,
      "num_tokens": 1642870.0,
      "step": 790
    },
    {
      "entropy": 6.5692647933959964,
      "epoch": 0.07211538461538461,
      "grad_norm": 1.21875,
      "learning_rate": 0.00039700000000000005,
      "loss": 6.4224,
      "mean_token_accuracy": 0.12195004150271416,
      "num_tokens": 1653390.0,
      "step": 795
    },
    {
      "entropy": 6.5903912544250485,
      "epoch": 0.07256894049346879,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003995,
      "loss": 6.5359,
      "mean_token_accuracy": 0.11937625855207443,
      "num_tokens": 1663382.0,
      "step": 800
    },
    {
      "entropy": 6.648660278320312,
      "epoch": 0.07302249637155298,
      "grad_norm": 1.0625,
      "learning_rate": 0.000402,
      "loss": 6.5463,
      "mean_token_accuracy": 0.11221321672201157,
      "num_tokens": 1673601.0,
      "step": 805
    },
    {
      "entropy": 6.414686155319214,
      "epoch": 0.07347605224963716,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004045,
      "loss": 6.4486,
      "mean_token_accuracy": 0.11948997676372528,
      "num_tokens": 1683565.0,
      "step": 810
    },
    {
      "entropy": 6.576791286468506,
      "epoch": 0.07392960812772134,
      "grad_norm": 1.9609375,
      "learning_rate": 0.00040699999999999997,
      "loss": 6.4393,
      "mean_token_accuracy": 0.11694891527295112,
      "num_tokens": 1693636.0,
      "step": 815
    },
    {
      "entropy": 6.726799058914184,
      "epoch": 0.07438316400580551,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004095,
      "loss": 6.543,
      "mean_token_accuracy": 0.1100502498447895,
      "num_tokens": 1703415.0,
      "step": 820
    },
    {
      "entropy": 6.53825044631958,
      "epoch": 0.0748367198838897,
      "grad_norm": 1.265625,
      "learning_rate": 0.000412,
      "loss": 6.5077,
      "mean_token_accuracy": 0.1171564668416977,
      "num_tokens": 1713579.0,
      "step": 825
    },
    {
      "entropy": 6.537848711013794,
      "epoch": 0.07529027576197388,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004145,
      "loss": 6.5234,
      "mean_token_accuracy": 0.11068456545472145,
      "num_tokens": 1724583.0,
      "step": 830
    },
    {
      "entropy": 6.56220064163208,
      "epoch": 0.07574383164005806,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000417,
      "loss": 6.4309,
      "mean_token_accuracy": 0.12289080545306205,
      "num_tokens": 1735293.0,
      "step": 835
    },
    {
      "entropy": 6.558898782730102,
      "epoch": 0.07619738751814223,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004195,
      "loss": 6.4543,
      "mean_token_accuracy": 0.11800618395209313,
      "num_tokens": 1745623.0,
      "step": 840
    },
    {
      "entropy": 6.5176109790802,
      "epoch": 0.07665094339622641,
      "grad_norm": 1.03125,
      "learning_rate": 0.000422,
      "loss": 6.4841,
      "mean_token_accuracy": 0.11565901786088943,
      "num_tokens": 1755713.0,
      "step": 845
    },
    {
      "entropy": 6.523326635360718,
      "epoch": 0.0771044992743106,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004245,
      "loss": 6.4853,
      "mean_token_accuracy": 0.11440500319004059,
      "num_tokens": 1765264.0,
      "step": 850
    },
    {
      "entropy": 6.563967895507813,
      "epoch": 0.07755805515239478,
      "grad_norm": 1.21875,
      "learning_rate": 0.000427,
      "loss": 6.5002,
      "mean_token_accuracy": 0.11388131231069565,
      "num_tokens": 1775769.0,
      "step": 855
    },
    {
      "entropy": 6.483576726913452,
      "epoch": 0.07801161103047896,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004295,
      "loss": 6.4249,
      "mean_token_accuracy": 0.11765370443463326,
      "num_tokens": 1786245.0,
      "step": 860
    },
    {
      "entropy": 6.497152996063233,
      "epoch": 0.07846516690856313,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000432,
      "loss": 6.4549,
      "mean_token_accuracy": 0.12084143534302712,
      "num_tokens": 1796040.0,
      "step": 865
    },
    {
      "entropy": 6.451007413864136,
      "epoch": 0.07891872278664731,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004345,
      "loss": 6.4846,
      "mean_token_accuracy": 0.11668049693107604,
      "num_tokens": 1806007.0,
      "step": 870
    },
    {
      "entropy": 6.6562886238098145,
      "epoch": 0.0793722786647315,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000437,
      "loss": 6.5641,
      "mean_token_accuracy": 0.10741591677069665,
      "num_tokens": 1816935.0,
      "step": 875
    },
    {
      "entropy": 6.411975765228272,
      "epoch": 0.07982583454281568,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004395,
      "loss": 6.4055,
      "mean_token_accuracy": 0.11737755388021469,
      "num_tokens": 1827419.0,
      "step": 880
    },
    {
      "entropy": 6.580957078933716,
      "epoch": 0.08027939042089985,
      "grad_norm": 1.234375,
      "learning_rate": 0.000442,
      "loss": 6.5167,
      "mean_token_accuracy": 0.11306339800357819,
      "num_tokens": 1837933.0,
      "step": 885
    },
    {
      "entropy": 6.480378532409668,
      "epoch": 0.08073294629898403,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004445,
      "loss": 6.5671,
      "mean_token_accuracy": 0.11678754836320877,
      "num_tokens": 1848551.0,
      "step": 890
    },
    {
      "entropy": 6.546552562713623,
      "epoch": 0.08118650217706821,
      "grad_norm": 1.21875,
      "learning_rate": 0.000447,
      "loss": 6.4439,
      "mean_token_accuracy": 0.11490961909294128,
      "num_tokens": 1858380.0,
      "step": 895
    },
    {
      "entropy": 6.448764801025391,
      "epoch": 0.0816400580551524,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044950000000000003,
      "loss": 6.3688,
      "mean_token_accuracy": 0.11925693526864052,
      "num_tokens": 1868291.0,
      "step": 900
    },
    {
      "entropy": 6.491116046905518,
      "epoch": 0.08209361393323658,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045200000000000004,
      "loss": 6.4651,
      "mean_token_accuracy": 0.11819546893239022,
      "num_tokens": 1879219.0,
      "step": 905
    },
    {
      "entropy": 6.526374053955078,
      "epoch": 0.08254716981132075,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00045450000000000004,
      "loss": 6.4579,
      "mean_token_accuracy": 0.1181636244058609,
      "num_tokens": 1889809.0,
      "step": 910
    },
    {
      "entropy": 6.414980411529541,
      "epoch": 0.08300072568940493,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045700000000000005,
      "loss": 6.431,
      "mean_token_accuracy": 0.11433882117271424,
      "num_tokens": 1899986.0,
      "step": 915
    },
    {
      "entropy": 6.483574247360229,
      "epoch": 0.08345428156748912,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045950000000000006,
      "loss": 6.4028,
      "mean_token_accuracy": 0.11887709498405456,
      "num_tokens": 1909314.0,
      "step": 920
    },
    {
      "entropy": 6.4847499370574955,
      "epoch": 0.0839078374455733,
      "grad_norm": 1.078125,
      "learning_rate": 0.000462,
      "loss": 6.4628,
      "mean_token_accuracy": 0.11896910145878792,
      "num_tokens": 1918443.0,
      "step": 925
    },
    {
      "entropy": 6.4445396900177006,
      "epoch": 0.08436139332365747,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004645,
      "loss": 6.4045,
      "mean_token_accuracy": 0.12024879306554795,
      "num_tokens": 1928683.0,
      "step": 930
    },
    {
      "entropy": 6.317193126678466,
      "epoch": 0.08481494920174165,
      "grad_norm": 1.140625,
      "learning_rate": 0.000467,
      "loss": 6.3318,
      "mean_token_accuracy": 0.1153000108897686,
      "num_tokens": 1938293.0,
      "step": 935
    },
    {
      "entropy": 6.431926250457764,
      "epoch": 0.08526850507982583,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004695,
      "loss": 6.4353,
      "mean_token_accuracy": 0.11641831621527672,
      "num_tokens": 1948233.0,
      "step": 940
    },
    {
      "entropy": 6.480042028427124,
      "epoch": 0.08572206095791002,
      "grad_norm": 1.109375,
      "learning_rate": 0.000472,
      "loss": 6.3302,
      "mean_token_accuracy": 0.12561678364872933,
      "num_tokens": 1958611.0,
      "step": 945
    },
    {
      "entropy": 6.443802785873413,
      "epoch": 0.0861756168359942,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004745,
      "loss": 6.4923,
      "mean_token_accuracy": 0.12038066908717156,
      "num_tokens": 1968957.0,
      "step": 950
    },
    {
      "entropy": 6.41972770690918,
      "epoch": 0.08662917271407837,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000477,
      "loss": 6.3452,
      "mean_token_accuracy": 0.12305001318454742,
      "num_tokens": 1978503.0,
      "step": 955
    },
    {
      "entropy": 6.331937170028686,
      "epoch": 0.08708272859216255,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004795,
      "loss": 6.3407,
      "mean_token_accuracy": 0.12022828236222267,
      "num_tokens": 1988458.0,
      "step": 960
    },
    {
      "entropy": 6.463285970687866,
      "epoch": 0.08753628447024674,
      "grad_norm": 1.046875,
      "learning_rate": 0.000482,
      "loss": 6.4719,
      "mean_token_accuracy": 0.11548013314604759,
      "num_tokens": 1999984.0,
      "step": 965
    },
    {
      "entropy": 6.286208868026733,
      "epoch": 0.08798984034833092,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004845,
      "loss": 6.3288,
      "mean_token_accuracy": 0.12638000547885894,
      "num_tokens": 2010645.0,
      "step": 970
    },
    {
      "entropy": 6.497215414047242,
      "epoch": 0.08844339622641509,
      "grad_norm": 1.109375,
      "learning_rate": 0.000487,
      "loss": 6.4547,
      "mean_token_accuracy": 0.11661298498511315,
      "num_tokens": 2020196.0,
      "step": 975
    },
    {
      "entropy": 6.386039924621582,
      "epoch": 0.08889695210449927,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004895,
      "loss": 6.347,
      "mean_token_accuracy": 0.11794270873069763,
      "num_tokens": 2030101.0,
      "step": 980
    },
    {
      "entropy": 6.388373708724975,
      "epoch": 0.08935050798258345,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000492,
      "loss": 6.4225,
      "mean_token_accuracy": 0.11813583448529244,
      "num_tokens": 2040454.0,
      "step": 985
    },
    {
      "entropy": 6.357129955291748,
      "epoch": 0.08980406386066764,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004945,
      "loss": 6.3631,
      "mean_token_accuracy": 0.11890257149934769,
      "num_tokens": 2050378.0,
      "step": 990
    },
    {
      "entropy": 6.4591563701629635,
      "epoch": 0.09025761973875182,
      "grad_norm": 1.0625,
      "learning_rate": 0.000497,
      "loss": 6.4206,
      "mean_token_accuracy": 0.11803730875253678,
      "num_tokens": 2060282.0,
      "step": 995
    },
    {
      "entropy": 6.331292963027954,
      "epoch": 0.09071117561683599,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004995,
      "loss": 6.4023,
      "mean_token_accuracy": 0.11980631649494171,
      "num_tokens": 2070563.0,
      "step": 1000
    },
    {
      "entropy": 6.491535806655884,
      "epoch": 0.09116473149492017,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999999985110217,
      "loss": 6.4912,
      "mean_token_accuracy": 0.11830355301499366,
      "num_tokens": 2080153.0,
      "step": 1005
    },
    {
      "entropy": 6.275622034072876,
      "epoch": 0.09161828737300436,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999999924620471,
      "loss": 6.3284,
      "mean_token_accuracy": 0.1242703415453434,
      "num_tokens": 2090590.0,
      "step": 1010
    },
    {
      "entropy": 6.441268587112427,
      "epoch": 0.09207184325108854,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999999817600155,
      "loss": 6.4116,
      "mean_token_accuracy": 0.11299227103590966,
      "num_tokens": 2101945.0,
      "step": 1015
    },
    {
      "entropy": 6.446867227554321,
      "epoch": 0.09252539912917271,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004999999664049268,
      "loss": 6.394,
      "mean_token_accuracy": 0.12263479754328728,
      "num_tokens": 2112514.0,
      "step": 1020
    },
    {
      "entropy": 6.262259101867675,
      "epoch": 0.09297895500725689,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999999463967816,
      "loss": 6.3454,
      "mean_token_accuracy": 0.12117168605327606,
      "num_tokens": 2122296.0,
      "step": 1025
    },
    {
      "entropy": 6.523186826705933,
      "epoch": 0.09343251088534107,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00049999992173558,
      "loss": 6.4913,
      "mean_token_accuracy": 0.10856073573231698,
      "num_tokens": 2134512.0,
      "step": 1030
    },
    {
      "entropy": 6.369787693023682,
      "epoch": 0.09388606676342526,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999998924213228,
      "loss": 6.421,
      "mean_token_accuracy": 0.1210160993039608,
      "num_tokens": 2143934.0,
      "step": 1035
    },
    {
      "entropy": 6.265434217453003,
      "epoch": 0.09433962264150944,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999998584540105,
      "loss": 6.298,
      "mean_token_accuracy": 0.1182013526558876,
      "num_tokens": 2155730.0,
      "step": 1040
    },
    {
      "entropy": 6.436200380325317,
      "epoch": 0.09479317851959361,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999998198336437,
      "loss": 6.323,
      "mean_token_accuracy": 0.1117444783449173,
      "num_tokens": 2166422.0,
      "step": 1045
    },
    {
      "entropy": 6.347907590866089,
      "epoch": 0.09524673439767779,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999997765602233,
      "loss": 6.4449,
      "mean_token_accuracy": 0.1139302246272564,
      "num_tokens": 2177842.0,
      "step": 1050
    },
    {
      "entropy": 6.457770681381225,
      "epoch": 0.09570029027576198,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999997286337502,
      "loss": 6.504,
      "mean_token_accuracy": 0.11967741474509239,
      "num_tokens": 2188614.0,
      "step": 1055
    },
    {
      "entropy": 6.377467250823974,
      "epoch": 0.09615384615384616,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999996760542254,
      "loss": 6.2878,
      "mean_token_accuracy": 0.12623362839221955,
      "num_tokens": 2197886.0,
      "step": 1060
    },
    {
      "entropy": 6.229464912414551,
      "epoch": 0.09660740203193033,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00049999961882165,
      "loss": 6.3599,
      "mean_token_accuracy": 0.12299013957381248,
      "num_tokens": 2209930.0,
      "step": 1065
    },
    {
      "entropy": 6.440615034103393,
      "epoch": 0.09706095791001451,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999995569360251,
      "loss": 6.3759,
      "mean_token_accuracy": 0.11413345336914063,
      "num_tokens": 2219767.0,
      "step": 1070
    },
    {
      "entropy": 6.328685569763183,
      "epoch": 0.0975145137880987,
      "grad_norm": 1.046875,
      "learning_rate": 0.000499999490397352,
      "loss": 6.3049,
      "mean_token_accuracy": 0.1205204777419567,
      "num_tokens": 2230082.0,
      "step": 1075
    },
    {
      "entropy": 6.243273115158081,
      "epoch": 0.09796806966618288,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999994192056321,
      "loss": 6.3255,
      "mean_token_accuracy": 0.12633560076355935,
      "num_tokens": 2240066.0,
      "step": 1080
    },
    {
      "entropy": 6.374665594100952,
      "epoch": 0.09842162554426705,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999993433608669,
      "loss": 6.3561,
      "mean_token_accuracy": 0.1187953770160675,
      "num_tokens": 2250732.0,
      "step": 1085
    },
    {
      "entropy": 6.321237421035766,
      "epoch": 0.09887518142235123,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999992628630579,
      "loss": 6.2454,
      "mean_token_accuracy": 0.11912132352590561,
      "num_tokens": 2261452.0,
      "step": 1090
    },
    {
      "entropy": 6.250462293624878,
      "epoch": 0.09932873730043541,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999991777122069,
      "loss": 6.3636,
      "mean_token_accuracy": 0.12190868705511093,
      "num_tokens": 2271929.0,
      "step": 1095
    },
    {
      "entropy": 6.321765327453614,
      "epoch": 0.0997822931785196,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004999990879083156,
      "loss": 6.3264,
      "mean_token_accuracy": 0.12127640098333359,
      "num_tokens": 2282987.0,
      "step": 1100
    },
    {
      "entropy": 6.319233846664429,
      "epoch": 0.10023584905660378,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999989934513857,
      "loss": 6.3136,
      "mean_token_accuracy": 0.1254054971039295,
      "num_tokens": 2292796.0,
      "step": 1105
    },
    {
      "entropy": 6.332448720932007,
      "epoch": 0.10068940493468795,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999988943414193,
      "loss": 6.3567,
      "mean_token_accuracy": 0.12186149284243583,
      "num_tokens": 2303057.0,
      "step": 1110
    },
    {
      "entropy": 6.252414035797119,
      "epoch": 0.10114296081277213,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999987905784184,
      "loss": 6.3422,
      "mean_token_accuracy": 0.12239331230521203,
      "num_tokens": 2314168.0,
      "step": 1115
    },
    {
      "entropy": 6.341477918624878,
      "epoch": 0.10159651669085631,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999986821623853,
      "loss": 6.3088,
      "mean_token_accuracy": 0.1192399226129055,
      "num_tokens": 2325241.0,
      "step": 1120
    },
    {
      "entropy": 6.351327323913575,
      "epoch": 0.1020500725689405,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004999985690933219,
      "loss": 6.3276,
      "mean_token_accuracy": 0.12044254839420318,
      "num_tokens": 2334866.0,
      "step": 1125
    },
    {
      "entropy": 6.211726427078247,
      "epoch": 0.10250362844702467,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999984513712311,
      "loss": 6.3425,
      "mean_token_accuracy": 0.11830557957291603,
      "num_tokens": 2345422.0,
      "step": 1130
    },
    {
      "entropy": 6.417554903030395,
      "epoch": 0.10295718432510885,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999983289961146,
      "loss": 6.349,
      "mean_token_accuracy": 0.11906943917274475,
      "num_tokens": 2356285.0,
      "step": 1135
    },
    {
      "entropy": 6.214526510238647,
      "epoch": 0.10341074020319303,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999982019679755,
      "loss": 6.2793,
      "mean_token_accuracy": 0.12734896019101144,
      "num_tokens": 2366861.0,
      "step": 1140
    },
    {
      "entropy": 6.289835214614868,
      "epoch": 0.10386429608127722,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999980702868164,
      "loss": 6.2976,
      "mean_token_accuracy": 0.12919359803199768,
      "num_tokens": 2376851.0,
      "step": 1145
    },
    {
      "entropy": 6.3659947395324705,
      "epoch": 0.1043178519593614,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999979339526396,
      "loss": 6.3591,
      "mean_token_accuracy": 0.12571399062871932,
      "num_tokens": 2386857.0,
      "step": 1150
    },
    {
      "entropy": 6.303523445129395,
      "epoch": 0.10477140783744557,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999977929654484,
      "loss": 6.268,
      "mean_token_accuracy": 0.12874414026737213,
      "num_tokens": 2397035.0,
      "step": 1155
    },
    {
      "entropy": 6.3552806854248045,
      "epoch": 0.10522496371552975,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004999976473252453,
      "loss": 6.3744,
      "mean_token_accuracy": 0.12097596004605293,
      "num_tokens": 2408177.0,
      "step": 1160
    },
    {
      "entropy": 6.359548139572143,
      "epoch": 0.10567851959361393,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999974970320338,
      "loss": 6.3076,
      "mean_token_accuracy": 0.127739118039608,
      "num_tokens": 2418445.0,
      "step": 1165
    },
    {
      "entropy": 6.25285906791687,
      "epoch": 0.10613207547169812,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999973420858165,
      "loss": 6.3116,
      "mean_token_accuracy": 0.12409163489937783,
      "num_tokens": 2428990.0,
      "step": 1170
    },
    {
      "entropy": 6.301597690582275,
      "epoch": 0.10658563134978229,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999971824865968,
      "loss": 6.3509,
      "mean_token_accuracy": 0.11999165937304497,
      "num_tokens": 2439643.0,
      "step": 1175
    },
    {
      "entropy": 6.453162050247192,
      "epoch": 0.10703918722786647,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999970182343782,
      "loss": 6.3972,
      "mean_token_accuracy": 0.12346342280507087,
      "num_tokens": 2450691.0,
      "step": 1180
    },
    {
      "entropy": 6.286817169189453,
      "epoch": 0.10749274310595065,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999968493291638,
      "loss": 6.3281,
      "mean_token_accuracy": 0.12339968159794808,
      "num_tokens": 2460706.0,
      "step": 1185
    },
    {
      "entropy": 6.101443719863892,
      "epoch": 0.10794629898403484,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004999966757709573,
      "loss": 6.223,
      "mean_token_accuracy": 0.12936578840017318,
      "num_tokens": 2470947.0,
      "step": 1190
    },
    {
      "entropy": 6.308624172210694,
      "epoch": 0.10839985486211902,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999964975597622,
      "loss": 6.1877,
      "mean_token_accuracy": 0.12678698301315308,
      "num_tokens": 2481132.0,
      "step": 1195
    },
    {
      "entropy": 6.23999056816101,
      "epoch": 0.10885341074020319,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999963146955821,
      "loss": 6.2221,
      "mean_token_accuracy": 0.13168013989925384,
      "num_tokens": 2491701.0,
      "step": 1200
    },
    {
      "entropy": 6.209512996673584,
      "epoch": 0.10930696661828737,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000499996127178421,
      "loss": 6.3154,
      "mean_token_accuracy": 0.12373346984386444,
      "num_tokens": 2502360.0,
      "step": 1205
    },
    {
      "entropy": 6.259164810180664,
      "epoch": 0.10976052249637155,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999959350082825,
      "loss": 6.1911,
      "mean_token_accuracy": 0.12604912146925926,
      "num_tokens": 2512140.0,
      "step": 1210
    },
    {
      "entropy": 6.277319002151489,
      "epoch": 0.11021407837445574,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999957381851709,
      "loss": 6.3226,
      "mean_token_accuracy": 0.11895870566368102,
      "num_tokens": 2522721.0,
      "step": 1215
    },
    {
      "entropy": 6.195718097686767,
      "epoch": 0.1106676342525399,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999955367090901,
      "loss": 6.1883,
      "mean_token_accuracy": 0.12740988954901694,
      "num_tokens": 2531786.0,
      "step": 1220
    },
    {
      "entropy": 6.27605619430542,
      "epoch": 0.11112119013062409,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004999953305800441,
      "loss": 6.3046,
      "mean_token_accuracy": 0.1145807795226574,
      "num_tokens": 2542895.0,
      "step": 1225
    },
    {
      "entropy": 6.222184085845948,
      "epoch": 0.11157474600870827,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999951197980374,
      "loss": 6.2913,
      "mean_token_accuracy": 0.12083523273468018,
      "num_tokens": 2554120.0,
      "step": 1230
    },
    {
      "entropy": 6.31572117805481,
      "epoch": 0.11202830188679246,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999949043630744,
      "loss": 6.3231,
      "mean_token_accuracy": 0.12528388127684592,
      "num_tokens": 2563935.0,
      "step": 1235
    },
    {
      "entropy": 6.163136672973633,
      "epoch": 0.11248185776487664,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999946842751593,
      "loss": 6.2399,
      "mean_token_accuracy": 0.12615554332733153,
      "num_tokens": 2574465.0,
      "step": 1240
    },
    {
      "entropy": 6.261920547485351,
      "epoch": 0.11293541364296081,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999944595342968,
      "loss": 6.1907,
      "mean_token_accuracy": 0.12804502844810486,
      "num_tokens": 2584694.0,
      "step": 1245
    },
    {
      "entropy": 6.215864896774292,
      "epoch": 0.11338896952104499,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999942301404916,
      "loss": 6.165,
      "mean_token_accuracy": 0.1279282294213772,
      "num_tokens": 2594540.0,
      "step": 1250
    },
    {
      "entropy": 6.151381015777588,
      "epoch": 0.11384252539912917,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999939960937484,
      "loss": 6.0988,
      "mean_token_accuracy": 0.1311326391994953,
      "num_tokens": 2604896.0,
      "step": 1255
    },
    {
      "entropy": 6.2653666019439695,
      "epoch": 0.11429608127721336,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999937573940721,
      "loss": 6.2806,
      "mean_token_accuracy": 0.13130278512835503,
      "num_tokens": 2614143.0,
      "step": 1260
    },
    {
      "entropy": 6.085949420928955,
      "epoch": 0.11474963715529753,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999935140414674,
      "loss": 6.1738,
      "mean_token_accuracy": 0.12467294484376908,
      "num_tokens": 2623999.0,
      "step": 1265
    },
    {
      "entropy": 6.378130960464477,
      "epoch": 0.11520319303338171,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999932660359395,
      "loss": 6.3881,
      "mean_token_accuracy": 0.12474165633320808,
      "num_tokens": 2636208.0,
      "step": 1270
    },
    {
      "entropy": 6.177397966384888,
      "epoch": 0.11565674891146589,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999930133774936,
      "loss": 6.2635,
      "mean_token_accuracy": 0.1310955010354519,
      "num_tokens": 2645854.0,
      "step": 1275
    },
    {
      "entropy": 6.334166765213013,
      "epoch": 0.11611030478955008,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999927560661348,
      "loss": 6.2513,
      "mean_token_accuracy": 0.13245694637298583,
      "num_tokens": 2655877.0,
      "step": 1280
    },
    {
      "entropy": 6.226203727722168,
      "epoch": 0.11656386066763426,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999924941018685,
      "loss": 6.2578,
      "mean_token_accuracy": 0.12858212292194365,
      "num_tokens": 2666157.0,
      "step": 1285
    },
    {
      "entropy": 6.180522871017456,
      "epoch": 0.11701741654571843,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999922274847,
      "loss": 6.1825,
      "mean_token_accuracy": 0.1282689355313778,
      "num_tokens": 2676647.0,
      "step": 1290
    },
    {
      "entropy": 6.1607826232910154,
      "epoch": 0.11747097242380261,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000499991956214635,
      "loss": 6.2123,
      "mean_token_accuracy": 0.14043712615966797,
      "num_tokens": 2686417.0,
      "step": 1295
    },
    {
      "entropy": 6.220397090911865,
      "epoch": 0.1179245283018868,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000499991680291679,
      "loss": 6.2217,
      "mean_token_accuracy": 0.1216107651591301,
      "num_tokens": 2697540.0,
      "step": 1300
    },
    {
      "entropy": 6.173789548873901,
      "epoch": 0.11837808417997098,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999913997158378,
      "loss": 6.2622,
      "mean_token_accuracy": 0.11780738160014152,
      "num_tokens": 2709397.0,
      "step": 1305
    },
    {
      "entropy": 6.271051549911499,
      "epoch": 0.11883164005805515,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999911144871169,
      "loss": 6.2158,
      "mean_token_accuracy": 0.1249967023730278,
      "num_tokens": 2720430.0,
      "step": 1310
    },
    {
      "entropy": 6.155363607406616,
      "epoch": 0.11928519593613933,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999908246055226,
      "loss": 6.1669,
      "mean_token_accuracy": 0.12677768394351005,
      "num_tokens": 2730648.0,
      "step": 1315
    },
    {
      "entropy": 6.153974342346191,
      "epoch": 0.11973875181422351,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999905300710607,
      "loss": 6.1472,
      "mean_token_accuracy": 0.1284649021923542,
      "num_tokens": 2739979.0,
      "step": 1320
    },
    {
      "entropy": 6.25038104057312,
      "epoch": 0.1201923076923077,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004999902308837373,
      "loss": 6.3028,
      "mean_token_accuracy": 0.12176627144217492,
      "num_tokens": 2750329.0,
      "step": 1325
    },
    {
      "entropy": 6.140483713150024,
      "epoch": 0.12064586357039188,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999899270435584,
      "loss": 6.1728,
      "mean_token_accuracy": 0.1314334973692894,
      "num_tokens": 2760535.0,
      "step": 1330
    },
    {
      "entropy": 6.230471897125244,
      "epoch": 0.12109941944847605,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999896185505307,
      "loss": 6.2044,
      "mean_token_accuracy": 0.12265649288892747,
      "num_tokens": 2770470.0,
      "step": 1335
    },
    {
      "entropy": 6.159397172927856,
      "epoch": 0.12155297532656023,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004999893054046603,
      "loss": 6.2145,
      "mean_token_accuracy": 0.12754668965935706,
      "num_tokens": 2781506.0,
      "step": 1340
    },
    {
      "entropy": 6.125198554992676,
      "epoch": 0.12200653120464441,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999889876059537,
      "loss": 6.1021,
      "mean_token_accuracy": 0.13616543412208557,
      "num_tokens": 2791245.0,
      "step": 1345
    },
    {
      "entropy": 6.152018737792969,
      "epoch": 0.1224600870827286,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999886651544176,
      "loss": 6.1054,
      "mean_token_accuracy": 0.13110152259469032,
      "num_tokens": 2801680.0,
      "step": 1350
    },
    {
      "entropy": 6.157893323898316,
      "epoch": 0.12291364296081277,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999883380500584,
      "loss": 6.1862,
      "mean_token_accuracy": 0.13278760090470315,
      "num_tokens": 2812805.0,
      "step": 1355
    },
    {
      "entropy": 6.238238143920898,
      "epoch": 0.12336719883889695,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999880062928831,
      "loss": 6.2099,
      "mean_token_accuracy": 0.12580350860953332,
      "num_tokens": 2822834.0,
      "step": 1360
    },
    {
      "entropy": 6.129199361801147,
      "epoch": 0.12382075471698113,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999876698828985,
      "loss": 6.1785,
      "mean_token_accuracy": 0.1322852909564972,
      "num_tokens": 2832312.0,
      "step": 1365
    },
    {
      "entropy": 6.211675024032592,
      "epoch": 0.12427431059506532,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999873288201116,
      "loss": 6.1859,
      "mean_token_accuracy": 0.13278617039322854,
      "num_tokens": 2842998.0,
      "step": 1370
    },
    {
      "entropy": 6.222190952301025,
      "epoch": 0.1247278664731495,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004999869831045294,
      "loss": 6.22,
      "mean_token_accuracy": 0.12911526262760162,
      "num_tokens": 2854218.0,
      "step": 1375
    },
    {
      "entropy": 6.1174554347991945,
      "epoch": 0.12518142235123367,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999866327361589,
      "loss": 6.0953,
      "mean_token_accuracy": 0.13779579773545264,
      "num_tokens": 2864484.0,
      "step": 1380
    },
    {
      "entropy": 6.212736892700195,
      "epoch": 0.12563497822931785,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999862777150077,
      "loss": 6.2057,
      "mean_token_accuracy": 0.12786780819296836,
      "num_tokens": 2873800.0,
      "step": 1385
    },
    {
      "entropy": 6.171827030181885,
      "epoch": 0.12608853410740203,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999859180410829,
      "loss": 6.2171,
      "mean_token_accuracy": 0.1288387209177017,
      "num_tokens": 2884247.0,
      "step": 1390
    },
    {
      "entropy": 6.141082382202148,
      "epoch": 0.12654208998548622,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999855537143919,
      "loss": 6.1257,
      "mean_token_accuracy": 0.12706563696265222,
      "num_tokens": 2894739.0,
      "step": 1395
    },
    {
      "entropy": 6.073680734634399,
      "epoch": 0.1269956458635704,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999851847349425,
      "loss": 6.1807,
      "mean_token_accuracy": 0.13040847554802895,
      "num_tokens": 2904376.0,
      "step": 1400
    },
    {
      "entropy": 6.245872640609742,
      "epoch": 0.12744920174165458,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999848111027419,
      "loss": 6.2325,
      "mean_token_accuracy": 0.127915970236063,
      "num_tokens": 2914304.0,
      "step": 1405
    },
    {
      "entropy": 6.183934211730957,
      "epoch": 0.12790275761973874,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999844328177983,
      "loss": 6.1235,
      "mean_token_accuracy": 0.12716101333498955,
      "num_tokens": 2924400.0,
      "step": 1410
    },
    {
      "entropy": 6.2829203605651855,
      "epoch": 0.12835631349782292,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999840498801191,
      "loss": 6.347,
      "mean_token_accuracy": 0.13103133216500282,
      "num_tokens": 2935040.0,
      "step": 1415
    },
    {
      "entropy": 6.135620164871216,
      "epoch": 0.1288098693759071,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999836622897126,
      "loss": 6.1366,
      "mean_token_accuracy": 0.13396168127655983,
      "num_tokens": 2945536.0,
      "step": 1420
    },
    {
      "entropy": 6.0857262134552,
      "epoch": 0.1292634252539913,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999832700465865,
      "loss": 6.0541,
      "mean_token_accuracy": 0.13348359540104865,
      "num_tokens": 2954969.0,
      "step": 1425
    },
    {
      "entropy": 6.16528549194336,
      "epoch": 0.12971698113207547,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999828731507491,
      "loss": 6.1774,
      "mean_token_accuracy": 0.12718980759382248,
      "num_tokens": 2966218.0,
      "step": 1430
    },
    {
      "entropy": 6.185882234573365,
      "epoch": 0.13017053701015965,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999824716022085,
      "loss": 6.1999,
      "mean_token_accuracy": 0.12183153331279754,
      "num_tokens": 2976640.0,
      "step": 1435
    },
    {
      "entropy": 6.093891048431397,
      "epoch": 0.13062409288824384,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999820654009731,
      "loss": 6.1687,
      "mean_token_accuracy": 0.131185482442379,
      "num_tokens": 2985883.0,
      "step": 1440
    },
    {
      "entropy": 6.16903977394104,
      "epoch": 0.13107764876632802,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999816545470513,
      "loss": 6.1739,
      "mean_token_accuracy": 0.12456371188163758,
      "num_tokens": 2996695.0,
      "step": 1445
    },
    {
      "entropy": 6.17958173751831,
      "epoch": 0.1315312046444122,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999812390404514,
      "loss": 6.1945,
      "mean_token_accuracy": 0.128679159283638,
      "num_tokens": 3007500.0,
      "step": 1450
    },
    {
      "entropy": 5.967514085769653,
      "epoch": 0.13198476052249636,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999808188811823,
      "loss": 6.0384,
      "mean_token_accuracy": 0.13110998794436454,
      "num_tokens": 3017299.0,
      "step": 1455
    },
    {
      "entropy": 6.2758769512176515,
      "epoch": 0.13243831640058054,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999803940692524,
      "loss": 6.3073,
      "mean_token_accuracy": 0.1230485461652279,
      "num_tokens": 3028389.0,
      "step": 1460
    },
    {
      "entropy": 6.220829582214355,
      "epoch": 0.13289187227866472,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999799646046707,
      "loss": 6.1307,
      "mean_token_accuracy": 0.13422513082623483,
      "num_tokens": 3038883.0,
      "step": 1465
    },
    {
      "entropy": 6.163057661056518,
      "epoch": 0.1333454281567489,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004999795304874461,
      "loss": 6.2457,
      "mean_token_accuracy": 0.13295281678438187,
      "num_tokens": 3049904.0,
      "step": 1470
    },
    {
      "entropy": 6.084758853912353,
      "epoch": 0.1337989840348331,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999790917175873,
      "loss": 6.0809,
      "mean_token_accuracy": 0.13047859445214272,
      "num_tokens": 3059788.0,
      "step": 1475
    },
    {
      "entropy": 6.1702046394348145,
      "epoch": 0.13425253991291727,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999786482951036,
      "loss": 6.0729,
      "mean_token_accuracy": 0.1356496147811413,
      "num_tokens": 3070413.0,
      "step": 1480
    },
    {
      "entropy": 6.121113586425781,
      "epoch": 0.13470609579100146,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004999782002200041,
      "loss": 6.2024,
      "mean_token_accuracy": 0.12601577937602998,
      "num_tokens": 3081093.0,
      "step": 1485
    },
    {
      "entropy": 6.136486053466797,
      "epoch": 0.13515965166908564,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999777474922982,
      "loss": 6.1304,
      "mean_token_accuracy": 0.12720816284418107,
      "num_tokens": 3090594.0,
      "step": 1490
    },
    {
      "entropy": 6.193129396438598,
      "epoch": 0.13561320754716982,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999772901119951,
      "loss": 6.2432,
      "mean_token_accuracy": 0.12697395980358123,
      "num_tokens": 3101061.0,
      "step": 1495
    },
    {
      "entropy": 6.138486433029175,
      "epoch": 0.13606676342525398,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999768280791043,
      "loss": 6.1538,
      "mean_token_accuracy": 0.13143179193139076,
      "num_tokens": 3111715.0,
      "step": 1500
    },
    {
      "entropy": 6.113825941085816,
      "epoch": 0.13652031930333816,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999763613936352,
      "loss": 6.0641,
      "mean_token_accuracy": 0.13530101105570794,
      "num_tokens": 3121378.0,
      "step": 1505
    },
    {
      "entropy": 6.079781913757325,
      "epoch": 0.13697387518142234,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999758900555979,
      "loss": 6.013,
      "mean_token_accuracy": 0.13859829008579255,
      "num_tokens": 3131229.0,
      "step": 1510
    },
    {
      "entropy": 6.068106603622437,
      "epoch": 0.13742743105950653,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004999754140650017,
      "loss": 6.1133,
      "mean_token_accuracy": 0.12841149419546127,
      "num_tokens": 3140859.0,
      "step": 1515
    },
    {
      "entropy": 6.186432981491089,
      "epoch": 0.1378809869375907,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004999749334218567,
      "loss": 6.1079,
      "mean_token_accuracy": 0.13175800815224648,
      "num_tokens": 3151692.0,
      "step": 1520
    },
    {
      "entropy": 6.034256410598755,
      "epoch": 0.1383345428156749,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999744481261726,
      "loss": 6.1773,
      "mean_token_accuracy": 0.12917248532176018,
      "num_tokens": 3161879.0,
      "step": 1525
    },
    {
      "entropy": 6.104421138763428,
      "epoch": 0.13878809869375908,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999739581779597,
      "loss": 6.0365,
      "mean_token_accuracy": 0.1392836257815361,
      "num_tokens": 3171282.0,
      "step": 1530
    },
    {
      "entropy": 6.098322439193725,
      "epoch": 0.13924165457184326,
      "grad_norm": 1.03125,
      "learning_rate": 0.000499973463577228,
      "loss": 6.0919,
      "mean_token_accuracy": 0.13791340962052345,
      "num_tokens": 3181398.0,
      "step": 1535
    },
    {
      "entropy": 6.116310262680054,
      "epoch": 0.13969521044992744,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999729643239877,
      "loss": 6.221,
      "mean_token_accuracy": 0.13014425411820413,
      "num_tokens": 3191431.0,
      "step": 1540
    },
    {
      "entropy": 6.162513589859008,
      "epoch": 0.1401487663280116,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999724604182492,
      "loss": 6.2272,
      "mean_token_accuracy": 0.12255278900265694,
      "num_tokens": 3202147.0,
      "step": 1545
    },
    {
      "entropy": 6.2134605884552006,
      "epoch": 0.14060232220609578,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999719518600229,
      "loss": 6.0894,
      "mean_token_accuracy": 0.13580598905682564,
      "num_tokens": 3212428.0,
      "step": 1550
    },
    {
      "entropy": 6.07160701751709,
      "epoch": 0.14105587808417996,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999714386493192,
      "loss": 6.1527,
      "mean_token_accuracy": 0.13032715693116187,
      "num_tokens": 3222099.0,
      "step": 1555
    },
    {
      "entropy": 6.135047817230225,
      "epoch": 0.14150943396226415,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999709207861489,
      "loss": 6.0821,
      "mean_token_accuracy": 0.13343799188733102,
      "num_tokens": 3232079.0,
      "step": 1560
    },
    {
      "entropy": 6.043800163269043,
      "epoch": 0.14196298984034833,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999703982705227,
      "loss": 6.0044,
      "mean_token_accuracy": 0.14089397639036177,
      "num_tokens": 3241762.0,
      "step": 1565
    },
    {
      "entropy": 6.056787490844727,
      "epoch": 0.1424165457184325,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999698711024513,
      "loss": 6.0942,
      "mean_token_accuracy": 0.13566121682524682,
      "num_tokens": 3252571.0,
      "step": 1570
    },
    {
      "entropy": 6.119260835647583,
      "epoch": 0.1428701015965167,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999693392819456,
      "loss": 6.1276,
      "mean_token_accuracy": 0.12969395369291306,
      "num_tokens": 3263542.0,
      "step": 1575
    },
    {
      "entropy": 6.129227113723755,
      "epoch": 0.14332365747460088,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999688028090166,
      "loss": 6.2,
      "mean_token_accuracy": 0.12882068231701851,
      "num_tokens": 3273295.0,
      "step": 1580
    },
    {
      "entropy": 6.217692804336548,
      "epoch": 0.14377721335268506,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999682616836755,
      "loss": 6.23,
      "mean_token_accuracy": 0.13299182802438736,
      "num_tokens": 3283605.0,
      "step": 1585
    },
    {
      "entropy": 6.094928216934204,
      "epoch": 0.14423076923076922,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999677159059334,
      "loss": 6.0657,
      "mean_token_accuracy": 0.1365733303129673,
      "num_tokens": 3293837.0,
      "step": 1590
    },
    {
      "entropy": 5.993037939071655,
      "epoch": 0.1446843251088534,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004999671654758016,
      "loss": 6.0685,
      "mean_token_accuracy": 0.14040320217609406,
      "num_tokens": 3304131.0,
      "step": 1595
    },
    {
      "entropy": 6.073614025115967,
      "epoch": 0.14513788098693758,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004999666103932915,
      "loss": 6.0992,
      "mean_token_accuracy": 0.13104889690876007,
      "num_tokens": 3313543.0,
      "step": 1600
    },
    {
      "entropy": 6.087307071685791,
      "epoch": 0.14559143686502177,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999660506584145,
      "loss": 6.0907,
      "mean_token_accuracy": 0.13283502534031869,
      "num_tokens": 3324487.0,
      "step": 1605
    },
    {
      "entropy": 6.13335018157959,
      "epoch": 0.14604499274310595,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999654862711823,
      "loss": 6.0993,
      "mean_token_accuracy": 0.13403137400746346,
      "num_tokens": 3335094.0,
      "step": 1610
    },
    {
      "entropy": 6.077793455123901,
      "epoch": 0.14649854862119013,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999649172316065,
      "loss": 6.033,
      "mean_token_accuracy": 0.13351959139108657,
      "num_tokens": 3344252.0,
      "step": 1615
    },
    {
      "entropy": 6.051349306106568,
      "epoch": 0.14695210449927432,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000499964343539699,
      "loss": 6.1092,
      "mean_token_accuracy": 0.12948338314890862,
      "num_tokens": 3354265.0,
      "step": 1620
    },
    {
      "entropy": 6.167207384109497,
      "epoch": 0.1474056603773585,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999637651954714,
      "loss": 6.2254,
      "mean_token_accuracy": 0.1327558323740959,
      "num_tokens": 3364616.0,
      "step": 1625
    },
    {
      "entropy": 6.135294818878174,
      "epoch": 0.14785921625544268,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999631821989358,
      "loss": 6.1636,
      "mean_token_accuracy": 0.1339091271162033,
      "num_tokens": 3373650.0,
      "step": 1630
    },
    {
      "entropy": 5.991321992874146,
      "epoch": 0.14831277213352684,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999625945501043,
      "loss": 6.0624,
      "mean_token_accuracy": 0.13642672672867776,
      "num_tokens": 3384590.0,
      "step": 1635
    },
    {
      "entropy": 6.221540021896362,
      "epoch": 0.14876632801161102,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000499962002248989,
      "loss": 6.1045,
      "mean_token_accuracy": 0.13118307739496232,
      "num_tokens": 3395616.0,
      "step": 1640
    },
    {
      "entropy": 6.03116774559021,
      "epoch": 0.1492198838896952,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999614052956024,
      "loss": 6.092,
      "mean_token_accuracy": 0.13425416350364686,
      "num_tokens": 3405810.0,
      "step": 1645
    },
    {
      "entropy": 6.0968505859375,
      "epoch": 0.1496734397677794,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999608036899563,
      "loss": 6.1287,
      "mean_token_accuracy": 0.13202429711818695,
      "num_tokens": 3415932.0,
      "step": 1650
    },
    {
      "entropy": 6.095308542251587,
      "epoch": 0.15012699564586357,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999601974320636,
      "loss": 6.0899,
      "mean_token_accuracy": 0.12914085909724235,
      "num_tokens": 3425626.0,
      "step": 1655
    },
    {
      "entropy": 6.025156164169312,
      "epoch": 0.15058055152394775,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999595865219367,
      "loss": 6.0764,
      "mean_token_accuracy": 0.12880702540278435,
      "num_tokens": 3434837.0,
      "step": 1660
    },
    {
      "entropy": 6.111604022979736,
      "epoch": 0.15103410740203194,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999589709595881,
      "loss": 6.1073,
      "mean_token_accuracy": 0.13633522242307664,
      "num_tokens": 3445238.0,
      "step": 1665
    },
    {
      "entropy": 6.0251708984375,
      "epoch": 0.15148766328011612,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999583507450308,
      "loss": 6.0934,
      "mean_token_accuracy": 0.1300239272415638,
      "num_tokens": 3456738.0,
      "step": 1670
    },
    {
      "entropy": 5.999286699295044,
      "epoch": 0.1519412191582003,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999577258782774,
      "loss": 5.957,
      "mean_token_accuracy": 0.13702991008758544,
      "num_tokens": 3466296.0,
      "step": 1675
    },
    {
      "entropy": 6.128127813339233,
      "epoch": 0.15239477503628446,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000499957096359341,
      "loss": 6.1536,
      "mean_token_accuracy": 0.12765773385763168,
      "num_tokens": 3476989.0,
      "step": 1680
    },
    {
      "entropy": 6.0421911716461185,
      "epoch": 0.15284833091436864,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004999564621882343,
      "loss": 6.0026,
      "mean_token_accuracy": 0.1342094786465168,
      "num_tokens": 3487050.0,
      "step": 1685
    },
    {
      "entropy": 6.144343376159668,
      "epoch": 0.15330188679245282,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999558233649708,
      "loss": 6.2183,
      "mean_token_accuracy": 0.12711812555789948,
      "num_tokens": 3497777.0,
      "step": 1690
    },
    {
      "entropy": 6.131715631484985,
      "epoch": 0.153755442670537,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999551798895636,
      "loss": 6.1137,
      "mean_token_accuracy": 0.12770443707704543,
      "num_tokens": 3509066.0,
      "step": 1695
    },
    {
      "entropy": 6.020261526107788,
      "epoch": 0.1542089985486212,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999545317620258,
      "loss": 6.0385,
      "mean_token_accuracy": 0.1325708270072937,
      "num_tokens": 3519567.0,
      "step": 1700
    },
    {
      "entropy": 6.05970869064331,
      "epoch": 0.15466255442670537,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999538789823711,
      "loss": 6.0921,
      "mean_token_accuracy": 0.13301847204566003,
      "num_tokens": 3529402.0,
      "step": 1705
    },
    {
      "entropy": 6.176769733428955,
      "epoch": 0.15511611030478956,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999532215506127,
      "loss": 6.0909,
      "mean_token_accuracy": 0.14401522800326347,
      "num_tokens": 3539805.0,
      "step": 1710
    },
    {
      "entropy": 5.969664430618286,
      "epoch": 0.15556966618287374,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999525594667645,
      "loss": 5.9514,
      "mean_token_accuracy": 0.13585785627365113,
      "num_tokens": 3550477.0,
      "step": 1715
    },
    {
      "entropy": 5.984322881698608,
      "epoch": 0.15602322206095792,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00049995189273084,
      "loss": 6.0522,
      "mean_token_accuracy": 0.13302028104662894,
      "num_tokens": 3560729.0,
      "step": 1720
    },
    {
      "entropy": 6.111390686035156,
      "epoch": 0.15647677793904208,
      "grad_norm": 1.046875,
      "learning_rate": 0.000499951221342853,
      "loss": 6.1395,
      "mean_token_accuracy": 0.12391486763954163,
      "num_tokens": 3571538.0,
      "step": 1725
    },
    {
      "entropy": 6.122786998748779,
      "epoch": 0.15693033381712626,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999505453028174,
      "loss": 6.1449,
      "mean_token_accuracy": 0.1329157382249832,
      "num_tokens": 3583128.0,
      "step": 1730
    },
    {
      "entropy": 6.065582609176635,
      "epoch": 0.15738388969521044,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999498646107472,
      "loss": 6.0667,
      "mean_token_accuracy": 0.13652407601475716,
      "num_tokens": 3594358.0,
      "step": 1735
    },
    {
      "entropy": 6.036913299560547,
      "epoch": 0.15783744557329463,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999491792666566,
      "loss": 6.0088,
      "mean_token_accuracy": 0.1374991677701473,
      "num_tokens": 3604345.0,
      "step": 1740
    },
    {
      "entropy": 6.039101409912109,
      "epoch": 0.1582910014513788,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999484892705597,
      "loss": 6.055,
      "mean_token_accuracy": 0.13434040173888206,
      "num_tokens": 3616129.0,
      "step": 1745
    },
    {
      "entropy": 6.09887547492981,
      "epoch": 0.158744557329463,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999477946224705,
      "loss": 6.1052,
      "mean_token_accuracy": 0.13473358303308486,
      "num_tokens": 3627458.0,
      "step": 1750
    },
    {
      "entropy": 6.10147271156311,
      "epoch": 0.15919811320754718,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999470953224037,
      "loss": 5.9868,
      "mean_token_accuracy": 0.1373836562037468,
      "num_tokens": 3636047.0,
      "step": 1755
    },
    {
      "entropy": 5.975709295272827,
      "epoch": 0.15965166908563136,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999463913703735,
      "loss": 5.9717,
      "mean_token_accuracy": 0.13532424792647363,
      "num_tokens": 3646930.0,
      "step": 1760
    },
    {
      "entropy": 6.0145656108856205,
      "epoch": 0.16010522496371554,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999456827663948,
      "loss": 6.0672,
      "mean_token_accuracy": 0.13636473268270494,
      "num_tokens": 3657279.0,
      "step": 1765
    },
    {
      "entropy": 6.016043090820313,
      "epoch": 0.1605587808417997,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000499944969510482,
      "loss": 5.9569,
      "mean_token_accuracy": 0.13713241890072822,
      "num_tokens": 3667070.0,
      "step": 1770
    },
    {
      "entropy": 6.058058595657348,
      "epoch": 0.16101233671988388,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004999442516026498,
      "loss": 5.9892,
      "mean_token_accuracy": 0.13559969365596772,
      "num_tokens": 3677560.0,
      "step": 1775
    },
    {
      "entropy": 5.985709571838379,
      "epoch": 0.16146589259796806,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999435290429133,
      "loss": 5.946,
      "mean_token_accuracy": 0.14185371547937392,
      "num_tokens": 3686701.0,
      "step": 1780
    },
    {
      "entropy": 5.999032068252563,
      "epoch": 0.16191944847605225,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999428018312872,
      "loss": 6.0462,
      "mean_token_accuracy": 0.14263416901230813,
      "num_tokens": 3696338.0,
      "step": 1785
    },
    {
      "entropy": 6.082455444335937,
      "epoch": 0.16237300435413643,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999420699677867,
      "loss": 6.0388,
      "mean_token_accuracy": 0.1367315612733364,
      "num_tokens": 3705752.0,
      "step": 1790
    },
    {
      "entropy": 6.089331150054932,
      "epoch": 0.1628265602322206,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999413334524269,
      "loss": 6.1127,
      "mean_token_accuracy": 0.13379622250795364,
      "num_tokens": 3716225.0,
      "step": 1795
    },
    {
      "entropy": 6.001873111724853,
      "epoch": 0.1632801161103048,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999405922852229,
      "loss": 5.9867,
      "mean_token_accuracy": 0.13304651603102685,
      "num_tokens": 3726593.0,
      "step": 1800
    },
    {
      "entropy": 5.997346496582031,
      "epoch": 0.16373367198838898,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004999398464661901,
      "loss": 6.0555,
      "mean_token_accuracy": 0.13038523942232133,
      "num_tokens": 3737563.0,
      "step": 1805
    },
    {
      "entropy": 6.145504331588745,
      "epoch": 0.16418722786647316,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999390959953441,
      "loss": 6.1043,
      "mean_token_accuracy": 0.13330675959587096,
      "num_tokens": 3747752.0,
      "step": 1810
    },
    {
      "entropy": 5.9703622341156,
      "epoch": 0.16464078374455732,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999383408727002,
      "loss": 6.0364,
      "mean_token_accuracy": 0.13549003675580024,
      "num_tokens": 3758795.0,
      "step": 1815
    },
    {
      "entropy": 6.059307003021241,
      "epoch": 0.1650943396226415,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004999375810982741,
      "loss": 6.0177,
      "mean_token_accuracy": 0.1396925114095211,
      "num_tokens": 3768690.0,
      "step": 1820
    },
    {
      "entropy": 6.01865873336792,
      "epoch": 0.16554789550072568,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999368166720815,
      "loss": 6.0841,
      "mean_token_accuracy": 0.13427145406603813,
      "num_tokens": 3779018.0,
      "step": 1825
    },
    {
      "entropy": 6.140135812759399,
      "epoch": 0.16600145137880987,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999360475941382,
      "loss": 5.9842,
      "mean_token_accuracy": 0.14041623398661612,
      "num_tokens": 3788985.0,
      "step": 1830
    },
    {
      "entropy": 6.026188707351684,
      "epoch": 0.16645500725689405,
      "grad_norm": 1.03125,
      "learning_rate": 0.00049993527386446,
      "loss": 6.1309,
      "mean_token_accuracy": 0.12635482400655745,
      "num_tokens": 3800787.0,
      "step": 1835
    },
    {
      "entropy": 6.024060583114624,
      "epoch": 0.16690856313497823,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999344954830633,
      "loss": 6.0076,
      "mean_token_accuracy": 0.13396736234426498,
      "num_tokens": 3811381.0,
      "step": 1840
    },
    {
      "entropy": 6.089121675491333,
      "epoch": 0.16736211901306242,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999337124499638,
      "loss": 6.1121,
      "mean_token_accuracy": 0.13973291739821433,
      "num_tokens": 3822055.0,
      "step": 1845
    },
    {
      "entropy": 5.998512506484985,
      "epoch": 0.1678156748911466,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999329247651777,
      "loss": 6.0637,
      "mean_token_accuracy": 0.1306314118206501,
      "num_tokens": 3831900.0,
      "step": 1850
    },
    {
      "entropy": 6.045335531234741,
      "epoch": 0.16826923076923078,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999321324287213,
      "loss": 6.0357,
      "mean_token_accuracy": 0.13573757112026213,
      "num_tokens": 3842702.0,
      "step": 1855
    },
    {
      "entropy": 6.1244423389434814,
      "epoch": 0.16872278664731494,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999313354406114,
      "loss": 5.9929,
      "mean_token_accuracy": 0.13379636257886887,
      "num_tokens": 3852690.0,
      "step": 1860
    },
    {
      "entropy": 5.9076738357543945,
      "epoch": 0.16917634252539912,
      "grad_norm": 1.046875,
      "learning_rate": 0.000499930533800864,
      "loss": 5.993,
      "mean_token_accuracy": 0.14352267235517502,
      "num_tokens": 3863040.0,
      "step": 1865
    },
    {
      "entropy": 6.074823522567749,
      "epoch": 0.1696298984034833,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999297275094958,
      "loss": 6.0803,
      "mean_token_accuracy": 0.13035911098122596,
      "num_tokens": 3873613.0,
      "step": 1870
    },
    {
      "entropy": 5.932005929946899,
      "epoch": 0.1700834542815675,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999289165665235,
      "loss": 5.8793,
      "mean_token_accuracy": 0.14385106191039085,
      "num_tokens": 3883936.0,
      "step": 1875
    },
    {
      "entropy": 5.965173816680908,
      "epoch": 0.17053701015965167,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004999281009719639,
      "loss": 5.8597,
      "mean_token_accuracy": 0.14646447375416755,
      "num_tokens": 3893788.0,
      "step": 1880
    },
    {
      "entropy": 5.964988851547242,
      "epoch": 0.17099056603773585,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999272807258339,
      "loss": 5.9165,
      "mean_token_accuracy": 0.14181002825498581,
      "num_tokens": 3903434.0,
      "step": 1885
    },
    {
      "entropy": 5.97712721824646,
      "epoch": 0.17144412191582004,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999264558281504,
      "loss": 5.9377,
      "mean_token_accuracy": 0.14012128487229347,
      "num_tokens": 3913476.0,
      "step": 1890
    },
    {
      "entropy": 5.872003698348999,
      "epoch": 0.17189767779390422,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999256262789305,
      "loss": 5.8922,
      "mean_token_accuracy": 0.14096907377243043,
      "num_tokens": 3923301.0,
      "step": 1895
    },
    {
      "entropy": 5.992045307159424,
      "epoch": 0.1723512336719884,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999247920781913,
      "loss": 6.0344,
      "mean_token_accuracy": 0.13852274641394616,
      "num_tokens": 3933335.0,
      "step": 1900
    },
    {
      "entropy": 6.088766717910767,
      "epoch": 0.17280478955007256,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999239532259502,
      "loss": 6.0646,
      "mean_token_accuracy": 0.1291509412229061,
      "num_tokens": 3944171.0,
      "step": 1905
    },
    {
      "entropy": 6.080254316329956,
      "epoch": 0.17325834542815674,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999231097222244,
      "loss": 6.1305,
      "mean_token_accuracy": 0.13909375816583633,
      "num_tokens": 3954160.0,
      "step": 1910
    },
    {
      "entropy": 6.020297527313232,
      "epoch": 0.17371190130624092,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999222615670312,
      "loss": 5.9253,
      "mean_token_accuracy": 0.14530864879488944,
      "num_tokens": 3964110.0,
      "step": 1915
    },
    {
      "entropy": 6.07570161819458,
      "epoch": 0.1741654571843251,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999214087603885,
      "loss": 6.0159,
      "mean_token_accuracy": 0.13602590560913086,
      "num_tokens": 3974293.0,
      "step": 1920
    },
    {
      "entropy": 5.922182607650757,
      "epoch": 0.1746190130624093,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999205513023135,
      "loss": 5.991,
      "mean_token_accuracy": 0.1404339738190174,
      "num_tokens": 3984346.0,
      "step": 1925
    },
    {
      "entropy": 5.97127251625061,
      "epoch": 0.17507256894049347,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999196891928245,
      "loss": 5.9094,
      "mean_token_accuracy": 0.1484488442540169,
      "num_tokens": 3992897.0,
      "step": 1930
    },
    {
      "entropy": 5.9334413528442385,
      "epoch": 0.17552612481857766,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999188224319388,
      "loss": 5.9525,
      "mean_token_accuracy": 0.1443849354982376,
      "num_tokens": 4002575.0,
      "step": 1935
    },
    {
      "entropy": 6.02183780670166,
      "epoch": 0.17597968069666184,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999179510196746,
      "loss": 6.027,
      "mean_token_accuracy": 0.13418469578027725,
      "num_tokens": 4013752.0,
      "step": 1940
    },
    {
      "entropy": 6.056840991973877,
      "epoch": 0.17643323657474602,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999170749560498,
      "loss": 5.9132,
      "mean_token_accuracy": 0.14471341520547867,
      "num_tokens": 4023718.0,
      "step": 1945
    },
    {
      "entropy": 5.9298981666564945,
      "epoch": 0.17688679245283018,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004999161942410827,
      "loss": 5.9974,
      "mean_token_accuracy": 0.13362503573298454,
      "num_tokens": 4034210.0,
      "step": 1950
    },
    {
      "entropy": 6.088183450698852,
      "epoch": 0.17734034833091436,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999153088747913,
      "loss": 6.0029,
      "mean_token_accuracy": 0.1354498915374279,
      "num_tokens": 4044868.0,
      "step": 1955
    },
    {
      "entropy": 5.887863874435425,
      "epoch": 0.17779390420899854,
      "grad_norm": 1.109375,
      "learning_rate": 0.000499914418857194,
      "loss": 6.0004,
      "mean_token_accuracy": 0.13930529579520226,
      "num_tokens": 4055049.0,
      "step": 1960
    },
    {
      "entropy": 6.098280572891236,
      "epoch": 0.17824746008708273,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999135241883092,
      "loss": 6.0537,
      "mean_token_accuracy": 0.13831883370876313,
      "num_tokens": 4065889.0,
      "step": 1965
    },
    {
      "entropy": 6.039399814605713,
      "epoch": 0.1787010159651669,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999126248681555,
      "loss": 6.0688,
      "mean_token_accuracy": 0.13603949919342995,
      "num_tokens": 4077288.0,
      "step": 1970
    },
    {
      "entropy": 6.006522798538208,
      "epoch": 0.1791545718432511,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999117208967513,
      "loss": 6.0125,
      "mean_token_accuracy": 0.13738583624362946,
      "num_tokens": 4087402.0,
      "step": 1975
    },
    {
      "entropy": 6.079785299301148,
      "epoch": 0.17960812772133528,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999108122741154,
      "loss": 6.0676,
      "mean_token_accuracy": 0.13442711606621743,
      "num_tokens": 4098150.0,
      "step": 1980
    },
    {
      "entropy": 5.926851367950439,
      "epoch": 0.18006168359941946,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999098990002666,
      "loss": 5.9194,
      "mean_token_accuracy": 0.14138512015342714,
      "num_tokens": 4108134.0,
      "step": 1985
    },
    {
      "entropy": 6.07075514793396,
      "epoch": 0.18051523947750364,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999089810752237,
      "loss": 6.0739,
      "mean_token_accuracy": 0.12807230651378632,
      "num_tokens": 4118251.0,
      "step": 1990
    },
    {
      "entropy": 5.945065689086914,
      "epoch": 0.1809687953555878,
      "grad_norm": 1.09375,
      "learning_rate": 0.000499908058499006,
      "loss": 5.9098,
      "mean_token_accuracy": 0.14311220571398736,
      "num_tokens": 4129228.0,
      "step": 1995
    },
    {
      "entropy": 5.959850263595581,
      "epoch": 0.18142235123367198,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999071312716321,
      "loss": 5.9727,
      "mean_token_accuracy": 0.14519958719611167,
      "num_tokens": 4139391.0,
      "step": 2000
    },
    {
      "entropy": 6.003384590148926,
      "epoch": 0.18187590711175616,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004999061993931215,
      "loss": 5.9024,
      "mean_token_accuracy": 0.1432528540492058,
      "num_tokens": 4148891.0,
      "step": 2005
    },
    {
      "entropy": 6.182823896408081,
      "epoch": 0.18232946298984035,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999052628634934,
      "loss": 6.2626,
      "mean_token_accuracy": 0.12997814267873764,
      "num_tokens": 4159551.0,
      "step": 2010
    },
    {
      "entropy": 6.064494609832764,
      "epoch": 0.18278301886792453,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999043216827671,
      "loss": 6.0504,
      "mean_token_accuracy": 0.13733192086219786,
      "num_tokens": 4170055.0,
      "step": 2015
    },
    {
      "entropy": 5.916315412521362,
      "epoch": 0.1832365747460087,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999033758509621,
      "loss": 5.9017,
      "mean_token_accuracy": 0.1404242791235447,
      "num_tokens": 4180767.0,
      "step": 2020
    },
    {
      "entropy": 5.9719846725463865,
      "epoch": 0.1836901306240929,
      "grad_norm": 1.03125,
      "learning_rate": 0.000499902425368098,
      "loss": 6.0294,
      "mean_token_accuracy": 0.1393634244799614,
      "num_tokens": 4190654.0,
      "step": 2025
    },
    {
      "entropy": 5.984382581710816,
      "epoch": 0.18414368650217708,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999014702341944,
      "loss": 5.9548,
      "mean_token_accuracy": 0.13900689631700516,
      "num_tokens": 4200753.0,
      "step": 2030
    },
    {
      "entropy": 6.028017377853393,
      "epoch": 0.18459724238026126,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999005104492712,
      "loss": 6.0411,
      "mean_token_accuracy": 0.1343386575579643,
      "num_tokens": 4210994.0,
      "step": 2035
    },
    {
      "entropy": 5.9925353050231935,
      "epoch": 0.18505079825834542,
      "grad_norm": 1.125,
      "learning_rate": 0.0004998995460133481,
      "loss": 5.9756,
      "mean_token_accuracy": 0.1370024561882019,
      "num_tokens": 4220883.0,
      "step": 2040
    },
    {
      "entropy": 6.067998695373535,
      "epoch": 0.1855043541364296,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004998985769264451,
      "loss": 6.0217,
      "mean_token_accuracy": 0.13243423029780388,
      "num_tokens": 4231269.0,
      "step": 2045
    },
    {
      "entropy": 5.975101900100708,
      "epoch": 0.18595791001451378,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004998976031885821,
      "loss": 6.0222,
      "mean_token_accuracy": 0.1368092842400074,
      "num_tokens": 4241559.0,
      "step": 2050
    },
    {
      "entropy": 5.964649868011475,
      "epoch": 0.18641146589259797,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004998966247997794,
      "loss": 5.8999,
      "mean_token_accuracy": 0.14066432118415834,
      "num_tokens": 4250870.0,
      "step": 2055
    },
    {
      "entropy": 5.934762763977051,
      "epoch": 0.18686502177068215,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004998956417600572,
      "loss": 5.8707,
      "mean_token_accuracy": 0.14538644403219222,
      "num_tokens": 4260848.0,
      "step": 2060
    },
    {
      "entropy": 5.93128457069397,
      "epoch": 0.18731857764876633,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004998946540694358,
      "loss": 5.9914,
      "mean_token_accuracy": 0.14326113611459732,
      "num_tokens": 4270973.0,
      "step": 2065
    },
    {
      "entropy": 6.073555898666382,
      "epoch": 0.18777213352685052,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998936617279357,
      "loss": 6.0634,
      "mean_token_accuracy": 0.13676538169384003,
      "num_tokens": 4281211.0,
      "step": 2070
    },
    {
      "entropy": 5.9082581996917725,
      "epoch": 0.1882256894049347,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998926647355773,
      "loss": 5.9195,
      "mean_token_accuracy": 0.14342071637511253,
      "num_tokens": 4291425.0,
      "step": 2075
    },
    {
      "entropy": 6.059259080886841,
      "epoch": 0.18867924528301888,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998916630923813,
      "loss": 6.0163,
      "mean_token_accuracy": 0.13589237108826638,
      "num_tokens": 4301079.0,
      "step": 2080
    },
    {
      "entropy": 5.99994306564331,
      "epoch": 0.18913280116110304,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004998906567983685,
      "loss": 6.0042,
      "mean_token_accuracy": 0.13732029348611832,
      "num_tokens": 4312374.0,
      "step": 2085
    },
    {
      "entropy": 5.953618669509888,
      "epoch": 0.18958635703918722,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004998896458535595,
      "loss": 5.9273,
      "mean_token_accuracy": 0.1498640865087509,
      "num_tokens": 4322810.0,
      "step": 2090
    },
    {
      "entropy": 5.99668779373169,
      "epoch": 0.1900399129172714,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004998886302579752,
      "loss": 5.9219,
      "mean_token_accuracy": 0.14000264406204224,
      "num_tokens": 4332082.0,
      "step": 2095
    },
    {
      "entropy": 6.013230466842652,
      "epoch": 0.19049346879535559,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000499887610011637,
      "loss": 6.0153,
      "mean_token_accuracy": 0.13956746608018875,
      "num_tokens": 4342632.0,
      "step": 2100
    },
    {
      "entropy": 5.974089479446411,
      "epoch": 0.19094702467343977,
      "grad_norm": 1.25,
      "learning_rate": 0.0004998865851145656,
      "loss": 5.9374,
      "mean_token_accuracy": 0.13679615035653114,
      "num_tokens": 4353544.0,
      "step": 2105
    },
    {
      "entropy": 5.911929941177368,
      "epoch": 0.19140058055152395,
      "grad_norm": 1.25,
      "learning_rate": 0.0004998855555667822,
      "loss": 6.0663,
      "mean_token_accuracy": 0.13838788717985154,
      "num_tokens": 4363636.0,
      "step": 2110
    },
    {
      "entropy": 6.095745468139649,
      "epoch": 0.19185413642960814,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004998845213683084,
      "loss": 5.9418,
      "mean_token_accuracy": 0.13569523617625237,
      "num_tokens": 4373672.0,
      "step": 2115
    },
    {
      "entropy": 5.816546630859375,
      "epoch": 0.19230769230769232,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004998834825191652,
      "loss": 5.9133,
      "mean_token_accuracy": 0.14208602756261826,
      "num_tokens": 4383199.0,
      "step": 2120
    },
    {
      "entropy": 6.015786647796631,
      "epoch": 0.1927612481857765,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004998824390193743,
      "loss": 5.983,
      "mean_token_accuracy": 0.13925845474004744,
      "num_tokens": 4394446.0,
      "step": 2125
    },
    {
      "entropy": 6.1271813869476315,
      "epoch": 0.19321480406386066,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004998813908689573,
      "loss": 6.0262,
      "mean_token_accuracy": 0.13644398227334023,
      "num_tokens": 4405882.0,
      "step": 2130
    },
    {
      "entropy": 5.8932281017303465,
      "epoch": 0.19366835994194484,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004998803380679359,
      "loss": 5.8922,
      "mean_token_accuracy": 0.14466665238142012,
      "num_tokens": 4415280.0,
      "step": 2135
    },
    {
      "entropy": 5.9923975467681885,
      "epoch": 0.19412191582002902,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004998792806163317,
      "loss": 6.0927,
      "mean_token_accuracy": 0.13269788771867752,
      "num_tokens": 4425755.0,
      "step": 2140
    },
    {
      "entropy": 6.003790664672851,
      "epoch": 0.1945754716981132,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004998782185141667,
      "loss": 5.961,
      "mean_token_accuracy": 0.13688114881515503,
      "num_tokens": 4436259.0,
      "step": 2145
    },
    {
      "entropy": 6.004357957839966,
      "epoch": 0.1950290275761974,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004998771517614629,
      "loss": 5.9919,
      "mean_token_accuracy": 0.13306296169757842,
      "num_tokens": 4446423.0,
      "step": 2150
    },
    {
      "entropy": 6.005547332763672,
      "epoch": 0.19548258345428157,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004998760803582422,
      "loss": 5.9289,
      "mean_token_accuracy": 0.14951801002025605,
      "num_tokens": 4456678.0,
      "step": 2155
    },
    {
      "entropy": 5.9408186912536625,
      "epoch": 0.19593613933236576,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004998750043045269,
      "loss": 5.924,
      "mean_token_accuracy": 0.14363801032304763,
      "num_tokens": 4467031.0,
      "step": 2160
    },
    {
      "entropy": 6.0042197704315186,
      "epoch": 0.19638969521044994,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998739236003392,
      "loss": 5.9952,
      "mean_token_accuracy": 0.13384947553277016,
      "num_tokens": 4477771.0,
      "step": 2165
    },
    {
      "entropy": 5.994749736785889,
      "epoch": 0.1968432510885341,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004998728382457014,
      "loss": 6.0132,
      "mean_token_accuracy": 0.13731033354997635,
      "num_tokens": 4488426.0,
      "step": 2170
    },
    {
      "entropy": 6.060128211975098,
      "epoch": 0.19729680696661828,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004998717482406362,
      "loss": 5.9842,
      "mean_token_accuracy": 0.13427486047148704,
      "num_tokens": 4499735.0,
      "step": 2175
    },
    {
      "entropy": 5.945842647552491,
      "epoch": 0.19775036284470246,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004998706535851657,
      "loss": 5.9344,
      "mean_token_accuracy": 0.13740400075912476,
      "num_tokens": 4509500.0,
      "step": 2180
    },
    {
      "entropy": 5.9569385051727295,
      "epoch": 0.19820391872278664,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998695542793129,
      "loss": 5.9392,
      "mean_token_accuracy": 0.13562830910086632,
      "num_tokens": 4520431.0,
      "step": 2185
    },
    {
      "entropy": 5.98747034072876,
      "epoch": 0.19865747460087083,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004998684503231003,
      "loss": 6.0309,
      "mean_token_accuracy": 0.13986979350447654,
      "num_tokens": 4532294.0,
      "step": 2190
    },
    {
      "entropy": 5.934312200546264,
      "epoch": 0.199111030478955,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000499867341716551,
      "loss": 5.948,
      "mean_token_accuracy": 0.1377948857843876,
      "num_tokens": 4542488.0,
      "step": 2195
    },
    {
      "entropy": 5.966859626770019,
      "epoch": 0.1995645863570392,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004998662284596878,
      "loss": 5.8828,
      "mean_token_accuracy": 0.13661633133888246,
      "num_tokens": 4552933.0,
      "step": 2200
    },
    {
      "entropy": 5.952669095993042,
      "epoch": 0.20001814223512338,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004998651105525335,
      "loss": 5.8786,
      "mean_token_accuracy": 0.1433972269296646,
      "num_tokens": 4562373.0,
      "step": 2205
    },
    {
      "entropy": 5.883717727661133,
      "epoch": 0.20047169811320756,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004998639879951115,
      "loss": 6.0106,
      "mean_token_accuracy": 0.13773834556341172,
      "num_tokens": 4572715.0,
      "step": 2210
    },
    {
      "entropy": 6.062313556671143,
      "epoch": 0.2009252539912917,
      "grad_norm": 1.125,
      "learning_rate": 0.000499862860787445,
      "loss": 5.8879,
      "mean_token_accuracy": 0.13721229061484336,
      "num_tokens": 4582782.0,
      "step": 2215
    },
    {
      "entropy": 5.980919361114502,
      "epoch": 0.2013788098693759,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004998617289295571,
      "loss": 5.9939,
      "mean_token_accuracy": 0.13345705717802048,
      "num_tokens": 4594108.0,
      "step": 2220
    },
    {
      "entropy": 6.007939147949219,
      "epoch": 0.20183236574746008,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004998605924214714,
      "loss": 5.9503,
      "mean_token_accuracy": 0.13709166496992112,
      "num_tokens": 4604732.0,
      "step": 2225
    },
    {
      "entropy": 5.984397983551025,
      "epoch": 0.20228592162554426,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004998594512632114,
      "loss": 6.0782,
      "mean_token_accuracy": 0.12725423350930215,
      "num_tokens": 4614998.0,
      "step": 2230
    },
    {
      "entropy": 5.951976251602173,
      "epoch": 0.20273947750362845,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004998583054548006,
      "loss": 5.9087,
      "mean_token_accuracy": 0.13761360049247742,
      "num_tokens": 4624665.0,
      "step": 2235
    },
    {
      "entropy": 5.996311187744141,
      "epoch": 0.20319303338171263,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004998571549962628,
      "loss": 5.9052,
      "mean_token_accuracy": 0.14659312665462493,
      "num_tokens": 4633731.0,
      "step": 2240
    },
    {
      "entropy": 5.99422254562378,
      "epoch": 0.2036465892597968,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004998559998876216,
      "loss": 5.9297,
      "mean_token_accuracy": 0.13845302835106849,
      "num_tokens": 4643421.0,
      "step": 2245
    },
    {
      "entropy": 5.973086881637573,
      "epoch": 0.204100145137881,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004998548401289012,
      "loss": 6.0172,
      "mean_token_accuracy": 0.13639747053384782,
      "num_tokens": 4653795.0,
      "step": 2250
    },
    {
      "entropy": 6.01025357246399,
      "epoch": 0.20455370101596518,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004998536757201253,
      "loss": 5.9909,
      "mean_token_accuracy": 0.13689780980348587,
      "num_tokens": 4664840.0,
      "step": 2255
    },
    {
      "entropy": 6.023299217224121,
      "epoch": 0.20500725689404933,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004998525066613181,
      "loss": 6.0004,
      "mean_token_accuracy": 0.1407467246055603,
      "num_tokens": 4675508.0,
      "step": 2260
    },
    {
      "entropy": 5.931134462356567,
      "epoch": 0.20546081277213352,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004998513329525038,
      "loss": 5.9157,
      "mean_token_accuracy": 0.14364449679851532,
      "num_tokens": 4685995.0,
      "step": 2265
    },
    {
      "entropy": 6.031408452987671,
      "epoch": 0.2059143686502177,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004998501545937066,
      "loss": 5.9397,
      "mean_token_accuracy": 0.13526451364159583,
      "num_tokens": 4696659.0,
      "step": 2270
    },
    {
      "entropy": 5.900203275680542,
      "epoch": 0.20636792452830188,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000499848971584951,
      "loss": 5.8941,
      "mean_token_accuracy": 0.14944595396518706,
      "num_tokens": 4706438.0,
      "step": 2275
    },
    {
      "entropy": 5.911033678054809,
      "epoch": 0.20682148040638607,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004998477839262611,
      "loss": 5.9105,
      "mean_token_accuracy": 0.14088572859764098,
      "num_tokens": 4716402.0,
      "step": 2280
    },
    {
      "entropy": 5.99247899055481,
      "epoch": 0.20727503628447025,
      "grad_norm": 1.109375,
      "learning_rate": 0.000499846591617662,
      "loss": 6.001,
      "mean_token_accuracy": 0.1439184844493866,
      "num_tokens": 4728488.0,
      "step": 2285
    },
    {
      "entropy": 5.915400600433349,
      "epoch": 0.20772859216255443,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004998453946591779,
      "loss": 6.0241,
      "mean_token_accuracy": 0.1419094666838646,
      "num_tokens": 4738693.0,
      "step": 2290
    },
    {
      "entropy": 5.9869701862335205,
      "epoch": 0.20818214804063861,
      "grad_norm": 1.125,
      "learning_rate": 0.0004998441930508339,
      "loss": 5.9132,
      "mean_token_accuracy": 0.1442976713180542,
      "num_tokens": 4748895.0,
      "step": 2295
    },
    {
      "entropy": 6.020207309722901,
      "epoch": 0.2086357039187228,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004998429867926547,
      "loss": 6.0203,
      "mean_token_accuracy": 0.13464172631502153,
      "num_tokens": 4760135.0,
      "step": 2300
    },
    {
      "entropy": 6.043022537231446,
      "epoch": 0.20908925979680695,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004998417758846651,
      "loss": 6.0728,
      "mean_token_accuracy": 0.14229417890310286,
      "num_tokens": 4769935.0,
      "step": 2305
    },
    {
      "entropy": 6.088847494125366,
      "epoch": 0.20954281567489114,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004998405603268903,
      "loss": 5.9793,
      "mean_token_accuracy": 0.14080933183431626,
      "num_tokens": 4779520.0,
      "step": 2310
    },
    {
      "entropy": 5.910649347305298,
      "epoch": 0.20999637155297532,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004998393401193554,
      "loss": 5.9004,
      "mean_token_accuracy": 0.14728002548217772,
      "num_tokens": 4791256.0,
      "step": 2315
    },
    {
      "entropy": 5.814291381835938,
      "epoch": 0.2104499274310595,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004998381152620857,
      "loss": 5.8551,
      "mean_token_accuracy": 0.14871017038822174,
      "num_tokens": 4801334.0,
      "step": 2320
    },
    {
      "entropy": 5.910569524765014,
      "epoch": 0.21090348330914369,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004998368857551067,
      "loss": 5.8394,
      "mean_token_accuracy": 0.145636598020792,
      "num_tokens": 4811361.0,
      "step": 2325
    },
    {
      "entropy": 5.91206111907959,
      "epoch": 0.21135703918722787,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004998356515984433,
      "loss": 5.9207,
      "mean_token_accuracy": 0.14299503937363625,
      "num_tokens": 4821190.0,
      "step": 2330
    },
    {
      "entropy": 5.959600400924683,
      "epoch": 0.21181059506531205,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004998344127921213,
      "loss": 5.9476,
      "mean_token_accuracy": 0.14232189506292342,
      "num_tokens": 4830694.0,
      "step": 2335
    },
    {
      "entropy": 5.944343614578247,
      "epoch": 0.21226415094339623,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004998331693361666,
      "loss": 5.9274,
      "mean_token_accuracy": 0.137859558314085,
      "num_tokens": 4841344.0,
      "step": 2340
    },
    {
      "entropy": 5.9577583312988285,
      "epoch": 0.21271770682148042,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004998319212306046,
      "loss": 5.8928,
      "mean_token_accuracy": 0.14234141260385513,
      "num_tokens": 4851161.0,
      "step": 2345
    },
    {
      "entropy": 5.885667610168457,
      "epoch": 0.21317126269956457,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000499830668475461,
      "loss": 5.8742,
      "mean_token_accuracy": 0.14236741438508033,
      "num_tokens": 4860740.0,
      "step": 2350
    },
    {
      "entropy": 5.944697475433349,
      "epoch": 0.21362481857764876,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000499829411070762,
      "loss": 5.9118,
      "mean_token_accuracy": 0.14243801087141036,
      "num_tokens": 4871733.0,
      "step": 2355
    },
    {
      "entropy": 5.945801448822022,
      "epoch": 0.21407837445573294,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004998281490165335,
      "loss": 5.9669,
      "mean_token_accuracy": 0.13472701087594033,
      "num_tokens": 4883617.0,
      "step": 2360
    },
    {
      "entropy": 5.939307308197021,
      "epoch": 0.21453193033381712,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004998268823128016,
      "loss": 5.9302,
      "mean_token_accuracy": 0.1350124605000019,
      "num_tokens": 4894519.0,
      "step": 2365
    },
    {
      "entropy": 5.977178955078125,
      "epoch": 0.2149854862119013,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004998256109595923,
      "loss": 5.8817,
      "mean_token_accuracy": 0.1414977490901947,
      "num_tokens": 4904965.0,
      "step": 2370
    },
    {
      "entropy": 5.86356201171875,
      "epoch": 0.2154390420899855,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004998243349569323,
      "loss": 5.7913,
      "mean_token_accuracy": 0.1485743671655655,
      "num_tokens": 4914618.0,
      "step": 2375
    },
    {
      "entropy": 5.848833799362183,
      "epoch": 0.21589259796806967,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004998230543048477,
      "loss": 5.8575,
      "mean_token_accuracy": 0.14849701523780823,
      "num_tokens": 4924341.0,
      "step": 2380
    },
    {
      "entropy": 5.913916826248169,
      "epoch": 0.21634615384615385,
      "grad_norm": 1.125,
      "learning_rate": 0.000499821769003365,
      "loss": 5.8344,
      "mean_token_accuracy": 0.14421416819095612,
      "num_tokens": 4934231.0,
      "step": 2385
    },
    {
      "entropy": 5.868072605133056,
      "epoch": 0.21679970972423804,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004998204790525109,
      "loss": 5.8489,
      "mean_token_accuracy": 0.1429114505648613,
      "num_tokens": 4944303.0,
      "step": 2390
    },
    {
      "entropy": 5.908360385894776,
      "epoch": 0.2172532656023222,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004998191844523119,
      "loss": 5.9351,
      "mean_token_accuracy": 0.14515570178627968,
      "num_tokens": 4954571.0,
      "step": 2395
    },
    {
      "entropy": 5.965419578552246,
      "epoch": 0.21770682148040638,
      "grad_norm": 1.203125,
      "learning_rate": 0.000499817885202795,
      "loss": 5.8968,
      "mean_token_accuracy": 0.14096556901931762,
      "num_tokens": 4964418.0,
      "step": 2400
    },
    {
      "entropy": 5.979696035385132,
      "epoch": 0.21816037735849056,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004998165813039869,
      "loss": 5.8835,
      "mean_token_accuracy": 0.14295963644981385,
      "num_tokens": 4975036.0,
      "step": 2405
    },
    {
      "entropy": 5.873979759216309,
      "epoch": 0.21861393323657474,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004998152727559145,
      "loss": 5.9914,
      "mean_token_accuracy": 0.13900846913456916,
      "num_tokens": 4985272.0,
      "step": 2410
    },
    {
      "entropy": 6.005450963973999,
      "epoch": 0.21906748911465893,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000499813959558605,
      "loss": 5.9213,
      "mean_token_accuracy": 0.14803757071495055,
      "num_tokens": 4995187.0,
      "step": 2415
    },
    {
      "entropy": 5.996357822418213,
      "epoch": 0.2195210449927431,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004998126417120856,
      "loss": 5.9681,
      "mean_token_accuracy": 0.14800111576914787,
      "num_tokens": 5005013.0,
      "step": 2420
    },
    {
      "entropy": 6.017085218429566,
      "epoch": 0.2199746008708273,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004998113192163835,
      "loss": 5.9755,
      "mean_token_accuracy": 0.13796148598194122,
      "num_tokens": 5016789.0,
      "step": 2425
    },
    {
      "entropy": 5.947689533233643,
      "epoch": 0.22042815674891147,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000499809992071526,
      "loss": 5.9331,
      "mean_token_accuracy": 0.14225904271006584,
      "num_tokens": 5026913.0,
      "step": 2430
    },
    {
      "entropy": 5.922537040710449,
      "epoch": 0.22088171262699566,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004998086602775406,
      "loss": 6.0365,
      "mean_token_accuracy": 0.13978781476616858,
      "num_tokens": 5037501.0,
      "step": 2435
    },
    {
      "entropy": 5.976426696777343,
      "epoch": 0.2213352685050798,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004998073238344547,
      "loss": 5.8812,
      "mean_token_accuracy": 0.13661929443478585,
      "num_tokens": 5048478.0,
      "step": 2440
    },
    {
      "entropy": 5.931379604339599,
      "epoch": 0.221788824383164,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004998059827422962,
      "loss": 5.9411,
      "mean_token_accuracy": 0.1370401792228222,
      "num_tokens": 5059118.0,
      "step": 2445
    },
    {
      "entropy": 5.984925699234009,
      "epoch": 0.22224238026124818,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004998046370010926,
      "loss": 5.8956,
      "mean_token_accuracy": 0.14508832097053528,
      "num_tokens": 5070180.0,
      "step": 2450
    },
    {
      "entropy": 5.882190847396851,
      "epoch": 0.22269593613933236,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004998032866108718,
      "loss": 5.8958,
      "mean_token_accuracy": 0.14208046942949296,
      "num_tokens": 5079912.0,
      "step": 2455
    },
    {
      "entropy": 5.956144332885742,
      "epoch": 0.22314949201741655,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004998019315716618,
      "loss": 5.9767,
      "mean_token_accuracy": 0.14252351969480515,
      "num_tokens": 5090549.0,
      "step": 2460
    },
    {
      "entropy": 5.902889060974121,
      "epoch": 0.22360304789550073,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004998005718834905,
      "loss": 5.8293,
      "mean_token_accuracy": 0.15000268071889877,
      "num_tokens": 5100880.0,
      "step": 2465
    },
    {
      "entropy": 5.912509965896606,
      "epoch": 0.2240566037735849,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004997992075463862,
      "loss": 5.9424,
      "mean_token_accuracy": 0.1433528684079647,
      "num_tokens": 5111796.0,
      "step": 2470
    },
    {
      "entropy": 5.958014440536499,
      "epoch": 0.2245101596516691,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004997978385603769,
      "loss": 5.935,
      "mean_token_accuracy": 0.14096981883049012,
      "num_tokens": 5123468.0,
      "step": 2475
    },
    {
      "entropy": 5.9337944984436035,
      "epoch": 0.22496371552975328,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004997964649254911,
      "loss": 5.8641,
      "mean_token_accuracy": 0.1491854891180992,
      "num_tokens": 5132925.0,
      "step": 2480
    },
    {
      "entropy": 6.011470174789428,
      "epoch": 0.22541727140783743,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004997950866417572,
      "loss": 5.9448,
      "mean_token_accuracy": 0.1389354422688484,
      "num_tokens": 5143445.0,
      "step": 2485
    },
    {
      "entropy": 5.92656307220459,
      "epoch": 0.22587082728592162,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004997937037092034,
      "loss": 5.8551,
      "mean_token_accuracy": 0.13968046829104425,
      "num_tokens": 5153176.0,
      "step": 2490
    },
    {
      "entropy": 5.953498363494873,
      "epoch": 0.2263243831640058,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004997923161278587,
      "loss": 5.9041,
      "mean_token_accuracy": 0.1428646430373192,
      "num_tokens": 5162765.0,
      "step": 2495
    },
    {
      "entropy": 5.978867244720459,
      "epoch": 0.22677793904208998,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004997909238977514,
      "loss": 5.9606,
      "mean_token_accuracy": 0.14364035278558732,
      "num_tokens": 5173390.0,
      "step": 2500
    },
    {
      "entropy": 5.906553840637207,
      "epoch": 0.22723149492017417,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004997895270189108,
      "loss": 5.8628,
      "mean_token_accuracy": 0.1373254232108593,
      "num_tokens": 5184212.0,
      "step": 2505
    },
    {
      "entropy": 5.853435754776001,
      "epoch": 0.22768505079825835,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004997881254913653,
      "loss": 5.8602,
      "mean_token_accuracy": 0.13991171047091483,
      "num_tokens": 5194311.0,
      "step": 2510
    },
    {
      "entropy": 5.888242435455322,
      "epoch": 0.22813860667634253,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004997867193151441,
      "loss": 5.8408,
      "mean_token_accuracy": 0.1456891916692257,
      "num_tokens": 5204708.0,
      "step": 2515
    },
    {
      "entropy": 5.830229997634888,
      "epoch": 0.22859216255442671,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004997853084902763,
      "loss": 5.843,
      "mean_token_accuracy": 0.14071134552359582,
      "num_tokens": 5214566.0,
      "step": 2520
    },
    {
      "entropy": 5.903006267547608,
      "epoch": 0.2290457184325109,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004997838930167911,
      "loss": 5.8779,
      "mean_token_accuracy": 0.14581192210316657,
      "num_tokens": 5224762.0,
      "step": 2525
    },
    {
      "entropy": 5.952599906921387,
      "epoch": 0.22949927431059505,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004997824728947176,
      "loss": 5.9372,
      "mean_token_accuracy": 0.13909635022282602,
      "num_tokens": 5235146.0,
      "step": 2530
    },
    {
      "entropy": 5.936500406265258,
      "epoch": 0.22995283018867924,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004997810481240853,
      "loss": 5.9115,
      "mean_token_accuracy": 0.1376205012202263,
      "num_tokens": 5243867.0,
      "step": 2535
    },
    {
      "entropy": 5.910394525527954,
      "epoch": 0.23040638606676342,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004997796187049238,
      "loss": 5.831,
      "mean_token_accuracy": 0.14185473769903184,
      "num_tokens": 5254279.0,
      "step": 2540
    },
    {
      "entropy": 6.032854747772217,
      "epoch": 0.2308599419448476,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004997781846372623,
      "loss": 5.9444,
      "mean_token_accuracy": 0.1413105957210064,
      "num_tokens": 5264769.0,
      "step": 2545
    },
    {
      "entropy": 5.892349815368652,
      "epoch": 0.23131349782293179,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004997767459211307,
      "loss": 5.8757,
      "mean_token_accuracy": 0.1384491316974163,
      "num_tokens": 5274291.0,
      "step": 2550
    },
    {
      "entropy": 5.9222578525543215,
      "epoch": 0.23176705370101597,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004997753025565587,
      "loss": 5.8226,
      "mean_token_accuracy": 0.14413735643029213,
      "num_tokens": 5284366.0,
      "step": 2555
    },
    {
      "entropy": 5.923757791519165,
      "epoch": 0.23222060957910015,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004997738545435763,
      "loss": 5.9218,
      "mean_token_accuracy": 0.14310500398278236,
      "num_tokens": 5294494.0,
      "step": 2560
    },
    {
      "entropy": 5.835726594924926,
      "epoch": 0.23267416545718433,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004997724018822133,
      "loss": 5.9402,
      "mean_token_accuracy": 0.14283815771341324,
      "num_tokens": 5305606.0,
      "step": 2565
    },
    {
      "entropy": 5.938771772384643,
      "epoch": 0.23312772133526852,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004997709445724996,
      "loss": 5.9326,
      "mean_token_accuracy": 0.13846345022320747,
      "num_tokens": 5315677.0,
      "step": 2570
    },
    {
      "entropy": 6.001786565780639,
      "epoch": 0.23358127721335267,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004997694826144656,
      "loss": 5.987,
      "mean_token_accuracy": 0.13663317039608955,
      "num_tokens": 5326453.0,
      "step": 2575
    },
    {
      "entropy": 5.945255422592163,
      "epoch": 0.23403483309143686,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004997680160081414,
      "loss": 5.8746,
      "mean_token_accuracy": 0.14533428698778153,
      "num_tokens": 5337591.0,
      "step": 2580
    },
    {
      "entropy": 5.880674076080322,
      "epoch": 0.23448838896952104,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004997665447535573,
      "loss": 5.8792,
      "mean_token_accuracy": 0.14644460529088973,
      "num_tokens": 5349400.0,
      "step": 2585
    },
    {
      "entropy": 5.976725244522095,
      "epoch": 0.23494194484760522,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004997650688507438,
      "loss": 5.8647,
      "mean_token_accuracy": 0.14361676946282387,
      "num_tokens": 5359984.0,
      "step": 2590
    },
    {
      "entropy": 5.860870790481568,
      "epoch": 0.2353955007256894,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004997635882997314,
      "loss": 5.9341,
      "mean_token_accuracy": 0.14296606034040452,
      "num_tokens": 5369254.0,
      "step": 2595
    },
    {
      "entropy": 5.9094048023223875,
      "epoch": 0.2358490566037736,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004997621031005507,
      "loss": 5.8609,
      "mean_token_accuracy": 0.14291658252477646,
      "num_tokens": 5379648.0,
      "step": 2600
    },
    {
      "entropy": 5.866974496841431,
      "epoch": 0.23630261248185777,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004997606132532325,
      "loss": 5.8309,
      "mean_token_accuracy": 0.14787738099694253,
      "num_tokens": 5390321.0,
      "step": 2605
    },
    {
      "entropy": 5.867215776443482,
      "epoch": 0.23675616835994195,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004997591187578075,
      "loss": 5.8596,
      "mean_token_accuracy": 0.14872283935546876,
      "num_tokens": 5401078.0,
      "step": 2610
    },
    {
      "entropy": 5.8993391513824465,
      "epoch": 0.23720972423802614,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004997576196143067,
      "loss": 5.8513,
      "mean_token_accuracy": 0.1461376279592514,
      "num_tokens": 5411977.0,
      "step": 2615
    },
    {
      "entropy": 5.88139500617981,
      "epoch": 0.2376632801161103,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000499756115822761,
      "loss": 5.841,
      "mean_token_accuracy": 0.13903868198394775,
      "num_tokens": 5422967.0,
      "step": 2620
    },
    {
      "entropy": 5.967788743972778,
      "epoch": 0.23811683599419448,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004997546073832015,
      "loss": 5.8848,
      "mean_token_accuracy": 0.15244564265012742,
      "num_tokens": 5432260.0,
      "step": 2625
    },
    {
      "entropy": 5.894135141372681,
      "epoch": 0.23857039187227866,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004997530942956595,
      "loss": 5.8465,
      "mean_token_accuracy": 0.14635367840528488,
      "num_tokens": 5441706.0,
      "step": 2630
    },
    {
      "entropy": 5.897868919372558,
      "epoch": 0.23902394775036284,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004997515765601662,
      "loss": 5.8558,
      "mean_token_accuracy": 0.15220582336187363,
      "num_tokens": 5452397.0,
      "step": 2635
    },
    {
      "entropy": 5.880718660354614,
      "epoch": 0.23947750362844702,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004997500541767531,
      "loss": 5.8202,
      "mean_token_accuracy": 0.14415100812911988,
      "num_tokens": 5463762.0,
      "step": 2640
    },
    {
      "entropy": 5.870383977890015,
      "epoch": 0.2399310595065312,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004997485271454516,
      "loss": 5.8788,
      "mean_token_accuracy": 0.13774784579873084,
      "num_tokens": 5474622.0,
      "step": 2645
    },
    {
      "entropy": 5.96224799156189,
      "epoch": 0.2403846153846154,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004997469954662933,
      "loss": 5.8695,
      "mean_token_accuracy": 0.14564383402466774,
      "num_tokens": 5484697.0,
      "step": 2650
    },
    {
      "entropy": 5.851238155364991,
      "epoch": 0.24083817126269957,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004997454591393097,
      "loss": 5.8667,
      "mean_token_accuracy": 0.1427861526608467,
      "num_tokens": 5496433.0,
      "step": 2655
    },
    {
      "entropy": 5.973064804077149,
      "epoch": 0.24129172714078376,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004997439181645329,
      "loss": 5.97,
      "mean_token_accuracy": 0.14119771718978882,
      "num_tokens": 5507917.0,
      "step": 2660
    },
    {
      "entropy": 6.047526073455811,
      "epoch": 0.2417452830188679,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004997423725419945,
      "loss": 5.9544,
      "mean_token_accuracy": 0.1415584534406662,
      "num_tokens": 5519159.0,
      "step": 2665
    },
    {
      "entropy": 5.842634201049805,
      "epoch": 0.2421988388969521,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004997408222717267,
      "loss": 5.9242,
      "mean_token_accuracy": 0.14665919542312622,
      "num_tokens": 5529180.0,
      "step": 2670
    },
    {
      "entropy": 5.897845602035522,
      "epoch": 0.24265239477503628,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004997392673537612,
      "loss": 5.8381,
      "mean_token_accuracy": 0.15111086070537566,
      "num_tokens": 5539363.0,
      "step": 2675
    },
    {
      "entropy": 5.8370285987854,
      "epoch": 0.24310595065312046,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004997377077881306,
      "loss": 5.946,
      "mean_token_accuracy": 0.14330666437745093,
      "num_tokens": 5549841.0,
      "step": 2680
    },
    {
      "entropy": 5.904944038391113,
      "epoch": 0.24355950653120464,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004997361435748668,
      "loss": 5.8349,
      "mean_token_accuracy": 0.14796256124973298,
      "num_tokens": 5560299.0,
      "step": 2685
    },
    {
      "entropy": 5.9934006690979,
      "epoch": 0.24401306240928883,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004997345747140024,
      "loss": 5.8923,
      "mean_token_accuracy": 0.14425792098045348,
      "num_tokens": 5571081.0,
      "step": 2690
    },
    {
      "entropy": 5.856541490554809,
      "epoch": 0.244466618287373,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004997330012055695,
      "loss": 5.8447,
      "mean_token_accuracy": 0.149576073884964,
      "num_tokens": 5581389.0,
      "step": 2695
    },
    {
      "entropy": 5.9023360252380375,
      "epoch": 0.2449201741654572,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004997314230496011,
      "loss": 5.8683,
      "mean_token_accuracy": 0.14250419288873672,
      "num_tokens": 5592347.0,
      "step": 2700
    },
    {
      "entropy": 5.878292608261108,
      "epoch": 0.24537373004354138,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004997298402461296,
      "loss": 5.8384,
      "mean_token_accuracy": 0.14773763567209244,
      "num_tokens": 5603411.0,
      "step": 2705
    },
    {
      "entropy": 5.920032787322998,
      "epoch": 0.24582728592162553,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004997282527951876,
      "loss": 5.844,
      "mean_token_accuracy": 0.14941513240337373,
      "num_tokens": 5614615.0,
      "step": 2710
    },
    {
      "entropy": 5.869028520584107,
      "epoch": 0.24628084179970972,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004997266606968081,
      "loss": 5.8137,
      "mean_token_accuracy": 0.1474539414048195,
      "num_tokens": 5624582.0,
      "step": 2715
    },
    {
      "entropy": 5.876409101486206,
      "epoch": 0.2467343976777939,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004997250639510241,
      "loss": 5.8213,
      "mean_token_accuracy": 0.14212239235639573,
      "num_tokens": 5634310.0,
      "step": 2720
    },
    {
      "entropy": 5.889693927764893,
      "epoch": 0.24718795355587808,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004997234625578684,
      "loss": 5.9165,
      "mean_token_accuracy": 0.14168269485235213,
      "num_tokens": 5644018.0,
      "step": 2725
    },
    {
      "entropy": 5.872516393661499,
      "epoch": 0.24764150943396226,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004997218565173742,
      "loss": 5.8501,
      "mean_token_accuracy": 0.1477284774184227,
      "num_tokens": 5653592.0,
      "step": 2730
    },
    {
      "entropy": 5.970296859741211,
      "epoch": 0.24809506531204645,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004997202458295748,
      "loss": 5.8983,
      "mean_token_accuracy": 0.14910974949598313,
      "num_tokens": 5662812.0,
      "step": 2735
    },
    {
      "entropy": 5.877613735198975,
      "epoch": 0.24854862119013063,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004997186304945034,
      "loss": 5.8589,
      "mean_token_accuracy": 0.14394548833370208,
      "num_tokens": 5672939.0,
      "step": 2740
    },
    {
      "entropy": 5.869716453552246,
      "epoch": 0.24900217706821481,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004997170105121935,
      "loss": 5.8577,
      "mean_token_accuracy": 0.14621407687664031,
      "num_tokens": 5683350.0,
      "step": 2745
    },
    {
      "entropy": 6.0468605041503904,
      "epoch": 0.249455732946299,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004997153858826786,
      "loss": 5.9484,
      "mean_token_accuracy": 0.13753940612077714,
      "num_tokens": 5694516.0,
      "step": 2750
    },
    {
      "entropy": 5.836617755889892,
      "epoch": 0.24990928882438315,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004997137566059922,
      "loss": 5.7658,
      "mean_token_accuracy": 0.1503409966826439,
      "num_tokens": 5705443.0,
      "step": 2755
    },
    {
      "entropy": 5.8198834419250485,
      "epoch": 0.25036284470246734,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004997121226821681,
      "loss": 5.8859,
      "mean_token_accuracy": 0.13909180611371993,
      "num_tokens": 5715933.0,
      "step": 2760
    },
    {
      "entropy": 5.972828674316406,
      "epoch": 0.25081640058055155,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00049971048411124,
      "loss": 5.9315,
      "mean_token_accuracy": 0.1365823417901993,
      "num_tokens": 5726630.0,
      "step": 2765
    },
    {
      "entropy": 6.025601530075074,
      "epoch": 0.2512699564586357,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004997088408932418,
      "loss": 5.8606,
      "mean_token_accuracy": 0.14626222401857375,
      "num_tokens": 5736890.0,
      "step": 2770
    },
    {
      "entropy": 5.7838927745819095,
      "epoch": 0.25172351233671986,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004997071930282076,
      "loss": 5.7814,
      "mean_token_accuracy": 0.15587268024683,
      "num_tokens": 5747042.0,
      "step": 2775
    },
    {
      "entropy": 5.836866760253907,
      "epoch": 0.25217706821480407,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004997055405161712,
      "loss": 5.904,
      "mean_token_accuracy": 0.14357635378837585,
      "num_tokens": 5757367.0,
      "step": 2780
    },
    {
      "entropy": 5.944021129608155,
      "epoch": 0.2526306240928882,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004997038833571671,
      "loss": 5.9615,
      "mean_token_accuracy": 0.1422647625207901,
      "num_tokens": 5769557.0,
      "step": 2785
    },
    {
      "entropy": 5.940666151046753,
      "epoch": 0.25308417997097243,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004997022215512295,
      "loss": 5.8503,
      "mean_token_accuracy": 0.15143156200647354,
      "num_tokens": 5780615.0,
      "step": 2790
    },
    {
      "entropy": 5.937691497802734,
      "epoch": 0.2535377358490566,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004997005550983925,
      "loss": 5.827,
      "mean_token_accuracy": 0.14720471799373627,
      "num_tokens": 5791271.0,
      "step": 2795
    },
    {
      "entropy": 5.830975341796875,
      "epoch": 0.2539912917271408,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000499698883998691,
      "loss": 5.7719,
      "mean_token_accuracy": 0.14439669102430344,
      "num_tokens": 5801117.0,
      "step": 2800
    },
    {
      "entropy": 5.828149271011353,
      "epoch": 0.25444484760522496,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004996972082521591,
      "loss": 5.7958,
      "mean_token_accuracy": 0.1420363336801529,
      "num_tokens": 5811696.0,
      "step": 2805
    },
    {
      "entropy": 5.868321561813355,
      "epoch": 0.25489840348330917,
      "grad_norm": 1.25,
      "learning_rate": 0.0004996955278588319,
      "loss": 5.8318,
      "mean_token_accuracy": 0.15537209510803224,
      "num_tokens": 5822207.0,
      "step": 2810
    },
    {
      "entropy": 5.808701229095459,
      "epoch": 0.2553519593613933,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004996938428187437,
      "loss": 5.7714,
      "mean_token_accuracy": 0.15279726982116698,
      "num_tokens": 5831927.0,
      "step": 2815
    },
    {
      "entropy": 5.886981678009033,
      "epoch": 0.2558055152394775,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004996921531319297,
      "loss": 5.8415,
      "mean_token_accuracy": 0.14455378502607347,
      "num_tokens": 5842093.0,
      "step": 2820
    },
    {
      "entropy": 5.86587119102478,
      "epoch": 0.2562590711175617,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004996904587984246,
      "loss": 5.7803,
      "mean_token_accuracy": 0.15339562743902208,
      "num_tokens": 5850551.0,
      "step": 2825
    },
    {
      "entropy": 5.764548873901367,
      "epoch": 0.25671262699564584,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004996887598182636,
      "loss": 5.7714,
      "mean_token_accuracy": 0.1484563708305359,
      "num_tokens": 5860651.0,
      "step": 2830
    },
    {
      "entropy": 5.840695333480835,
      "epoch": 0.25716618287373005,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004996870561914819,
      "loss": 5.7878,
      "mean_token_accuracy": 0.14319953173398972,
      "num_tokens": 5871500.0,
      "step": 2835
    },
    {
      "entropy": 5.857464742660523,
      "epoch": 0.2576197387518142,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004996853479181145,
      "loss": 5.7638,
      "mean_token_accuracy": 0.1445399969816208,
      "num_tokens": 5881523.0,
      "step": 2840
    },
    {
      "entropy": 5.928007984161377,
      "epoch": 0.2580732946298984,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004996836349981969,
      "loss": 5.9088,
      "mean_token_accuracy": 0.14473725184798242,
      "num_tokens": 5891162.0,
      "step": 2845
    },
    {
      "entropy": 5.846505928039551,
      "epoch": 0.2585268505079826,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004996819174317645,
      "loss": 5.7977,
      "mean_token_accuracy": 0.14666571021080016,
      "num_tokens": 5901818.0,
      "step": 2850
    },
    {
      "entropy": 5.829140758514404,
      "epoch": 0.2589804063860668,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004996801952188528,
      "loss": 5.895,
      "mean_token_accuracy": 0.14497522860765458,
      "num_tokens": 5913889.0,
      "step": 2855
    },
    {
      "entropy": 5.888280963897705,
      "epoch": 0.25943396226415094,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004996784683594973,
      "loss": 5.8485,
      "mean_token_accuracy": 0.14567365497350693,
      "num_tokens": 5923429.0,
      "step": 2860
    },
    {
      "entropy": 5.998221731185913,
      "epoch": 0.2598875181422351,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004996767368537338,
      "loss": 5.8844,
      "mean_token_accuracy": 0.14624838829040526,
      "num_tokens": 5933847.0,
      "step": 2865
    },
    {
      "entropy": 5.8100090503692625,
      "epoch": 0.2603410740203193,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004996750007015983,
      "loss": 5.7683,
      "mean_token_accuracy": 0.14520363360643387,
      "num_tokens": 5943874.0,
      "step": 2870
    },
    {
      "entropy": 5.848591136932373,
      "epoch": 0.26079462989840346,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004996732599031264,
      "loss": 5.8297,
      "mean_token_accuracy": 0.14542661756277084,
      "num_tokens": 5953653.0,
      "step": 2875
    },
    {
      "entropy": 5.917650747299194,
      "epoch": 0.2612481857764877,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004996715144583543,
      "loss": 5.7734,
      "mean_token_accuracy": 0.14755018949508666,
      "num_tokens": 5964039.0,
      "step": 2880
    },
    {
      "entropy": 5.900532245635986,
      "epoch": 0.26170174165457183,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000499669764367318,
      "loss": 5.9012,
      "mean_token_accuracy": 0.14746888726949692,
      "num_tokens": 5974083.0,
      "step": 2885
    },
    {
      "entropy": 5.8273969173431395,
      "epoch": 0.26215529753265604,
      "grad_norm": 1.25,
      "learning_rate": 0.0004996680096300537,
      "loss": 5.8462,
      "mean_token_accuracy": 0.14346672594547272,
      "num_tokens": 5984629.0,
      "step": 2890
    },
    {
      "entropy": 5.839718961715699,
      "epoch": 0.2626088534107402,
      "grad_norm": 1.5,
      "learning_rate": 0.0004996662502465978,
      "loss": 5.7524,
      "mean_token_accuracy": 0.16117057502269744,
      "num_tokens": 5994295.0,
      "step": 2895
    },
    {
      "entropy": 5.838645267486572,
      "epoch": 0.2630624092888244,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004996644862169864,
      "loss": 5.7866,
      "mean_token_accuracy": 0.14446460008621215,
      "num_tokens": 6004156.0,
      "step": 2900
    },
    {
      "entropy": 5.901477813720703,
      "epoch": 0.26351596516690856,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004996627175412563,
      "loss": 5.8837,
      "mean_token_accuracy": 0.14486180245876312,
      "num_tokens": 6014188.0,
      "step": 2905
    },
    {
      "entropy": 5.949041032791138,
      "epoch": 0.2639695210449927,
      "grad_norm": 1.21875,
      "learning_rate": 0.000499660944219444,
      "loss": 5.9405,
      "mean_token_accuracy": 0.1410454109311104,
      "num_tokens": 6024879.0,
      "step": 2910
    },
    {
      "entropy": 5.967418909072876,
      "epoch": 0.2644230769230769,
      "grad_norm": 1.125,
      "learning_rate": 0.0004996591662515861,
      "loss": 5.9944,
      "mean_token_accuracy": 0.13865989670157433,
      "num_tokens": 6036374.0,
      "step": 2915
    },
    {
      "entropy": 5.901403713226318,
      "epoch": 0.2648766328011611,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004996573836377193,
      "loss": 5.8373,
      "mean_token_accuracy": 0.14499613642692566,
      "num_tokens": 6046432.0,
      "step": 2920
    },
    {
      "entropy": 5.943468523025513,
      "epoch": 0.2653301886792453,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004996555963778806,
      "loss": 5.8592,
      "mean_token_accuracy": 0.14138642624020575,
      "num_tokens": 6056864.0,
      "step": 2925
    },
    {
      "entropy": 5.995022583007812,
      "epoch": 0.26578374455732945,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000499653804472107,
      "loss": 5.8844,
      "mean_token_accuracy": 0.15080390572547914,
      "num_tokens": 6067294.0,
      "step": 2930
    },
    {
      "entropy": 5.905005168914795,
      "epoch": 0.26623730043541366,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004996520079204353,
      "loss": 5.8871,
      "mean_token_accuracy": 0.14075151458382607,
      "num_tokens": 6077587.0,
      "step": 2935
    },
    {
      "entropy": 5.918993663787842,
      "epoch": 0.2666908563134978,
      "grad_norm": 1.28125,
      "learning_rate": 0.000499650206722903,
      "loss": 5.8385,
      "mean_token_accuracy": 0.14215887933969498,
      "num_tokens": 6089306.0,
      "step": 2940
    },
    {
      "entropy": 5.827713394165039,
      "epoch": 0.267144412191582,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004996484008795471,
      "loss": 5.7875,
      "mean_token_accuracy": 0.1460879400372505,
      "num_tokens": 6099651.0,
      "step": 2945
    },
    {
      "entropy": 5.838403844833374,
      "epoch": 0.2675979680696662,
      "grad_norm": 1.34375,
      "learning_rate": 0.000499646590390405,
      "loss": 5.7932,
      "mean_token_accuracy": 0.14292585104703903,
      "num_tokens": 6110063.0,
      "step": 2950
    },
    {
      "entropy": 5.874997091293335,
      "epoch": 0.26805152394775034,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004996447752555142,
      "loss": 5.8194,
      "mean_token_accuracy": 0.1502259336411953,
      "num_tokens": 6120031.0,
      "step": 2955
    },
    {
      "entropy": 5.802283811569214,
      "epoch": 0.26850507982583455,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004996429554749122,
      "loss": 5.8655,
      "mean_token_accuracy": 0.15023885816335678,
      "num_tokens": 6131274.0,
      "step": 2960
    },
    {
      "entropy": 5.964536857604981,
      "epoch": 0.2689586357039187,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004996411310486367,
      "loss": 5.8932,
      "mean_token_accuracy": 0.1471738301217556,
      "num_tokens": 6140878.0,
      "step": 2965
    },
    {
      "entropy": 5.937205791473389,
      "epoch": 0.2694121915820029,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004996393019767253,
      "loss": 5.8875,
      "mean_token_accuracy": 0.14035830050706863,
      "num_tokens": 6152229.0,
      "step": 2970
    },
    {
      "entropy": 5.76689043045044,
      "epoch": 0.26986574746008707,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004996374682592158,
      "loss": 5.7879,
      "mean_token_accuracy": 0.14796984195709229,
      "num_tokens": 6162311.0,
      "step": 2975
    },
    {
      "entropy": 5.934810733795166,
      "epoch": 0.2703193033381713,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004996356298961463,
      "loss": 5.9202,
      "mean_token_accuracy": 0.14373642206192017,
      "num_tokens": 6172806.0,
      "step": 2980
    },
    {
      "entropy": 5.891864681243897,
      "epoch": 0.27077285921625543,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004996337868875549,
      "loss": 5.879,
      "mean_token_accuracy": 0.14369452446699144,
      "num_tokens": 6183479.0,
      "step": 2985
    },
    {
      "entropy": 5.89184536933899,
      "epoch": 0.27122641509433965,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004996319392334792,
      "loss": 5.8281,
      "mean_token_accuracy": 0.14598268270492554,
      "num_tokens": 6194201.0,
      "step": 2990
    },
    {
      "entropy": 5.824180364608765,
      "epoch": 0.2716799709724238,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000499630086933958,
      "loss": 5.7868,
      "mean_token_accuracy": 0.14525872468948364,
      "num_tokens": 6204774.0,
      "step": 2995
    },
    {
      "entropy": 5.86897759437561,
      "epoch": 0.27213352685050796,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004996282299890292,
      "loss": 5.7866,
      "mean_token_accuracy": 0.1425388991832733,
      "num_tokens": 6215056.0,
      "step": 3000
    },
    {
      "epoch": 0.27213352685050796,
      "eval_entropy": 5.760963649555569,
      "eval_loss": 5.879345417022705,
      "eval_mean_token_accuracy": 0.15095411169330053,
      "eval_num_tokens": 6215056.0,
      "eval_runtime": 26.6633,
      "eval_samples_per_second": 1326.167,
      "eval_steps_per_second": 165.771,
      "step": 3000
    },
    {
      "entropy": 5.954884576797485,
      "epoch": 0.27258708272859217,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004996263683987316,
      "loss": 5.8977,
      "mean_token_accuracy": 0.14913439005613327,
      "num_tokens": 6226264.0,
      "step": 3005
    },
    {
      "entropy": 5.886638212203979,
      "epoch": 0.2730406386066763,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004996245021631033,
      "loss": 5.8156,
      "mean_token_accuracy": 0.14778477251529692,
      "num_tokens": 6235137.0,
      "step": 3010
    },
    {
      "entropy": 5.848877334594727,
      "epoch": 0.27349419448476053,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000499622631282183,
      "loss": 5.8335,
      "mean_token_accuracy": 0.1477284774184227,
      "num_tokens": 6245411.0,
      "step": 3015
    },
    {
      "entropy": 5.880431079864502,
      "epoch": 0.2739477503628447,
      "grad_norm": 1.375,
      "learning_rate": 0.0004996207557560095,
      "loss": 5.9158,
      "mean_token_accuracy": 0.14132191836833954,
      "num_tokens": 6255633.0,
      "step": 3020
    },
    {
      "entropy": 5.930035352706909,
      "epoch": 0.2744013062409289,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004996188755846216,
      "loss": 5.8541,
      "mean_token_accuracy": 0.14648665934801103,
      "num_tokens": 6265109.0,
      "step": 3025
    },
    {
      "entropy": 5.880398845672607,
      "epoch": 0.27485486211901305,
      "grad_norm": 1.359375,
      "learning_rate": 0.000499616990768058,
      "loss": 5.9021,
      "mean_token_accuracy": 0.1377116046845913,
      "num_tokens": 6275241.0,
      "step": 3030
    },
    {
      "entropy": 5.896439170837402,
      "epoch": 0.27530841799709727,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004996151013063579,
      "loss": 5.9281,
      "mean_token_accuracy": 0.13803014755249024,
      "num_tokens": 6286603.0,
      "step": 3035
    },
    {
      "entropy": 5.930904483795166,
      "epoch": 0.2757619738751814,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004996132071995601,
      "loss": 5.7778,
      "mean_token_accuracy": 0.15983342826366426,
      "num_tokens": 6297219.0,
      "step": 3040
    },
    {
      "entropy": 5.833100652694702,
      "epoch": 0.2762155297532656,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004996113084477042,
      "loss": 5.8305,
      "mean_token_accuracy": 0.15047609359025954,
      "num_tokens": 6307694.0,
      "step": 3045
    },
    {
      "entropy": 5.810063695907592,
      "epoch": 0.2766690856313498,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004996094050508291,
      "loss": 5.7577,
      "mean_token_accuracy": 0.15339901745319368,
      "num_tokens": 6317384.0,
      "step": 3050
    },
    {
      "entropy": 5.917490768432617,
      "epoch": 0.27712264150943394,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004996074970089741,
      "loss": 5.9031,
      "mean_token_accuracy": 0.1421275481581688,
      "num_tokens": 6327558.0,
      "step": 3055
    },
    {
      "entropy": 5.8848024845123295,
      "epoch": 0.27757619738751815,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004996055843221789,
      "loss": 5.7545,
      "mean_token_accuracy": 0.15326765477657317,
      "num_tokens": 6337828.0,
      "step": 3060
    },
    {
      "entropy": 5.889102983474731,
      "epoch": 0.2780297532656023,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004996036669904831,
      "loss": 5.8111,
      "mean_token_accuracy": 0.14792668223381042,
      "num_tokens": 6348678.0,
      "step": 3065
    },
    {
      "entropy": 5.846058511734009,
      "epoch": 0.2784833091436865,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004996017450139261,
      "loss": 5.9027,
      "mean_token_accuracy": 0.13910972326993942,
      "num_tokens": 6359670.0,
      "step": 3070
    },
    {
      "entropy": 5.913860940933228,
      "epoch": 0.2789368650217707,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004995998183925478,
      "loss": 5.854,
      "mean_token_accuracy": 0.14639595597982408,
      "num_tokens": 6371034.0,
      "step": 3075
    },
    {
      "entropy": 5.873360013961792,
      "epoch": 0.2793904208998549,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004995978871263881,
      "loss": 5.8086,
      "mean_token_accuracy": 0.14766118600964545,
      "num_tokens": 6380672.0,
      "step": 3080
    },
    {
      "entropy": 5.863108396530151,
      "epoch": 0.27984397677793904,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004995959512154868,
      "loss": 5.9087,
      "mean_token_accuracy": 0.1358805164694786,
      "num_tokens": 6390799.0,
      "step": 3085
    },
    {
      "entropy": 5.892030858993531,
      "epoch": 0.2802975326560232,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000499594010659884,
      "loss": 5.7886,
      "mean_token_accuracy": 0.14606204628944397,
      "num_tokens": 6400965.0,
      "step": 3090
    },
    {
      "entropy": 5.858356714248657,
      "epoch": 0.2807510885341074,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004995920654596199,
      "loss": 5.8181,
      "mean_token_accuracy": 0.14298465102910995,
      "num_tokens": 6411567.0,
      "step": 3095
    },
    {
      "entropy": 5.93129563331604,
      "epoch": 0.28120464441219156,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004995901156147345,
      "loss": 5.8549,
      "mean_token_accuracy": 0.14818204939365387,
      "num_tokens": 6421193.0,
      "step": 3100
    },
    {
      "entropy": 5.919429969787598,
      "epoch": 0.2816582002902758,
      "grad_norm": 1.5,
      "learning_rate": 0.0004995881611252684,
      "loss": 5.8732,
      "mean_token_accuracy": 0.15032653510570526,
      "num_tokens": 6430281.0,
      "step": 3105
    },
    {
      "entropy": 5.843878746032715,
      "epoch": 0.28211175616835993,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004995862019912619,
      "loss": 5.8664,
      "mean_token_accuracy": 0.13923195675015448,
      "num_tokens": 6441055.0,
      "step": 3110
    },
    {
      "entropy": 5.923148584365845,
      "epoch": 0.28256531204644414,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004995842382127555,
      "loss": 5.7781,
      "mean_token_accuracy": 0.14486080035567284,
      "num_tokens": 6451082.0,
      "step": 3115
    },
    {
      "entropy": 5.824050951004028,
      "epoch": 0.2830188679245283,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004995822697897897,
      "loss": 5.8814,
      "mean_token_accuracy": 0.14269932955503464,
      "num_tokens": 6461617.0,
      "step": 3120
    },
    {
      "entropy": 5.99531660079956,
      "epoch": 0.2834724238026125,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004995802967224056,
      "loss": 5.8715,
      "mean_token_accuracy": 0.14604993909597397,
      "num_tokens": 6472780.0,
      "step": 3125
    },
    {
      "entropy": 5.821425580978394,
      "epoch": 0.28392597968069666,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004995783190106434,
      "loss": 5.7767,
      "mean_token_accuracy": 0.1456163577735424,
      "num_tokens": 6484078.0,
      "step": 3130
    },
    {
      "entropy": 5.887919139862061,
      "epoch": 0.2843795355587808,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004995763366545445,
      "loss": 5.8759,
      "mean_token_accuracy": 0.14020358249545098,
      "num_tokens": 6494366.0,
      "step": 3135
    },
    {
      "entropy": 5.793185806274414,
      "epoch": 0.284833091436865,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004995743496541497,
      "loss": 5.8095,
      "mean_token_accuracy": 0.14859849438071251,
      "num_tokens": 6505461.0,
      "step": 3140
    },
    {
      "entropy": 5.851748323440551,
      "epoch": 0.2852866473149492,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004995723580095001,
      "loss": 5.8046,
      "mean_token_accuracy": 0.14792132675647734,
      "num_tokens": 6516141.0,
      "step": 3145
    },
    {
      "entropy": 5.804424190521241,
      "epoch": 0.2857402031930334,
      "grad_norm": 1.34375,
      "learning_rate": 0.000499570361720637,
      "loss": 5.8024,
      "mean_token_accuracy": 0.1473324827849865,
      "num_tokens": 6525536.0,
      "step": 3150
    },
    {
      "entropy": 5.917052936553955,
      "epoch": 0.28619375907111755,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004995683607876015,
      "loss": 5.8618,
      "mean_token_accuracy": 0.14245882481336594,
      "num_tokens": 6535852.0,
      "step": 3155
    },
    {
      "entropy": 5.9194934368133545,
      "epoch": 0.28664731494920176,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004995663552104352,
      "loss": 5.8379,
      "mean_token_accuracy": 0.15086517333984376,
      "num_tokens": 6546353.0,
      "step": 3160
    },
    {
      "entropy": 5.818840503692627,
      "epoch": 0.2871008708272859,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004995643449891793,
      "loss": 5.7942,
      "mean_token_accuracy": 0.1443500891327858,
      "num_tokens": 6557782.0,
      "step": 3165
    },
    {
      "entropy": 5.908010578155517,
      "epoch": 0.2875544267053701,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004995623301238757,
      "loss": 5.8142,
      "mean_token_accuracy": 0.150578635931015,
      "num_tokens": 6568100.0,
      "step": 3170
    },
    {
      "entropy": 5.893179893493652,
      "epoch": 0.2880079825834543,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004995603106145657,
      "loss": 5.8712,
      "mean_token_accuracy": 0.1439751148223877,
      "num_tokens": 6578698.0,
      "step": 3175
    },
    {
      "entropy": 5.9084978103637695,
      "epoch": 0.28846153846153844,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004995582864612913,
      "loss": 5.8446,
      "mean_token_accuracy": 0.145198056101799,
      "num_tokens": 6588908.0,
      "step": 3180
    },
    {
      "entropy": 5.821733474731445,
      "epoch": 0.28891509433962265,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004995562576640943,
      "loss": 5.7096,
      "mean_token_accuracy": 0.14953436851501464,
      "num_tokens": 6599445.0,
      "step": 3185
    },
    {
      "entropy": 5.891123914718628,
      "epoch": 0.2893686502177068,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004995542242230167,
      "loss": 5.8133,
      "mean_token_accuracy": 0.14369880110025407,
      "num_tokens": 6610246.0,
      "step": 3190
    },
    {
      "entropy": 5.856010913848877,
      "epoch": 0.289822206095791,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004995521861381006,
      "loss": 5.8187,
      "mean_token_accuracy": 0.1429808184504509,
      "num_tokens": 6620501.0,
      "step": 3195
    },
    {
      "entropy": 6.004367733001709,
      "epoch": 0.29027576197387517,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000499550143409388,
      "loss": 5.8964,
      "mean_token_accuracy": 0.1352742463350296,
      "num_tokens": 6631206.0,
      "step": 3200
    },
    {
      "entropy": 5.821225309371949,
      "epoch": 0.2907293178519594,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004995480960369211,
      "loss": 5.8262,
      "mean_token_accuracy": 0.14419071823358537,
      "num_tokens": 6641600.0,
      "step": 3205
    },
    {
      "entropy": 5.829140472412109,
      "epoch": 0.29118287373004353,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004995460440207425,
      "loss": 5.8931,
      "mean_token_accuracy": 0.14225514978170395,
      "num_tokens": 6652319.0,
      "step": 3210
    },
    {
      "entropy": 5.9609822750091555,
      "epoch": 0.29163642960812775,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004995439873608945,
      "loss": 5.8314,
      "mean_token_accuracy": 0.1385738007724285,
      "num_tokens": 6663702.0,
      "step": 3215
    },
    {
      "entropy": 5.875910472869873,
      "epoch": 0.2920899854862119,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004995419260574196,
      "loss": 5.8177,
      "mean_token_accuracy": 0.1408117711544037,
      "num_tokens": 6673908.0,
      "step": 3220
    },
    {
      "entropy": 5.849833536148071,
      "epoch": 0.29254354136429606,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004995398601103603,
      "loss": 5.8474,
      "mean_token_accuracy": 0.1477855384349823,
      "num_tokens": 6683237.0,
      "step": 3225
    },
    {
      "entropy": 5.877266454696655,
      "epoch": 0.29299709724238027,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004995377895197597,
      "loss": 5.7741,
      "mean_token_accuracy": 0.1441804811358452,
      "num_tokens": 6693451.0,
      "step": 3230
    },
    {
      "entropy": 5.826403617858887,
      "epoch": 0.2934506531204644,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004995357142856602,
      "loss": 5.7418,
      "mean_token_accuracy": 0.1471038743853569,
      "num_tokens": 6703479.0,
      "step": 3235
    },
    {
      "entropy": 5.786555290222168,
      "epoch": 0.29390420899854863,
      "grad_norm": 1.3125,
      "learning_rate": 0.000499533634408105,
      "loss": 5.7384,
      "mean_token_accuracy": 0.1510653540492058,
      "num_tokens": 6715321.0,
      "step": 3240
    },
    {
      "entropy": 5.851634502410889,
      "epoch": 0.2943577648766328,
      "grad_norm": 1.546875,
      "learning_rate": 0.000499531549887137,
      "loss": 5.7926,
      "mean_token_accuracy": 0.14810622185468675,
      "num_tokens": 6725541.0,
      "step": 3245
    },
    {
      "entropy": 5.840380907058716,
      "epoch": 0.294811320754717,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004995294607227993,
      "loss": 5.8448,
      "mean_token_accuracy": 0.1491126760840416,
      "num_tokens": 6735165.0,
      "step": 3250
    },
    {
      "entropy": 5.938561058044433,
      "epoch": 0.29526487663280115,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004995273669151351,
      "loss": 5.8398,
      "mean_token_accuracy": 0.1489756792783737,
      "num_tokens": 6744245.0,
      "step": 3255
    },
    {
      "entropy": 5.934394836425781,
      "epoch": 0.29571843251088537,
      "grad_norm": 1.4375,
      "learning_rate": 0.000499525268464188,
      "loss": 5.7883,
      "mean_token_accuracy": 0.1483618885278702,
      "num_tokens": 6754894.0,
      "step": 3260
    },
    {
      "entropy": 5.832844591140747,
      "epoch": 0.2961719883889695,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004995231653700009,
      "loss": 5.8538,
      "mean_token_accuracy": 0.14105366319417953,
      "num_tokens": 6764442.0,
      "step": 3265
    },
    {
      "entropy": 5.8802886486053465,
      "epoch": 0.2966255442670537,
      "grad_norm": 1.375,
      "learning_rate": 0.0004995210576326176,
      "loss": 5.8285,
      "mean_token_accuracy": 0.14174019694328308,
      "num_tokens": 6775054.0,
      "step": 3270
    },
    {
      "entropy": 5.894062757492065,
      "epoch": 0.2970791001451379,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004995189452520816,
      "loss": 5.7856,
      "mean_token_accuracy": 0.160012948513031,
      "num_tokens": 6784825.0,
      "step": 3275
    },
    {
      "entropy": 5.887639808654785,
      "epoch": 0.29753265602322204,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004995168282284366,
      "loss": 5.8532,
      "mean_token_accuracy": 0.14631997644901276,
      "num_tokens": 6795928.0,
      "step": 3280
    },
    {
      "entropy": 5.8099130153656,
      "epoch": 0.29798621190130625,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004995147065617263,
      "loss": 5.7761,
      "mean_token_accuracy": 0.14812222272157669,
      "num_tokens": 6805456.0,
      "step": 3285
    },
    {
      "entropy": 5.922118043899536,
      "epoch": 0.2984397677793904,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004995125802519946,
      "loss": 5.8439,
      "mean_token_accuracy": 0.14502991884946823,
      "num_tokens": 6816215.0,
      "step": 3290
    },
    {
      "entropy": 5.972725343704224,
      "epoch": 0.2988933236574746,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004995104492992857,
      "loss": 5.8794,
      "mean_token_accuracy": 0.14430792406201362,
      "num_tokens": 6826545.0,
      "step": 3295
    },
    {
      "entropy": 5.7858147621154785,
      "epoch": 0.2993468795355588,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004995083137036434,
      "loss": 5.7952,
      "mean_token_accuracy": 0.1501883916556835,
      "num_tokens": 6836184.0,
      "step": 3300
    },
    {
      "entropy": 5.940231418609619,
      "epoch": 0.299800435413643,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004995061734651121,
      "loss": 5.8701,
      "mean_token_accuracy": 0.14531150236725807,
      "num_tokens": 6846646.0,
      "step": 3305
    },
    {
      "entropy": 5.887357616424561,
      "epoch": 0.30025399129172714,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004995040285837358,
      "loss": 5.7889,
      "mean_token_accuracy": 0.1509875774383545,
      "num_tokens": 6857597.0,
      "step": 3310
    },
    {
      "entropy": 5.866164779663086,
      "epoch": 0.3007075471698113,
      "grad_norm": 1.625,
      "learning_rate": 0.0004995018790595589,
      "loss": 5.8584,
      "mean_token_accuracy": 0.14564741998910904,
      "num_tokens": 6867523.0,
      "step": 3315
    },
    {
      "entropy": 5.948118543624878,
      "epoch": 0.3011611030478955,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004994997248926261,
      "loss": 5.9656,
      "mean_token_accuracy": 0.14200707226991655,
      "num_tokens": 6878190.0,
      "step": 3320
    },
    {
      "entropy": 5.900257682800293,
      "epoch": 0.30161465892597966,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004994975660829817,
      "loss": 5.76,
      "mean_token_accuracy": 0.15460998713970184,
      "num_tokens": 6888483.0,
      "step": 3325
    },
    {
      "entropy": 5.831659650802612,
      "epoch": 0.3020682148040639,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004994954026306705,
      "loss": 5.7383,
      "mean_token_accuracy": 0.14502749666571618,
      "num_tokens": 6898337.0,
      "step": 3330
    },
    {
      "entropy": 5.909942626953125,
      "epoch": 0.30252177068214803,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004994932345357371,
      "loss": 5.8354,
      "mean_token_accuracy": 0.14530461728572847,
      "num_tokens": 6909131.0,
      "step": 3335
    },
    {
      "entropy": 5.851958847045898,
      "epoch": 0.30297532656023224,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004994910617982264,
      "loss": 5.7986,
      "mean_token_accuracy": 0.14323278665542602,
      "num_tokens": 6919790.0,
      "step": 3340
    },
    {
      "entropy": 5.855467844009399,
      "epoch": 0.3034288824383164,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004994888844181833,
      "loss": 5.8352,
      "mean_token_accuracy": 0.1461854785680771,
      "num_tokens": 6930072.0,
      "step": 3345
    },
    {
      "entropy": 5.8171202659606935,
      "epoch": 0.3038824383164006,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004994867023956529,
      "loss": 5.708,
      "mean_token_accuracy": 0.1426757961511612,
      "num_tokens": 6940646.0,
      "step": 3350
    },
    {
      "entropy": 5.799750900268554,
      "epoch": 0.30433599419448476,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004994845157306803,
      "loss": 5.7194,
      "mean_token_accuracy": 0.1513740189373493,
      "num_tokens": 6951645.0,
      "step": 3355
    },
    {
      "entropy": 5.950794792175293,
      "epoch": 0.3047895500725689,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004994823244233107,
      "loss": 5.8849,
      "mean_token_accuracy": 0.14364167153835297,
      "num_tokens": 6963020.0,
      "step": 3360
    },
    {
      "entropy": 5.939484977722168,
      "epoch": 0.3052431059506531,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004994801284735895,
      "loss": 5.8,
      "mean_token_accuracy": 0.15134814828634263,
      "num_tokens": 6972895.0,
      "step": 3365
    },
    {
      "entropy": 5.890461444854736,
      "epoch": 0.3056966618287373,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004994779278815619,
      "loss": 5.9493,
      "mean_token_accuracy": 0.13638330921530722,
      "num_tokens": 6985363.0,
      "step": 3370
    },
    {
      "entropy": 5.943120670318604,
      "epoch": 0.3061502177068215,
      "grad_norm": 2.125,
      "learning_rate": 0.0004994757226472736,
      "loss": 5.8885,
      "mean_token_accuracy": 0.14796184450387956,
      "num_tokens": 6997040.0,
      "step": 3375
    },
    {
      "entropy": 5.883255338668823,
      "epoch": 0.30660377358490565,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004994735127707702,
      "loss": 5.8132,
      "mean_token_accuracy": 0.14786105006933212,
      "num_tokens": 7007591.0,
      "step": 3380
    },
    {
      "entropy": 5.877407550811768,
      "epoch": 0.30705732946298986,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004994712982520972,
      "loss": 5.8239,
      "mean_token_accuracy": 0.1532495453953743,
      "num_tokens": 7017858.0,
      "step": 3385
    },
    {
      "entropy": 5.839944124221802,
      "epoch": 0.307510885341074,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004994690790913008,
      "loss": 5.782,
      "mean_token_accuracy": 0.145537555962801,
      "num_tokens": 7027365.0,
      "step": 3390
    },
    {
      "entropy": 5.779080820083618,
      "epoch": 0.3079644412191582,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004994668552884265,
      "loss": 5.7916,
      "mean_token_accuracy": 0.14543195366859435,
      "num_tokens": 7038728.0,
      "step": 3395
    },
    {
      "entropy": 5.920030498504639,
      "epoch": 0.3084179970972424,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004994646268435204,
      "loss": 5.7284,
      "mean_token_accuracy": 0.15286441892385483,
      "num_tokens": 7048736.0,
      "step": 3400
    },
    {
      "entropy": 5.903434181213379,
      "epoch": 0.30887155297532654,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004994623937566286,
      "loss": 5.8435,
      "mean_token_accuracy": 0.14856164008378983,
      "num_tokens": 7059205.0,
      "step": 3405
    },
    {
      "entropy": 5.834399271011352,
      "epoch": 0.30932510885341075,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004994601560277974,
      "loss": 5.8519,
      "mean_token_accuracy": 0.13926964849233628,
      "num_tokens": 7069513.0,
      "step": 3410
    },
    {
      "entropy": 5.890263462066651,
      "epoch": 0.3097786647314949,
      "grad_norm": 1.453125,
      "learning_rate": 0.000499457913657073,
      "loss": 5.7705,
      "mean_token_accuracy": 0.147540445625782,
      "num_tokens": 7079940.0,
      "step": 3415
    },
    {
      "entropy": 5.894211339950561,
      "epoch": 0.3102322206095791,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004994556666445016,
      "loss": 5.7803,
      "mean_token_accuracy": 0.1465594932436943,
      "num_tokens": 7090792.0,
      "step": 3420
    },
    {
      "entropy": 5.789063596725464,
      "epoch": 0.31068577648766327,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004994534149901299,
      "loss": 5.7383,
      "mean_token_accuracy": 0.14637766480445863,
      "num_tokens": 7101964.0,
      "step": 3425
    },
    {
      "entropy": 5.8579018115997314,
      "epoch": 0.3111393323657475,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004994511586940044,
      "loss": 5.7893,
      "mean_token_accuracy": 0.14257095232605935,
      "num_tokens": 7113332.0,
      "step": 3430
    },
    {
      "entropy": 5.97416934967041,
      "epoch": 0.31159288824383163,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004994488977561717,
      "loss": 5.9521,
      "mean_token_accuracy": 0.14183745980262757,
      "num_tokens": 7123481.0,
      "step": 3435
    },
    {
      "entropy": 5.814324283599854,
      "epoch": 0.31204644412191584,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004994466321766786,
      "loss": 5.7119,
      "mean_token_accuracy": 0.15085338801145554,
      "num_tokens": 7133694.0,
      "step": 3440
    },
    {
      "entropy": 5.849011659622192,
      "epoch": 0.3125,
      "grad_norm": 1.6171875,
      "learning_rate": 0.000499444361955572,
      "loss": 5.8351,
      "mean_token_accuracy": 0.15081941783428193,
      "num_tokens": 7144491.0,
      "step": 3445
    },
    {
      "entropy": 6.005057287216187,
      "epoch": 0.31295355587808416,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004994420870928988,
      "loss": 5.9086,
      "mean_token_accuracy": 0.14320788010954857,
      "num_tokens": 7154716.0,
      "step": 3450
    },
    {
      "entropy": 5.896556758880616,
      "epoch": 0.31340711175616837,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004994398075887059,
      "loss": 5.9023,
      "mean_token_accuracy": 0.14134786948561667,
      "num_tokens": 7164258.0,
      "step": 3455
    },
    {
      "entropy": 5.83188419342041,
      "epoch": 0.3138606676342525,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004994375234430407,
      "loss": 5.7675,
      "mean_token_accuracy": 0.1461493268609047,
      "num_tokens": 7175098.0,
      "step": 3460
    },
    {
      "entropy": 5.892137575149536,
      "epoch": 0.31431422351233673,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004994352346559504,
      "loss": 5.8343,
      "mean_token_accuracy": 0.14542945474386215,
      "num_tokens": 7186480.0,
      "step": 3465
    },
    {
      "entropy": 5.799042797088623,
      "epoch": 0.3147677793904209,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004994329412274821,
      "loss": 5.8072,
      "mean_token_accuracy": 0.14921786934137343,
      "num_tokens": 7197056.0,
      "step": 3470
    },
    {
      "entropy": 5.921955919265747,
      "epoch": 0.3152213352685051,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004994306431576835,
      "loss": 5.908,
      "mean_token_accuracy": 0.14707638397812844,
      "num_tokens": 7206040.0,
      "step": 3475
    },
    {
      "entropy": 5.987424421310425,
      "epoch": 0.31567489114658925,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000499428340446602,
      "loss": 5.773,
      "mean_token_accuracy": 0.15154775828123093,
      "num_tokens": 7216344.0,
      "step": 3480
    },
    {
      "entropy": 5.893925428390503,
      "epoch": 0.31612844702467346,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004994260330942851,
      "loss": 5.8535,
      "mean_token_accuracy": 0.14545960426330568,
      "num_tokens": 7226729.0,
      "step": 3485
    },
    {
      "entropy": 5.822306489944458,
      "epoch": 0.3165820029027576,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004994237211007808,
      "loss": 5.8019,
      "mean_token_accuracy": 0.14516061842441558,
      "num_tokens": 7237616.0,
      "step": 3490
    },
    {
      "entropy": 5.896799755096436,
      "epoch": 0.3170355587808418,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004994214044661367,
      "loss": 5.7364,
      "mean_token_accuracy": 0.14803769513964654,
      "num_tokens": 7247857.0,
      "step": 3495
    },
    {
      "entropy": 5.767730283737182,
      "epoch": 0.317489114658926,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004994190831904007,
      "loss": 5.711,
      "mean_token_accuracy": 0.15267411172389983,
      "num_tokens": 7257858.0,
      "step": 3500
    },
    {
      "entropy": 5.71488733291626,
      "epoch": 0.31794267053701014,
      "grad_norm": 1.6171875,
      "learning_rate": 0.000499416757273621,
      "loss": 5.7186,
      "mean_token_accuracy": 0.1502937451004982,
      "num_tokens": 7268440.0,
      "step": 3505
    },
    {
      "entropy": 5.9244773387908936,
      "epoch": 0.31839622641509435,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004994144267158454,
      "loss": 5.7995,
      "mean_token_accuracy": 0.15224562138319014,
      "num_tokens": 7278560.0,
      "step": 3510
    },
    {
      "entropy": 5.970515632629395,
      "epoch": 0.3188497822931785,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004994120915171224,
      "loss": 5.7927,
      "mean_token_accuracy": 0.14984412491321564,
      "num_tokens": 7288422.0,
      "step": 3515
    },
    {
      "entropy": 5.850483465194702,
      "epoch": 0.3193033381712627,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004994097516775002,
      "loss": 5.8555,
      "mean_token_accuracy": 0.14620980769395828,
      "num_tokens": 7298613.0,
      "step": 3520
    },
    {
      "entropy": 5.9037727355957035,
      "epoch": 0.3197568940493469,
      "grad_norm": 1.625,
      "learning_rate": 0.0004994074071970271,
      "loss": 5.8943,
      "mean_token_accuracy": 0.1411413997411728,
      "num_tokens": 7309509.0,
      "step": 3525
    },
    {
      "entropy": 5.948377180099487,
      "epoch": 0.3202104499274311,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004994050580757515,
      "loss": 5.8095,
      "mean_token_accuracy": 0.14360484182834626,
      "num_tokens": 7319015.0,
      "step": 3530
    },
    {
      "entropy": 5.825267028808594,
      "epoch": 0.32066400580551524,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004994027043137223,
      "loss": 5.7781,
      "mean_token_accuracy": 0.15057334899902344,
      "num_tokens": 7329026.0,
      "step": 3535
    },
    {
      "entropy": 5.940208768844604,
      "epoch": 0.3211175616835994,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004994003459109879,
      "loss": 5.8815,
      "mean_token_accuracy": 0.14752597510814666,
      "num_tokens": 7339638.0,
      "step": 3540
    },
    {
      "entropy": 5.925622272491455,
      "epoch": 0.3215711175616836,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004993979828675972,
      "loss": 5.7797,
      "mean_token_accuracy": 0.1494549334049225,
      "num_tokens": 7349979.0,
      "step": 3545
    },
    {
      "entropy": 5.7519083499908445,
      "epoch": 0.32202467343976776,
      "grad_norm": 2.53125,
      "learning_rate": 0.000499395615183599,
      "loss": 5.7062,
      "mean_token_accuracy": 0.15184318274259567,
      "num_tokens": 7359398.0,
      "step": 3550
    },
    {
      "entropy": 5.7866541862487795,
      "epoch": 0.32247822931785197,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004993932428590422,
      "loss": 5.7652,
      "mean_token_accuracy": 0.1473848804831505,
      "num_tokens": 7370418.0,
      "step": 3555
    },
    {
      "entropy": 5.946097373962402,
      "epoch": 0.3229317851959361,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004993908658939762,
      "loss": 5.8028,
      "mean_token_accuracy": 0.1365408256649971,
      "num_tokens": 7380461.0,
      "step": 3560
    },
    {
      "entropy": 5.865476608276367,
      "epoch": 0.32338534107402034,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004993884842884496,
      "loss": 5.7909,
      "mean_token_accuracy": 0.14746063798666,
      "num_tokens": 7390646.0,
      "step": 3565
    },
    {
      "entropy": 5.833335304260254,
      "epoch": 0.3238388969521045,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004993860980425121,
      "loss": 5.9112,
      "mean_token_accuracy": 0.14362555146217346,
      "num_tokens": 7402106.0,
      "step": 3570
    },
    {
      "entropy": 5.881076526641846,
      "epoch": 0.3242924528301887,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004993837071562131,
      "loss": 5.8834,
      "mean_token_accuracy": 0.14547116905450821,
      "num_tokens": 7412936.0,
      "step": 3575
    },
    {
      "entropy": 5.909448480606079,
      "epoch": 0.32474600870827286,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004993813116296017,
      "loss": 5.7451,
      "mean_token_accuracy": 0.15334491580724716,
      "num_tokens": 7422303.0,
      "step": 3580
    },
    {
      "entropy": 5.858001613616944,
      "epoch": 0.325199564586357,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004993789114627275,
      "loss": 5.8084,
      "mean_token_accuracy": 0.1455884724855423,
      "num_tokens": 7432829.0,
      "step": 3585
    },
    {
      "entropy": 5.897142839431763,
      "epoch": 0.3256531204644412,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004993765066556403,
      "loss": 5.8427,
      "mean_token_accuracy": 0.14214815869927405,
      "num_tokens": 7443218.0,
      "step": 3590
    },
    {
      "entropy": 5.87728099822998,
      "epoch": 0.3261066763425254,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004993740972083899,
      "loss": 5.8019,
      "mean_token_accuracy": 0.13930851370096206,
      "num_tokens": 7454531.0,
      "step": 3595
    },
    {
      "entropy": 5.862864971160889,
      "epoch": 0.3265602322206096,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004993716831210259,
      "loss": 5.8417,
      "mean_token_accuracy": 0.1489178165793419,
      "num_tokens": 7464993.0,
      "step": 3600
    },
    {
      "entropy": 5.848693084716797,
      "epoch": 0.32701378809869375,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004993692643935983,
      "loss": 5.8819,
      "mean_token_accuracy": 0.14340077340602875,
      "num_tokens": 7475755.0,
      "step": 3605
    },
    {
      "entropy": 5.898052215576172,
      "epoch": 0.32746734397677796,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004993668410261571,
      "loss": 5.8648,
      "mean_token_accuracy": 0.14696988239884376,
      "num_tokens": 7486778.0,
      "step": 3610
    },
    {
      "entropy": 5.989887142181397,
      "epoch": 0.3279208998548621,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004993644130187525,
      "loss": 5.826,
      "mean_token_accuracy": 0.14816126823425294,
      "num_tokens": 7497439.0,
      "step": 3615
    },
    {
      "entropy": 5.916240262985229,
      "epoch": 0.3283744557329463,
      "grad_norm": 1.875,
      "learning_rate": 0.0004993619803714347,
      "loss": 5.8852,
      "mean_token_accuracy": 0.14901129752397538,
      "num_tokens": 7508274.0,
      "step": 3620
    },
    {
      "entropy": 5.900164651870727,
      "epoch": 0.3288280116110305,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004993595430842539,
      "loss": 5.8396,
      "mean_token_accuracy": 0.14458931982517242,
      "num_tokens": 7519315.0,
      "step": 3625
    },
    {
      "entropy": 5.966023588180542,
      "epoch": 0.32928156748911463,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004993571011572606,
      "loss": 5.8417,
      "mean_token_accuracy": 0.14624714255332946,
      "num_tokens": 7529910.0,
      "step": 3630
    },
    {
      "entropy": 5.87901463508606,
      "epoch": 0.32973512336719885,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004993546545905053,
      "loss": 5.7665,
      "mean_token_accuracy": 0.15389569848775864,
      "num_tokens": 7538966.0,
      "step": 3635
    },
    {
      "entropy": 5.901384925842285,
      "epoch": 0.330188679245283,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004993522033840386,
      "loss": 5.8063,
      "mean_token_accuracy": 0.1490241676568985,
      "num_tokens": 7549683.0,
      "step": 3640
    },
    {
      "entropy": 5.766738080978394,
      "epoch": 0.3306422351233672,
      "grad_norm": 1.40625,
      "learning_rate": 0.000499349747537911,
      "loss": 5.788,
      "mean_token_accuracy": 0.14788973480463027,
      "num_tokens": 7560629.0,
      "step": 3645
    },
    {
      "entropy": 5.873026323318482,
      "epoch": 0.33109579100145137,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004993472870521736,
      "loss": 5.7384,
      "mean_token_accuracy": 0.15600567162036896,
      "num_tokens": 7570613.0,
      "step": 3650
    },
    {
      "entropy": 5.851700639724731,
      "epoch": 0.3315493468795356,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004993448219268771,
      "loss": 5.7104,
      "mean_token_accuracy": 0.1479036659002304,
      "num_tokens": 7580640.0,
      "step": 3655
    },
    {
      "entropy": 5.855727005004883,
      "epoch": 0.33200290275761973,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004993423521620726,
      "loss": 5.7926,
      "mean_token_accuracy": 0.14974100142717361,
      "num_tokens": 7590164.0,
      "step": 3660
    },
    {
      "entropy": 5.854600191116333,
      "epoch": 0.33245645863570394,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004993398777578109,
      "loss": 5.8563,
      "mean_token_accuracy": 0.1418274387717247,
      "num_tokens": 7600173.0,
      "step": 3665
    },
    {
      "entropy": 5.840301847457885,
      "epoch": 0.3329100145137881,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004993373987141435,
      "loss": 5.6667,
      "mean_token_accuracy": 0.15604306310415267,
      "num_tokens": 7609958.0,
      "step": 3670
    },
    {
      "entropy": 5.945655298233032,
      "epoch": 0.33336357039187225,
      "grad_norm": 1.875,
      "learning_rate": 0.0004993349150311214,
      "loss": 5.7995,
      "mean_token_accuracy": 0.14565619081258774,
      "num_tokens": 7620237.0,
      "step": 3675
    },
    {
      "entropy": 5.800672483444214,
      "epoch": 0.33381712626995647,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004993324267087962,
      "loss": 5.6914,
      "mean_token_accuracy": 0.15652529150247574,
      "num_tokens": 7629886.0,
      "step": 3680
    },
    {
      "entropy": 5.822660684585571,
      "epoch": 0.3342706821480406,
      "grad_norm": 1.875,
      "learning_rate": 0.0004993299337472192,
      "loss": 5.77,
      "mean_token_accuracy": 0.15085675567388535,
      "num_tokens": 7640057.0,
      "step": 3685
    },
    {
      "entropy": 5.817735576629639,
      "epoch": 0.33472423802612483,
      "grad_norm": 1.6015625,
      "learning_rate": 0.000499327436146442,
      "loss": 5.7257,
      "mean_token_accuracy": 0.1475510910153389,
      "num_tokens": 7649766.0,
      "step": 3690
    },
    {
      "entropy": 5.8362211227417,
      "epoch": 0.335177793904209,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004993249339065162,
      "loss": 5.7994,
      "mean_token_accuracy": 0.14809372201561927,
      "num_tokens": 7660890.0,
      "step": 3695
    },
    {
      "entropy": 5.8948447704315186,
      "epoch": 0.3356313497822932,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004993224270274937,
      "loss": 5.8063,
      "mean_token_accuracy": 0.14590197429060936,
      "num_tokens": 7670737.0,
      "step": 3700
    },
    {
      "entropy": 5.797480678558349,
      "epoch": 0.33608490566037735,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004993199155094263,
      "loss": 5.7728,
      "mean_token_accuracy": 0.14692779183387755,
      "num_tokens": 7681407.0,
      "step": 3705
    },
    {
      "entropy": 5.8480466365814205,
      "epoch": 0.33653846153846156,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004993173993523658,
      "loss": 5.7183,
      "mean_token_accuracy": 0.16032444536685944,
      "num_tokens": 7691758.0,
      "step": 3710
    },
    {
      "entropy": 5.812997913360595,
      "epoch": 0.3369920174165457,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004993148785563643,
      "loss": 5.7426,
      "mean_token_accuracy": 0.14660506919026375,
      "num_tokens": 7702934.0,
      "step": 3715
    },
    {
      "entropy": 5.723813915252686,
      "epoch": 0.3374455732946299,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000499312353121474,
      "loss": 5.7393,
      "mean_token_accuracy": 0.15231507793068885,
      "num_tokens": 7713001.0,
      "step": 3720
    },
    {
      "entropy": 5.890734529495239,
      "epoch": 0.3378991291727141,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004993098230477471,
      "loss": 5.794,
      "mean_token_accuracy": 0.14045751169323922,
      "num_tokens": 7724005.0,
      "step": 3725
    },
    {
      "entropy": 5.900547075271606,
      "epoch": 0.33835268505079824,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004993072883352358,
      "loss": 5.7467,
      "mean_token_accuracy": 0.1473555825650692,
      "num_tokens": 7733743.0,
      "step": 3730
    },
    {
      "entropy": 5.855947017669678,
      "epoch": 0.33880624092888245,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004993047489839927,
      "loss": 5.9383,
      "mean_token_accuracy": 0.13958059400320053,
      "num_tokens": 7743762.0,
      "step": 3735
    },
    {
      "entropy": 5.930490493774414,
      "epoch": 0.3392597968069666,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004993022049940702,
      "loss": 5.817,
      "mean_token_accuracy": 0.14824567139148712,
      "num_tokens": 7753758.0,
      "step": 3740
    },
    {
      "entropy": 5.871481132507324,
      "epoch": 0.3397133526850508,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004992996563655209,
      "loss": 5.7719,
      "mean_token_accuracy": 0.1495535232126713,
      "num_tokens": 7764692.0,
      "step": 3745
    },
    {
      "entropy": 5.848981046676636,
      "epoch": 0.340166908563135,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004992971030983975,
      "loss": 5.8128,
      "mean_token_accuracy": 0.14353037178516387,
      "num_tokens": 7776892.0,
      "step": 3750
    },
    {
      "entropy": 5.902813720703125,
      "epoch": 0.3406204644412192,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004992945451927531,
      "loss": 5.8642,
      "mean_token_accuracy": 0.1436462000012398,
      "num_tokens": 7787143.0,
      "step": 3755
    },
    {
      "entropy": 5.806349992752075,
      "epoch": 0.34107402031930334,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004992919826486402,
      "loss": 5.7675,
      "mean_token_accuracy": 0.1482495039701462,
      "num_tokens": 7798537.0,
      "step": 3760
    },
    {
      "entropy": 5.8226241111755375,
      "epoch": 0.3415275761973875,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004992894154661118,
      "loss": 5.7344,
      "mean_token_accuracy": 0.15073239281773568,
      "num_tokens": 7808434.0,
      "step": 3765
    },
    {
      "entropy": 5.856382083892822,
      "epoch": 0.3419811320754717,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004992868436452213,
      "loss": 5.851,
      "mean_token_accuracy": 0.1481686621904373,
      "num_tokens": 7819592.0,
      "step": 3770
    },
    {
      "entropy": 5.890800809860229,
      "epoch": 0.34243468795355586,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004992842671860217,
      "loss": 5.7417,
      "mean_token_accuracy": 0.15032289028167725,
      "num_tokens": 7830602.0,
      "step": 3775
    },
    {
      "entropy": 5.879576730728149,
      "epoch": 0.34288824383164007,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004992816860885662,
      "loss": 5.8953,
      "mean_token_accuracy": 0.1439703606069088,
      "num_tokens": 7840850.0,
      "step": 3780
    },
    {
      "entropy": 6.028309392929077,
      "epoch": 0.3433417997097242,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004992791003529084,
      "loss": 5.8544,
      "mean_token_accuracy": 0.14057560488581658,
      "num_tokens": 7852371.0,
      "step": 3785
    },
    {
      "entropy": 5.9221659183502195,
      "epoch": 0.34379535558780844,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004992765099791017,
      "loss": 5.8467,
      "mean_token_accuracy": 0.14126079380512238,
      "num_tokens": 7863290.0,
      "step": 3790
    },
    {
      "entropy": 5.783722972869873,
      "epoch": 0.3442489114658926,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004992739149671995,
      "loss": 5.687,
      "mean_token_accuracy": 0.1527349516749382,
      "num_tokens": 7873927.0,
      "step": 3795
    },
    {
      "entropy": 5.791727685928345,
      "epoch": 0.3447024673439768,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004992713153172556,
      "loss": 5.7173,
      "mean_token_accuracy": 0.14965997114777566,
      "num_tokens": 7884977.0,
      "step": 3800
    },
    {
      "entropy": 5.827057123184204,
      "epoch": 0.34515602322206096,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004992687110293238,
      "loss": 5.7985,
      "mean_token_accuracy": 0.15286190062761307,
      "num_tokens": 7894620.0,
      "step": 3805
    },
    {
      "entropy": 5.84814682006836,
      "epoch": 0.3456095791001451,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004992661021034579,
      "loss": 5.7404,
      "mean_token_accuracy": 0.15445290505886078,
      "num_tokens": 7904541.0,
      "step": 3810
    },
    {
      "entropy": 5.818321228027344,
      "epoch": 0.3460631349782293,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004992634885397118,
      "loss": 5.7626,
      "mean_token_accuracy": 0.14903516918420792,
      "num_tokens": 7915435.0,
      "step": 3815
    },
    {
      "entropy": 5.83319125175476,
      "epoch": 0.3465166908563135,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004992608703381396,
      "loss": 5.742,
      "mean_token_accuracy": 0.14871969372034072,
      "num_tokens": 7925870.0,
      "step": 3820
    },
    {
      "entropy": 5.8578869819641115,
      "epoch": 0.3469702467343977,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004992582474987955,
      "loss": 5.7545,
      "mean_token_accuracy": 0.14518966972827912,
      "num_tokens": 7937046.0,
      "step": 3825
    },
    {
      "entropy": 5.77307505607605,
      "epoch": 0.34742380261248185,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004992556200217337,
      "loss": 5.7064,
      "mean_token_accuracy": 0.14739801585674286,
      "num_tokens": 7947150.0,
      "step": 3830
    },
    {
      "entropy": 5.862250852584839,
      "epoch": 0.34787735849056606,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004992529879070085,
      "loss": 5.8049,
      "mean_token_accuracy": 0.1511588916182518,
      "num_tokens": 7956815.0,
      "step": 3835
    },
    {
      "entropy": 5.875207948684692,
      "epoch": 0.3483309143686502,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004992503511546745,
      "loss": 5.8322,
      "mean_token_accuracy": 0.14654328525066376,
      "num_tokens": 7968114.0,
      "step": 3840
    },
    {
      "entropy": 5.881295871734619,
      "epoch": 0.3487844702467344,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004992477097647859,
      "loss": 5.7647,
      "mean_token_accuracy": 0.14300089776515962,
      "num_tokens": 7977840.0,
      "step": 3845
    },
    {
      "entropy": 5.8811851978302006,
      "epoch": 0.3492380261248186,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004992450637373976,
      "loss": 5.8822,
      "mean_token_accuracy": 0.14847241342067719,
      "num_tokens": 7988663.0,
      "step": 3850
    },
    {
      "entropy": 5.808912992477417,
      "epoch": 0.34969158200290273,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004992424130725642,
      "loss": 5.6955,
      "mean_token_accuracy": 0.14893131777644159,
      "num_tokens": 7999693.0,
      "step": 3855
    },
    {
      "entropy": 5.872442388534546,
      "epoch": 0.35014513788098695,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004992397577703406,
      "loss": 5.8621,
      "mean_token_accuracy": 0.14381872415542601,
      "num_tokens": 8010154.0,
      "step": 3860
    },
    {
      "entropy": 5.946900177001953,
      "epoch": 0.3505986937590711,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004992370978307816,
      "loss": 5.8415,
      "mean_token_accuracy": 0.14593325406312943,
      "num_tokens": 8020921.0,
      "step": 3865
    },
    {
      "entropy": 5.8881330490112305,
      "epoch": 0.3510522496371553,
      "grad_norm": 2.5,
      "learning_rate": 0.0004992344332539424,
      "loss": 5.7708,
      "mean_token_accuracy": 0.15435807704925536,
      "num_tokens": 8031404.0,
      "step": 3870
    },
    {
      "entropy": 5.730764579772949,
      "epoch": 0.35150580551523947,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004992317640398779,
      "loss": 5.7562,
      "mean_token_accuracy": 0.1508772239089012,
      "num_tokens": 8041310.0,
      "step": 3875
    },
    {
      "entropy": 5.828841018676758,
      "epoch": 0.3519593613933237,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004992290901886434,
      "loss": 5.7351,
      "mean_token_accuracy": 0.14970532357692717,
      "num_tokens": 8051054.0,
      "step": 3880
    },
    {
      "entropy": 5.846589374542236,
      "epoch": 0.35241291727140783,
      "grad_norm": 1.859375,
      "learning_rate": 0.000499226411700294,
      "loss": 5.7467,
      "mean_token_accuracy": 0.14929102063179017,
      "num_tokens": 8061063.0,
      "step": 3885
    },
    {
      "entropy": 5.819320011138916,
      "epoch": 0.35286647314949204,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004992237285748855,
      "loss": 5.749,
      "mean_token_accuracy": 0.15423073768615722,
      "num_tokens": 8071715.0,
      "step": 3890
    },
    {
      "entropy": 5.911148452758789,
      "epoch": 0.3533200290275762,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004992210408124731,
      "loss": 5.8243,
      "mean_token_accuracy": 0.14293495416641236,
      "num_tokens": 8082679.0,
      "step": 3895
    },
    {
      "entropy": 5.898702573776245,
      "epoch": 0.35377358490566035,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004992183484131123,
      "loss": 5.8528,
      "mean_token_accuracy": 0.14598763659596442,
      "num_tokens": 8093785.0,
      "step": 3900
    },
    {
      "entropy": 5.834564685821533,
      "epoch": 0.35422714078374457,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004992156513768591,
      "loss": 5.7622,
      "mean_token_accuracy": 0.1508698433637619,
      "num_tokens": 8105822.0,
      "step": 3905
    },
    {
      "entropy": 5.7624317646026615,
      "epoch": 0.3546806966618287,
      "grad_norm": 1.7109375,
      "learning_rate": 0.000499212949703769,
      "loss": 5.682,
      "mean_token_accuracy": 0.15023178458213807,
      "num_tokens": 8115521.0,
      "step": 3910
    },
    {
      "entropy": 5.811681795120239,
      "epoch": 0.35513425253991293,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004992102433938981,
      "loss": 5.7786,
      "mean_token_accuracy": 0.14909463599324227,
      "num_tokens": 8125524.0,
      "step": 3915
    },
    {
      "entropy": 5.856146621704101,
      "epoch": 0.3555878084179971,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004992075324473021,
      "loss": 5.6655,
      "mean_token_accuracy": 0.16061010211706161,
      "num_tokens": 8135552.0,
      "step": 3920
    },
    {
      "entropy": 5.878396081924438,
      "epoch": 0.3560413642960813,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004992048168640371,
      "loss": 5.8021,
      "mean_token_accuracy": 0.14835941046476364,
      "num_tokens": 8145617.0,
      "step": 3925
    },
    {
      "entropy": 5.8598052024841305,
      "epoch": 0.35649492017416545,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004992020966441595,
      "loss": 5.8254,
      "mean_token_accuracy": 0.14229780659079552,
      "num_tokens": 8156729.0,
      "step": 3930
    },
    {
      "entropy": 5.834389066696167,
      "epoch": 0.35694847605224966,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004991993717877254,
      "loss": 5.8103,
      "mean_token_accuracy": 0.1433609440922737,
      "num_tokens": 8167637.0,
      "step": 3935
    },
    {
      "entropy": 5.883994483947754,
      "epoch": 0.3574020319303338,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004991966422947911,
      "loss": 5.7989,
      "mean_token_accuracy": 0.14962732493877412,
      "num_tokens": 8177354.0,
      "step": 3940
    },
    {
      "entropy": 5.8065185546875,
      "epoch": 0.357855587808418,
      "grad_norm": 2.0,
      "learning_rate": 0.0004991939081654132,
      "loss": 5.7199,
      "mean_token_accuracy": 0.16128290593624114,
      "num_tokens": 8187580.0,
      "step": 3945
    },
    {
      "entropy": 5.752160549163818,
      "epoch": 0.3583091436865022,
      "grad_norm": 1.796875,
      "learning_rate": 0.000499191169399648,
      "loss": 5.8216,
      "mean_token_accuracy": 0.14404607713222503,
      "num_tokens": 8199382.0,
      "step": 3950
    },
    {
      "entropy": 5.846578645706177,
      "epoch": 0.35876269956458634,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004991884259975525,
      "loss": 5.7256,
      "mean_token_accuracy": 0.14961068257689475,
      "num_tokens": 8208484.0,
      "step": 3955
    },
    {
      "entropy": 5.8810042381286625,
      "epoch": 0.35921625544267055,
      "grad_norm": 1.7421875,
      "learning_rate": 0.000499185677959183,
      "loss": 5.7582,
      "mean_token_accuracy": 0.14644067436456681,
      "num_tokens": 8219056.0,
      "step": 3960
    },
    {
      "entropy": 5.849036693572998,
      "epoch": 0.3596698113207547,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004991829252845966,
      "loss": 5.7966,
      "mean_token_accuracy": 0.14414289593696594,
      "num_tokens": 8230149.0,
      "step": 3965
    },
    {
      "entropy": 5.865396165847779,
      "epoch": 0.3601233671988389,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004991801679738502,
      "loss": 5.8497,
      "mean_token_accuracy": 0.1402447283267975,
      "num_tokens": 8240803.0,
      "step": 3970
    },
    {
      "entropy": 5.896198892593384,
      "epoch": 0.3605769230769231,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004991774060270008,
      "loss": 5.8232,
      "mean_token_accuracy": 0.14320749416947365,
      "num_tokens": 8251563.0,
      "step": 3975
    },
    {
      "entropy": 5.7957603454589846,
      "epoch": 0.3610304789550073,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004991746394441055,
      "loss": 5.7534,
      "mean_token_accuracy": 0.14512395709753037,
      "num_tokens": 8261342.0,
      "step": 3980
    },
    {
      "entropy": 5.8211053848266605,
      "epoch": 0.36148403483309144,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004991718682252216,
      "loss": 5.8592,
      "mean_token_accuracy": 0.14282869845628737,
      "num_tokens": 8271540.0,
      "step": 3985
    },
    {
      "entropy": 5.969235181808472,
      "epoch": 0.3619375907111756,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004991690923704063,
      "loss": 5.774,
      "mean_token_accuracy": 0.1459478497505188,
      "num_tokens": 8280593.0,
      "step": 3990
    },
    {
      "entropy": 5.8458741188049315,
      "epoch": 0.3623911465892598,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004991663118797171,
      "loss": 5.7937,
      "mean_token_accuracy": 0.1484941601753235,
      "num_tokens": 8290126.0,
      "step": 3995
    },
    {
      "entropy": 5.810283374786377,
      "epoch": 0.36284470246734396,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004991635267532113,
      "loss": 5.7276,
      "mean_token_accuracy": 0.1489349991083145,
      "num_tokens": 8301416.0,
      "step": 4000
    },
    {
      "entropy": 5.92182149887085,
      "epoch": 0.36329825834542817,
      "grad_norm": 1.75,
      "learning_rate": 0.0004991607369909468,
      "loss": 5.8051,
      "mean_token_accuracy": 0.14902321994304657,
      "num_tokens": 8311034.0,
      "step": 4005
    },
    {
      "entropy": 5.9140537738800045,
      "epoch": 0.3637518142235123,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004991579425929811,
      "loss": 5.8382,
      "mean_token_accuracy": 0.14275244921445845,
      "num_tokens": 8322203.0,
      "step": 4010
    },
    {
      "entropy": 5.910188245773315,
      "epoch": 0.36420537010159654,
      "grad_norm": 1.5546875,
      "learning_rate": 0.000499155143559372,
      "loss": 5.703,
      "mean_token_accuracy": 0.15037306100130082,
      "num_tokens": 8330981.0,
      "step": 4015
    },
    {
      "entropy": 5.727109909057617,
      "epoch": 0.3646589259796807,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004991523398901773,
      "loss": 5.7422,
      "mean_token_accuracy": 0.14837262779474258,
      "num_tokens": 8341294.0,
      "step": 4020
    },
    {
      "entropy": 5.937906169891358,
      "epoch": 0.3651124818577649,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004991495315854553,
      "loss": 5.8772,
      "mean_token_accuracy": 0.14473702684044837,
      "num_tokens": 8351824.0,
      "step": 4025
    },
    {
      "entropy": 5.887541055679321,
      "epoch": 0.36556603773584906,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004991467186452637,
      "loss": 5.7527,
      "mean_token_accuracy": 0.14745676666498184,
      "num_tokens": 8361975.0,
      "step": 4030
    },
    {
      "entropy": 5.788140058517456,
      "epoch": 0.3660195936139332,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004991439010696609,
      "loss": 5.7179,
      "mean_token_accuracy": 0.15196914672851564,
      "num_tokens": 8372145.0,
      "step": 4035
    },
    {
      "entropy": 5.883623647689819,
      "epoch": 0.3664731494920174,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004991410788587051,
      "loss": 5.7945,
      "mean_token_accuracy": 0.1431012563407421,
      "num_tokens": 8382542.0,
      "step": 4040
    },
    {
      "entropy": 5.880132675170898,
      "epoch": 0.3669267053701016,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004991382520124547,
      "loss": 5.8053,
      "mean_token_accuracy": 0.14454633221030236,
      "num_tokens": 8393450.0,
      "step": 4045
    },
    {
      "entropy": 5.843823099136353,
      "epoch": 0.3673802612481858,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004991354205309681,
      "loss": 5.7325,
      "mean_token_accuracy": 0.1445598617196083,
      "num_tokens": 8404929.0,
      "step": 4050
    },
    {
      "entropy": 5.8328451156616214,
      "epoch": 0.36783381712626995,
      "grad_norm": 2.015625,
      "learning_rate": 0.000499132584414304,
      "loss": 5.763,
      "mean_token_accuracy": 0.14964553862810134,
      "num_tokens": 8415703.0,
      "step": 4055
    },
    {
      "entropy": 5.886098432540893,
      "epoch": 0.36828737300435416,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004991297436625207,
      "loss": 5.8256,
      "mean_token_accuracy": 0.14880608171224594,
      "num_tokens": 8426220.0,
      "step": 4060
    },
    {
      "entropy": 5.749038791656494,
      "epoch": 0.3687409288824383,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004991268982756774,
      "loss": 5.7145,
      "mean_token_accuracy": 0.15660099536180497,
      "num_tokens": 8436220.0,
      "step": 4065
    },
    {
      "entropy": 5.905976867675781,
      "epoch": 0.3691944847605225,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004991240482538327,
      "loss": 5.8618,
      "mean_token_accuracy": 0.140651698410511,
      "num_tokens": 8446724.0,
      "step": 4070
    },
    {
      "entropy": 5.880495882034301,
      "epoch": 0.3696480406386067,
      "grad_norm": 1.625,
      "learning_rate": 0.0004991211935970455,
      "loss": 5.7217,
      "mean_token_accuracy": 0.1466690830886364,
      "num_tokens": 8457270.0,
      "step": 4075
    },
    {
      "entropy": 5.808180665969848,
      "epoch": 0.37010159651669083,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004991183343053748,
      "loss": 5.7933,
      "mean_token_accuracy": 0.1437254838645458,
      "num_tokens": 8468137.0,
      "step": 4080
    },
    {
      "entropy": 5.832044219970703,
      "epoch": 0.37055515239477504,
      "grad_norm": 1.59375,
      "learning_rate": 0.00049911547037888,
      "loss": 5.7281,
      "mean_token_accuracy": 0.1513017401099205,
      "num_tokens": 8477574.0,
      "step": 4085
    },
    {
      "entropy": 5.84261417388916,
      "epoch": 0.3710087082728592,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004991126018176202,
      "loss": 5.8255,
      "mean_token_accuracy": 0.14621062725782394,
      "num_tokens": 8487268.0,
      "step": 4090
    },
    {
      "entropy": 5.94973406791687,
      "epoch": 0.3714622641509434,
      "grad_norm": 1.75,
      "learning_rate": 0.0004991097286216545,
      "loss": 5.8993,
      "mean_token_accuracy": 0.14729729145765305,
      "num_tokens": 8497785.0,
      "step": 4095
    },
    {
      "entropy": 5.92133355140686,
      "epoch": 0.37191582002902757,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004991068507910427,
      "loss": 5.7538,
      "mean_token_accuracy": 0.14549212902784348,
      "num_tokens": 8508169.0,
      "step": 4100
    },
    {
      "entropy": 5.796505546569824,
      "epoch": 0.3723693759071118,
      "grad_norm": 1.515625,
      "learning_rate": 0.000499103968325844,
      "loss": 5.7271,
      "mean_token_accuracy": 0.14605697244405746,
      "num_tokens": 8519201.0,
      "step": 4105
    },
    {
      "entropy": 5.881065845489502,
      "epoch": 0.37282293178519593,
      "grad_norm": 1.5,
      "learning_rate": 0.0004991010812261181,
      "loss": 5.8454,
      "mean_token_accuracy": 0.13941508531570435,
      "num_tokens": 8531093.0,
      "step": 4110
    },
    {
      "entropy": 5.850164127349854,
      "epoch": 0.37327648766328014,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004990981894919248,
      "loss": 5.6946,
      "mean_token_accuracy": 0.14870947301387788,
      "num_tokens": 8541730.0,
      "step": 4115
    },
    {
      "entropy": 5.792994403839112,
      "epoch": 0.3737300435413643,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004990952931233238,
      "loss": 5.7656,
      "mean_token_accuracy": 0.150176040828228,
      "num_tokens": 8552570.0,
      "step": 4120
    },
    {
      "entropy": 5.824308729171753,
      "epoch": 0.37418359941944845,
      "grad_norm": 2.015625,
      "learning_rate": 0.000499092392120375,
      "loss": 5.6597,
      "mean_token_accuracy": 0.1537941262125969,
      "num_tokens": 8562340.0,
      "step": 4125
    },
    {
      "entropy": 5.876683855056763,
      "epoch": 0.37463715529753266,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004990894864831385,
      "loss": 5.8779,
      "mean_token_accuracy": 0.14065658450126647,
      "num_tokens": 8573240.0,
      "step": 4130
    },
    {
      "entropy": 5.889408159255981,
      "epoch": 0.3750907111756168,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004990865762116743,
      "loss": 5.7275,
      "mean_token_accuracy": 0.15267123952507972,
      "num_tokens": 8583194.0,
      "step": 4135
    },
    {
      "entropy": 5.83022403717041,
      "epoch": 0.37554426705370103,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004990836613060426,
      "loss": 5.7907,
      "mean_token_accuracy": 0.1415669023990631,
      "num_tokens": 8593603.0,
      "step": 4140
    },
    {
      "entropy": 5.852129411697388,
      "epoch": 0.3759978229317852,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004990807417663036,
      "loss": 5.8044,
      "mean_token_accuracy": 0.14670791774988173,
      "num_tokens": 8603176.0,
      "step": 4145
    },
    {
      "entropy": 5.894939708709717,
      "epoch": 0.3764513788098694,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004990778175925179,
      "loss": 5.8056,
      "mean_token_accuracy": 0.15206588059663773,
      "num_tokens": 8614801.0,
      "step": 4150
    },
    {
      "entropy": 5.8759352684021,
      "epoch": 0.37690493468795355,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004990748887847457,
      "loss": 5.8379,
      "mean_token_accuracy": 0.1481950357556343,
      "num_tokens": 8624999.0,
      "step": 4155
    },
    {
      "entropy": 5.830811405181885,
      "epoch": 0.37735849056603776,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004990719553430477,
      "loss": 5.7012,
      "mean_token_accuracy": 0.1591061070561409,
      "num_tokens": 8635358.0,
      "step": 4160
    },
    {
      "entropy": 5.838520669937134,
      "epoch": 0.3778120464441219,
      "grad_norm": 1.625,
      "learning_rate": 0.0004990690172674847,
      "loss": 5.7631,
      "mean_token_accuracy": 0.1457097850739956,
      "num_tokens": 8645430.0,
      "step": 4165
    },
    {
      "entropy": 5.834043788909912,
      "epoch": 0.3782656023222061,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004990660745581172,
      "loss": 5.7158,
      "mean_token_accuracy": 0.15678425431251525,
      "num_tokens": 8655115.0,
      "step": 4170
    },
    {
      "entropy": 5.725192070007324,
      "epoch": 0.3787191582002903,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004990631272150062,
      "loss": 5.7122,
      "mean_token_accuracy": 0.14877525568008423,
      "num_tokens": 8665391.0,
      "step": 4175
    },
    {
      "entropy": 5.905309581756592,
      "epoch": 0.37917271407837444,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004990601752382127,
      "loss": 5.8272,
      "mean_token_accuracy": 0.14464897736907006,
      "num_tokens": 8676819.0,
      "step": 4180
    },
    {
      "entropy": 5.917379760742188,
      "epoch": 0.37962626995645865,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004990572186277975,
      "loss": 5.7598,
      "mean_token_accuracy": 0.15172497928142548,
      "num_tokens": 8687052.0,
      "step": 4185
    },
    {
      "entropy": 5.85373740196228,
      "epoch": 0.3800798258345428,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004990542573838221,
      "loss": 5.8318,
      "mean_token_accuracy": 0.14298003166913986,
      "num_tokens": 8697756.0,
      "step": 4190
    },
    {
      "entropy": 5.784401750564575,
      "epoch": 0.380533381712627,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004990512915063475,
      "loss": 5.7493,
      "mean_token_accuracy": 0.15153614282608033,
      "num_tokens": 8708388.0,
      "step": 4195
    },
    {
      "entropy": 5.872270154953003,
      "epoch": 0.38098693759071117,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004990483209954352,
      "loss": 5.8258,
      "mean_token_accuracy": 0.1415306583046913,
      "num_tokens": 8719475.0,
      "step": 4200
    },
    {
      "entropy": 5.8587422370910645,
      "epoch": 0.3814404934687954,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004990453458511464,
      "loss": 5.6861,
      "mean_token_accuracy": 0.15934394747018815,
      "num_tokens": 8729395.0,
      "step": 4205
    },
    {
      "entropy": 5.873230600357056,
      "epoch": 0.38189404934687954,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004990423660735428,
      "loss": 5.7549,
      "mean_token_accuracy": 0.1521786004304886,
      "num_tokens": 8740351.0,
      "step": 4210
    },
    {
      "entropy": 5.826301527023316,
      "epoch": 0.3823476052249637,
      "grad_norm": 1.78125,
      "learning_rate": 0.000499039381662686,
      "loss": 5.7769,
      "mean_token_accuracy": 0.14992594420909883,
      "num_tokens": 8749260.0,
      "step": 4215
    },
    {
      "entropy": 5.838196468353272,
      "epoch": 0.3828011611030479,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004990363926186377,
      "loss": 5.8601,
      "mean_token_accuracy": 0.13967606276273728,
      "num_tokens": 8759604.0,
      "step": 4220
    },
    {
      "entropy": 5.865482902526855,
      "epoch": 0.38325471698113206,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004990333989414597,
      "loss": 5.8563,
      "mean_token_accuracy": 0.14229664355516433,
      "num_tokens": 8769815.0,
      "step": 4225
    },
    {
      "entropy": 5.842839145660401,
      "epoch": 0.38370827285921627,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004990304006312138,
      "loss": 5.8042,
      "mean_token_accuracy": 0.144217911362648,
      "num_tokens": 8780557.0,
      "step": 4230
    },
    {
      "entropy": 5.893247365951538,
      "epoch": 0.3841618287373004,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004990273976879622,
      "loss": 5.746,
      "mean_token_accuracy": 0.15741802453994752,
      "num_tokens": 8790993.0,
      "step": 4235
    },
    {
      "entropy": 5.908751630783081,
      "epoch": 0.38461538461538464,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004990243901117669,
      "loss": 5.7993,
      "mean_token_accuracy": 0.14164488315582274,
      "num_tokens": 8800657.0,
      "step": 4240
    },
    {
      "entropy": 5.749859476089478,
      "epoch": 0.3850689404934688,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004990213779026903,
      "loss": 5.8134,
      "mean_token_accuracy": 0.14273017644882202,
      "num_tokens": 8812983.0,
      "step": 4245
    },
    {
      "entropy": 5.878562164306641,
      "epoch": 0.385522496371553,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004990183610607944,
      "loss": 5.7723,
      "mean_token_accuracy": 0.14839280098676683,
      "num_tokens": 8822980.0,
      "step": 4250
    },
    {
      "entropy": 5.868063688278198,
      "epoch": 0.38597605224963716,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004990153395861415,
      "loss": 5.7693,
      "mean_token_accuracy": 0.15079350471496583,
      "num_tokens": 8833821.0,
      "step": 4255
    },
    {
      "entropy": 5.78454327583313,
      "epoch": 0.3864296081277213,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004990123134787946,
      "loss": 5.7049,
      "mean_token_accuracy": 0.15205924212932587,
      "num_tokens": 8843929.0,
      "step": 4260
    },
    {
      "entropy": 5.8183831691741945,
      "epoch": 0.3868831640058055,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004990092827388159,
      "loss": 5.6926,
      "mean_token_accuracy": 0.14838198870420455,
      "num_tokens": 8854180.0,
      "step": 4265
    },
    {
      "entropy": 5.767665433883667,
      "epoch": 0.3873367198838897,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004990062473662681,
      "loss": 5.708,
      "mean_token_accuracy": 0.15309692025184632,
      "num_tokens": 8865018.0,
      "step": 4270
    },
    {
      "entropy": 5.879072809219361,
      "epoch": 0.3877902757619739,
      "grad_norm": 1.765625,
      "learning_rate": 0.000499003207361214,
      "loss": 5.8918,
      "mean_token_accuracy": 0.14059318900108336,
      "num_tokens": 8875423.0,
      "step": 4275
    },
    {
      "entropy": 5.988429594039917,
      "epoch": 0.38824383164005805,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004990001627237165,
      "loss": 5.8433,
      "mean_token_accuracy": 0.14742626398801803,
      "num_tokens": 8887850.0,
      "step": 4280
    },
    {
      "entropy": 5.77903642654419,
      "epoch": 0.38869738751814226,
      "grad_norm": 2.125,
      "learning_rate": 0.0004989971134538386,
      "loss": 5.7332,
      "mean_token_accuracy": 0.152449893951416,
      "num_tokens": 8897429.0,
      "step": 4285
    },
    {
      "entropy": 5.795289611816406,
      "epoch": 0.3891509433962264,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004989940595516432,
      "loss": 5.7322,
      "mean_token_accuracy": 0.15252401381731034,
      "num_tokens": 8907048.0,
      "step": 4290
    },
    {
      "entropy": 5.867383098602295,
      "epoch": 0.3896044992743106,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004989910010171938,
      "loss": 5.7955,
      "mean_token_accuracy": 0.14241870269179344,
      "num_tokens": 8917987.0,
      "step": 4295
    },
    {
      "entropy": 5.902405786514282,
      "epoch": 0.3900580551523948,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004989879378505532,
      "loss": 5.7042,
      "mean_token_accuracy": 0.145450921356678,
      "num_tokens": 8929339.0,
      "step": 4300
    },
    {
      "entropy": 5.787789440155029,
      "epoch": 0.39051161103047893,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004989848700517849,
      "loss": 5.7059,
      "mean_token_accuracy": 0.15002443864941598,
      "num_tokens": 8939473.0,
      "step": 4305
    },
    {
      "entropy": 5.778521919250489,
      "epoch": 0.39096516690856314,
      "grad_norm": 3.09375,
      "learning_rate": 0.0004989817976209526,
      "loss": 5.7477,
      "mean_token_accuracy": 0.14690720587968825,
      "num_tokens": 8949173.0,
      "step": 4310
    },
    {
      "entropy": 5.890463495254517,
      "epoch": 0.3914187227866473,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004989787205581196,
      "loss": 5.7321,
      "mean_token_accuracy": 0.15126073211431504,
      "num_tokens": 8959879.0,
      "step": 4315
    },
    {
      "entropy": 5.9013659954071045,
      "epoch": 0.3918722786647315,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004989756388633496,
      "loss": 5.7962,
      "mean_token_accuracy": 0.1466445192694664,
      "num_tokens": 8970643.0,
      "step": 4320
    },
    {
      "entropy": 5.776039171218872,
      "epoch": 0.39232583454281567,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004989725525367063,
      "loss": 5.662,
      "mean_token_accuracy": 0.15230050086975097,
      "num_tokens": 8979882.0,
      "step": 4325
    },
    {
      "entropy": 5.822905206680298,
      "epoch": 0.3927793904208999,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004989694615782534,
      "loss": 5.8141,
      "mean_token_accuracy": 0.14447067975997924,
      "num_tokens": 8990925.0,
      "step": 4330
    },
    {
      "entropy": 5.899914932250977,
      "epoch": 0.39323294629898403,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004989663659880552,
      "loss": 5.84,
      "mean_token_accuracy": 0.1423853024840355,
      "num_tokens": 9001365.0,
      "step": 4335
    },
    {
      "entropy": 5.923264074325561,
      "epoch": 0.3936865021770682,
      "grad_norm": 3.0,
      "learning_rate": 0.0004989632657661754,
      "loss": 5.8286,
      "mean_token_accuracy": 0.14973220527172088,
      "num_tokens": 9011942.0,
      "step": 4340
    },
    {
      "entropy": 5.935524415969849,
      "epoch": 0.3941400580551524,
      "grad_norm": 1.8828125,
      "learning_rate": 0.000498960160912678,
      "loss": 5.8534,
      "mean_token_accuracy": 0.14469540640711784,
      "num_tokens": 9022610.0,
      "step": 4345
    },
    {
      "entropy": 5.852841472625732,
      "epoch": 0.39459361393323655,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004989570514276276,
      "loss": 5.7577,
      "mean_token_accuracy": 0.14448110163211822,
      "num_tokens": 9033034.0,
      "step": 4350
    },
    {
      "entropy": 5.914149904251099,
      "epoch": 0.39504716981132076,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004989539373110883,
      "loss": 5.7943,
      "mean_token_accuracy": 0.1416556268930435,
      "num_tokens": 9043418.0,
      "step": 4355
    },
    {
      "entropy": 5.945955562591553,
      "epoch": 0.3955007256894049,
      "grad_norm": 1.75,
      "learning_rate": 0.0004989508185631245,
      "loss": 5.8783,
      "mean_token_accuracy": 0.14210682958364487,
      "num_tokens": 9052708.0,
      "step": 4360
    },
    {
      "entropy": 5.764258909225464,
      "epoch": 0.39595428156748913,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004989476951838006,
      "loss": 5.741,
      "mean_token_accuracy": 0.14578800797462463,
      "num_tokens": 9062883.0,
      "step": 4365
    },
    {
      "entropy": 5.867785787582397,
      "epoch": 0.3964078374455733,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004989445671731814,
      "loss": 5.7054,
      "mean_token_accuracy": 0.15434126257896424,
      "num_tokens": 9072830.0,
      "step": 4370
    },
    {
      "entropy": 5.7734722137451175,
      "epoch": 0.3968613933236575,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004989414345313315,
      "loss": 5.6661,
      "mean_token_accuracy": 0.1422044441103935,
      "num_tokens": 9082749.0,
      "step": 4375
    },
    {
      "entropy": 5.874911880493164,
      "epoch": 0.39731494920174165,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004989382972583156,
      "loss": 5.7632,
      "mean_token_accuracy": 0.14635880067944526,
      "num_tokens": 9094188.0,
      "step": 4380
    },
    {
      "entropy": 5.875842332839966,
      "epoch": 0.3977685050798258,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004989351553541986,
      "loss": 5.7743,
      "mean_token_accuracy": 0.1443245954811573,
      "num_tokens": 9105023.0,
      "step": 4385
    },
    {
      "entropy": 5.7874414920806885,
      "epoch": 0.39822206095791,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004989320088190456,
      "loss": 5.725,
      "mean_token_accuracy": 0.1519322693347931,
      "num_tokens": 9115563.0,
      "step": 4390
    },
    {
      "entropy": 5.8170891284942625,
      "epoch": 0.3986756168359942,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004989288576529216,
      "loss": 5.7447,
      "mean_token_accuracy": 0.14550501853227615,
      "num_tokens": 9125394.0,
      "step": 4395
    },
    {
      "entropy": 5.876050138473511,
      "epoch": 0.3991291727140784,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004989257018558917,
      "loss": 5.7457,
      "mean_token_accuracy": 0.14662075340747832,
      "num_tokens": 9137061.0,
      "step": 4400
    },
    {
      "entropy": 5.858101940155029,
      "epoch": 0.39958272859216254,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004989225414280213,
      "loss": 5.7085,
      "mean_token_accuracy": 0.14827828407287597,
      "num_tokens": 9148196.0,
      "step": 4405
    },
    {
      "entropy": 5.830370330810547,
      "epoch": 0.40003628447024675,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004989193763693756,
      "loss": 5.7425,
      "mean_token_accuracy": 0.15035545378923415,
      "num_tokens": 9158013.0,
      "step": 4410
    },
    {
      "entropy": 5.81728343963623,
      "epoch": 0.4004898403483309,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004989162066800202,
      "loss": 5.6504,
      "mean_token_accuracy": 0.1513180986046791,
      "num_tokens": 9167813.0,
      "step": 4415
    },
    {
      "entropy": 5.791998386383057,
      "epoch": 0.4009433962264151,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004989130323600205,
      "loss": 5.7369,
      "mean_token_accuracy": 0.15377046391367913,
      "num_tokens": 9177783.0,
      "step": 4420
    },
    {
      "entropy": 5.829497003555298,
      "epoch": 0.40139695210449927,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004989098534094424,
      "loss": 5.7658,
      "mean_token_accuracy": 0.15063516944646835,
      "num_tokens": 9187556.0,
      "step": 4425
    },
    {
      "entropy": 5.810343599319458,
      "epoch": 0.4018505079825834,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004989066698283512,
      "loss": 5.8178,
      "mean_token_accuracy": 0.14447112008929253,
      "num_tokens": 9198659.0,
      "step": 4430
    },
    {
      "entropy": 5.862035846710205,
      "epoch": 0.40230406386066764,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004989034816168132,
      "loss": 5.7796,
      "mean_token_accuracy": 0.14605519771575928,
      "num_tokens": 9209546.0,
      "step": 4435
    },
    {
      "entropy": 5.83269739151001,
      "epoch": 0.4027576197387518,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004989002887748941,
      "loss": 5.7002,
      "mean_token_accuracy": 0.1521790012717247,
      "num_tokens": 9219339.0,
      "step": 4440
    },
    {
      "entropy": 5.877965784072876,
      "epoch": 0.403211175616836,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004988970913026601,
      "loss": 5.8796,
      "mean_token_accuracy": 0.14370514899492265,
      "num_tokens": 9230364.0,
      "step": 4445
    },
    {
      "entropy": 5.9046632766723635,
      "epoch": 0.40366473149492016,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004988938892001771,
      "loss": 5.7459,
      "mean_token_accuracy": 0.14931874126195907,
      "num_tokens": 9240732.0,
      "step": 4450
    },
    {
      "entropy": 5.796473550796509,
      "epoch": 0.40411828737300437,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004988906824675113,
      "loss": 5.6842,
      "mean_token_accuracy": 0.15042189061641692,
      "num_tokens": 9251049.0,
      "step": 4455
    },
    {
      "entropy": 5.852061557769775,
      "epoch": 0.4045718432510885,
      "grad_norm": 2.125,
      "learning_rate": 0.0004988874711047293,
      "loss": 5.727,
      "mean_token_accuracy": 0.15044807642698288,
      "num_tokens": 9261660.0,
      "step": 4460
    },
    {
      "entropy": 5.830964851379394,
      "epoch": 0.40502539912917274,
      "grad_norm": 1.625,
      "learning_rate": 0.0004988842551118972,
      "loss": 5.798,
      "mean_token_accuracy": 0.14962558969855308,
      "num_tokens": 9272114.0,
      "step": 4465
    },
    {
      "entropy": 5.867624425888062,
      "epoch": 0.4054789550072569,
      "grad_norm": 2.125,
      "learning_rate": 0.0004988810344890818,
      "loss": 5.767,
      "mean_token_accuracy": 0.15117275416851045,
      "num_tokens": 9282125.0,
      "step": 4470
    },
    {
      "entropy": 5.913155126571655,
      "epoch": 0.40593251088534105,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004988778092363494,
      "loss": 5.8271,
      "mean_token_accuracy": 0.14333474934101104,
      "num_tokens": 9294078.0,
      "step": 4475
    },
    {
      "entropy": 5.834015846252441,
      "epoch": 0.40638606676342526,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004988745793537671,
      "loss": 5.7602,
      "mean_token_accuracy": 0.14512751549482344,
      "num_tokens": 9304026.0,
      "step": 4480
    },
    {
      "entropy": 5.8868653774261475,
      "epoch": 0.4068396226415094,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004988713448414011,
      "loss": 5.7957,
      "mean_token_accuracy": 0.15431096106767656,
      "num_tokens": 9313933.0,
      "step": 4485
    },
    {
      "entropy": 5.919203186035157,
      "epoch": 0.4072931785195936,
      "grad_norm": 2.390625,
      "learning_rate": 0.000498868105699319,
      "loss": 5.8216,
      "mean_token_accuracy": 0.14666167348623277,
      "num_tokens": 9323932.0,
      "step": 4490
    },
    {
      "entropy": 5.755291604995728,
      "epoch": 0.4077467343976778,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004988648619275872,
      "loss": 5.7265,
      "mean_token_accuracy": 0.15795115977525712,
      "num_tokens": 9333467.0,
      "step": 4495
    },
    {
      "entropy": 5.849260568618774,
      "epoch": 0.408200290275762,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004988616135262729,
      "loss": 5.7941,
      "mean_token_accuracy": 0.1402094468474388,
      "num_tokens": 9344909.0,
      "step": 4500
    },
    {
      "entropy": 5.984080076217651,
      "epoch": 0.40865384615384615,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004988583604954436,
      "loss": 5.8291,
      "mean_token_accuracy": 0.1422007627785206,
      "num_tokens": 9355630.0,
      "step": 4505
    },
    {
      "entropy": 5.847146606445312,
      "epoch": 0.40910740203193036,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004988551028351663,
      "loss": 5.7674,
      "mean_token_accuracy": 0.14505108445882797,
      "num_tokens": 9365700.0,
      "step": 4510
    },
    {
      "entropy": 5.8058990955352785,
      "epoch": 0.4095609579100145,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004988518405455084,
      "loss": 5.8489,
      "mean_token_accuracy": 0.14073796719312667,
      "num_tokens": 9377088.0,
      "step": 4515
    },
    {
      "entropy": 5.843289089202881,
      "epoch": 0.41001451378809867,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004988485736265374,
      "loss": 5.6492,
      "mean_token_accuracy": 0.1532072402536869,
      "num_tokens": 9386855.0,
      "step": 4520
    },
    {
      "entropy": 5.841462135314941,
      "epoch": 0.4104680696661829,
      "grad_norm": 1.6875,
      "learning_rate": 0.000498845302078321,
      "loss": 5.772,
      "mean_token_accuracy": 0.14480848461389542,
      "num_tokens": 9398661.0,
      "step": 4525
    },
    {
      "entropy": 5.877380418777466,
      "epoch": 0.41092162554426703,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004988420259009266,
      "loss": 5.7597,
      "mean_token_accuracy": 0.15487963408231736,
      "num_tokens": 9408670.0,
      "step": 4530
    },
    {
      "entropy": 5.917161273956299,
      "epoch": 0.41137518142235124,
      "grad_norm": 3.078125,
      "learning_rate": 0.000498838745094422,
      "loss": 5.7748,
      "mean_token_accuracy": 0.14555237144231797,
      "num_tokens": 9418740.0,
      "step": 4535
    },
    {
      "entropy": 5.8036597728729244,
      "epoch": 0.4118287373004354,
      "grad_norm": 3.25,
      "learning_rate": 0.0004988354596588751,
      "loss": 5.7659,
      "mean_token_accuracy": 0.15228321254253388,
      "num_tokens": 9429339.0,
      "step": 4540
    },
    {
      "entropy": 5.7854917526245115,
      "epoch": 0.4122822931785196,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004988321695943539,
      "loss": 5.7236,
      "mean_token_accuracy": 0.15588060170412063,
      "num_tokens": 9438996.0,
      "step": 4545
    },
    {
      "entropy": 5.850178670883179,
      "epoch": 0.41273584905660377,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004988288749009264,
      "loss": 5.7185,
      "mean_token_accuracy": 0.15071613490581512,
      "num_tokens": 9448669.0,
      "step": 4550
    },
    {
      "entropy": 5.810451078414917,
      "epoch": 0.413189404934688,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004988255755786608,
      "loss": 5.7114,
      "mean_token_accuracy": 0.14924689680337905,
      "num_tokens": 9458857.0,
      "step": 4555
    },
    {
      "entropy": 5.801498889923096,
      "epoch": 0.41364296081277213,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004988222716276251,
      "loss": 5.7071,
      "mean_token_accuracy": 0.15637744963169098,
      "num_tokens": 9470241.0,
      "step": 4560
    },
    {
      "entropy": 5.892170333862305,
      "epoch": 0.4140965166908563,
      "grad_norm": 2.5625,
      "learning_rate": 0.000498818963047888,
      "loss": 5.8641,
      "mean_token_accuracy": 0.14676743596792222,
      "num_tokens": 9480528.0,
      "step": 4565
    },
    {
      "entropy": 5.931382036209106,
      "epoch": 0.4145500725689405,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004988156498395176,
      "loss": 5.786,
      "mean_token_accuracy": 0.14832354038953782,
      "num_tokens": 9491000.0,
      "step": 4570
    },
    {
      "entropy": 5.862825870513916,
      "epoch": 0.41500362844702465,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004988123320025825,
      "loss": 5.7668,
      "mean_token_accuracy": 0.1513970524072647,
      "num_tokens": 9501612.0,
      "step": 4575
    },
    {
      "entropy": 5.8863836288452145,
      "epoch": 0.41545718432510886,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004988090095371514,
      "loss": 5.8554,
      "mean_token_accuracy": 0.14679653570055962,
      "num_tokens": 9513999.0,
      "step": 4580
    },
    {
      "entropy": 5.8441126346588135,
      "epoch": 0.415910740203193,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004988056824432928,
      "loss": 5.7543,
      "mean_token_accuracy": 0.14250365793704986,
      "num_tokens": 9524392.0,
      "step": 4585
    },
    {
      "entropy": 5.861783695220947,
      "epoch": 0.41636429608127723,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004988023507210758,
      "loss": 5.7453,
      "mean_token_accuracy": 0.14709829092025756,
      "num_tokens": 9534227.0,
      "step": 4590
    },
    {
      "entropy": 5.811742496490479,
      "epoch": 0.4168178519593614,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004987990143705692,
      "loss": 5.7413,
      "mean_token_accuracy": 0.15399961918592453,
      "num_tokens": 9544242.0,
      "step": 4595
    },
    {
      "entropy": 5.838329458236695,
      "epoch": 0.4172714078374456,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004987956733918418,
      "loss": 5.6717,
      "mean_token_accuracy": 0.1506224110722542,
      "num_tokens": 9553921.0,
      "step": 4600
    },
    {
      "entropy": 5.757907295227051,
      "epoch": 0.41772496371552975,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004987923277849629,
      "loss": 5.7169,
      "mean_token_accuracy": 0.14515915364027024,
      "num_tokens": 9564768.0,
      "step": 4605
    },
    {
      "entropy": 5.806964540481568,
      "epoch": 0.4181785195936139,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004987889775500017,
      "loss": 5.7797,
      "mean_token_accuracy": 0.14279748350381852,
      "num_tokens": 9576360.0,
      "step": 4610
    },
    {
      "entropy": 5.898333644866943,
      "epoch": 0.4186320754716981,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004987856226870272,
      "loss": 5.7643,
      "mean_token_accuracy": 0.14810260534286498,
      "num_tokens": 9587623.0,
      "step": 4615
    },
    {
      "entropy": 5.835229158401489,
      "epoch": 0.4190856313497823,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004987822631961092,
      "loss": 5.7959,
      "mean_token_accuracy": 0.14757825285196305,
      "num_tokens": 9597837.0,
      "step": 4620
    },
    {
      "entropy": 5.865486764907837,
      "epoch": 0.4195391872278665,
      "grad_norm": 5.03125,
      "learning_rate": 0.000498778899077317,
      "loss": 5.7521,
      "mean_token_accuracy": 0.1463379107415676,
      "num_tokens": 9609509.0,
      "step": 4625
    },
    {
      "entropy": 5.824074077606201,
      "epoch": 0.41999274310595064,
      "grad_norm": 1.9765625,
      "learning_rate": 0.00049877553033072,
      "loss": 5.7348,
      "mean_token_accuracy": 0.15134868174791336,
      "num_tokens": 9619954.0,
      "step": 4630
    },
    {
      "entropy": 5.8004528999328615,
      "epoch": 0.42044629898403485,
      "grad_norm": 1.875,
      "learning_rate": 0.0004987721569563882,
      "loss": 5.7403,
      "mean_token_accuracy": 0.14655848294496537,
      "num_tokens": 9629742.0,
      "step": 4635
    },
    {
      "entropy": 5.837172031402588,
      "epoch": 0.420899854862119,
      "grad_norm": 1.6484375,
      "learning_rate": 0.000498768778954391,
      "loss": 5.7608,
      "mean_token_accuracy": 0.14900261014699936,
      "num_tokens": 9641378.0,
      "step": 4640
    },
    {
      "entropy": 5.948794746398926,
      "epoch": 0.4213534107402032,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004987653963247985,
      "loss": 5.8219,
      "mean_token_accuracy": 0.14978400021791458,
      "num_tokens": 9652279.0,
      "step": 4645
    },
    {
      "entropy": 5.867143392562866,
      "epoch": 0.42180696661828737,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004987620090676805,
      "loss": 5.838,
      "mean_token_accuracy": 0.14259155243635177,
      "num_tokens": 9663416.0,
      "step": 4650
    },
    {
      "entropy": 5.858006381988526,
      "epoch": 0.4222605224963715,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004987586171831072,
      "loss": 5.7857,
      "mean_token_accuracy": 0.1458583153784275,
      "num_tokens": 9674647.0,
      "step": 4655
    },
    {
      "entropy": 5.827457094192505,
      "epoch": 0.42271407837445574,
      "grad_norm": 1.5,
      "learning_rate": 0.0004987552206711487,
      "loss": 5.625,
      "mean_token_accuracy": 0.1527285732328892,
      "num_tokens": 9685084.0,
      "step": 4660
    },
    {
      "entropy": 5.700690031051636,
      "epoch": 0.4231676342525399,
      "grad_norm": 1.75,
      "learning_rate": 0.0004987518195318752,
      "loss": 5.6882,
      "mean_token_accuracy": 0.1495150238275528,
      "num_tokens": 9695661.0,
      "step": 4665
    },
    {
      "entropy": 5.858991527557373,
      "epoch": 0.4236211901306241,
      "grad_norm": 1.671875,
      "learning_rate": 0.000498748413765357,
      "loss": 5.7407,
      "mean_token_accuracy": 0.14949771612882615,
      "num_tokens": 9705773.0,
      "step": 4670
    },
    {
      "entropy": 5.8706464767456055,
      "epoch": 0.42407474600870826,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004987450033716646,
      "loss": 5.8052,
      "mean_token_accuracy": 0.14631288051605223,
      "num_tokens": 9715799.0,
      "step": 4675
    },
    {
      "entropy": 5.8481348037719725,
      "epoch": 0.42452830188679247,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004987415883508686,
      "loss": 5.7683,
      "mean_token_accuracy": 0.15039038732647897,
      "num_tokens": 9727012.0,
      "step": 4680
    },
    {
      "entropy": 5.842692613601685,
      "epoch": 0.4249818577648766,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004987381687030394,
      "loss": 5.7363,
      "mean_token_accuracy": 0.148531074821949,
      "num_tokens": 9737044.0,
      "step": 4685
    },
    {
      "entropy": 5.8995438575744625,
      "epoch": 0.42543541364296084,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004987347444282479,
      "loss": 5.8423,
      "mean_token_accuracy": 0.1406104564666748,
      "num_tokens": 9747648.0,
      "step": 4690
    },
    {
      "entropy": 5.810405111312866,
      "epoch": 0.425888969521045,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004987313155265649,
      "loss": 5.7585,
      "mean_token_accuracy": 0.14476184248924256,
      "num_tokens": 9757812.0,
      "step": 4695
    },
    {
      "entropy": 5.901001262664795,
      "epoch": 0.42634252539912915,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004987278819980613,
      "loss": 5.7523,
      "mean_token_accuracy": 0.15064995884895324,
      "num_tokens": 9768337.0,
      "step": 4700
    },
    {
      "entropy": 5.868827390670776,
      "epoch": 0.42679608127721336,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004987244438428081,
      "loss": 5.7938,
      "mean_token_accuracy": 0.14378955513238906,
      "num_tokens": 9779433.0,
      "step": 4705
    },
    {
      "entropy": 5.774157524108887,
      "epoch": 0.4272496371552975,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004987210010608764,
      "loss": 5.6983,
      "mean_token_accuracy": 0.14819056391716004,
      "num_tokens": 9790069.0,
      "step": 4710
    },
    {
      "entropy": 5.828324174880981,
      "epoch": 0.4277031930333817,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004987175536523373,
      "loss": 5.6902,
      "mean_token_accuracy": 0.15661128610372543,
      "num_tokens": 9799669.0,
      "step": 4715
    },
    {
      "entropy": 5.810670280456543,
      "epoch": 0.4281567489114659,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004987141016172622,
      "loss": 5.7101,
      "mean_token_accuracy": 0.15046556890010834,
      "num_tokens": 9809615.0,
      "step": 4720
    },
    {
      "entropy": 5.774604225158692,
      "epoch": 0.4286103047895501,
      "grad_norm": 3.5625,
      "learning_rate": 0.0004987106449557225,
      "loss": 5.6934,
      "mean_token_accuracy": 0.14824203699827193,
      "num_tokens": 9819666.0,
      "step": 4725
    },
    {
      "entropy": 5.898928689956665,
      "epoch": 0.42906386066763424,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004987071836677896,
      "loss": 5.8448,
      "mean_token_accuracy": 0.14774569869041443,
      "num_tokens": 9831015.0,
      "step": 4730
    },
    {
      "entropy": 5.842265319824219,
      "epoch": 0.42951741654571846,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004987037177535353,
      "loss": 5.7116,
      "mean_token_accuracy": 0.1472062274813652,
      "num_tokens": 9841354.0,
      "step": 4735
    },
    {
      "entropy": 5.716348314285279,
      "epoch": 0.4299709724238026,
      "grad_norm": 3.171875,
      "learning_rate": 0.0004987002472130309,
      "loss": 5.7335,
      "mean_token_accuracy": 0.14573415368795395,
      "num_tokens": 9850467.0,
      "step": 4740
    },
    {
      "entropy": 5.853896808624268,
      "epoch": 0.43042452830188677,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004986967720463485,
      "loss": 5.7836,
      "mean_token_accuracy": 0.14540919959545134,
      "num_tokens": 9861115.0,
      "step": 4745
    },
    {
      "entropy": 5.881145763397217,
      "epoch": 0.430878084179971,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004986932922535598,
      "loss": 5.7928,
      "mean_token_accuracy": 0.1494162768125534,
      "num_tokens": 9871172.0,
      "step": 4750
    },
    {
      "entropy": 5.8132791996002195,
      "epoch": 0.43133164005805513,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004986898078347368,
      "loss": 5.6845,
      "mean_token_accuracy": 0.15211455523967743,
      "num_tokens": 9881997.0,
      "step": 4755
    },
    {
      "entropy": 5.795827627182007,
      "epoch": 0.43178519593613934,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004986863187899516,
      "loss": 5.6945,
      "mean_token_accuracy": 0.15255386531352996,
      "num_tokens": 9892310.0,
      "step": 4760
    },
    {
      "entropy": 5.739163732528686,
      "epoch": 0.4322387518142235,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004986828251192764,
      "loss": 5.7317,
      "mean_token_accuracy": 0.15006684958934785,
      "num_tokens": 9902664.0,
      "step": 4765
    },
    {
      "entropy": 5.765441131591797,
      "epoch": 0.4326923076923077,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004986793268227833,
      "loss": 5.7226,
      "mean_token_accuracy": 0.14934629648923875,
      "num_tokens": 9912491.0,
      "step": 4770
    },
    {
      "entropy": 5.810837459564209,
      "epoch": 0.43314586357039186,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004986758239005448,
      "loss": 5.6559,
      "mean_token_accuracy": 0.15213295966386794,
      "num_tokens": 9922057.0,
      "step": 4775
    },
    {
      "entropy": 5.772084760665893,
      "epoch": 0.4335994194484761,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004986723163526331,
      "loss": 5.7265,
      "mean_token_accuracy": 0.1473941296339035,
      "num_tokens": 9933661.0,
      "step": 4780
    },
    {
      "entropy": 5.798309230804444,
      "epoch": 0.43405297532656023,
      "grad_norm": 1.5,
      "learning_rate": 0.0004986688041791209,
      "loss": 5.7003,
      "mean_token_accuracy": 0.14621387347579,
      "num_tokens": 9945177.0,
      "step": 4785
    },
    {
      "entropy": 5.851141834259034,
      "epoch": 0.4345065312046444,
      "grad_norm": 1.78125,
      "learning_rate": 0.000498665287380081,
      "loss": 5.7308,
      "mean_token_accuracy": 0.14761214107275009,
      "num_tokens": 9956226.0,
      "step": 4790
    },
    {
      "entropy": 5.874855804443359,
      "epoch": 0.4349600870827286,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004986617659555859,
      "loss": 5.7916,
      "mean_token_accuracy": 0.14956312030553817,
      "num_tokens": 9965877.0,
      "step": 4795
    },
    {
      "entropy": 5.805899953842163,
      "epoch": 0.43541364296081275,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004986582399057085,
      "loss": 5.7519,
      "mean_token_accuracy": 0.160671429336071,
      "num_tokens": 9975592.0,
      "step": 4800
    },
    {
      "entropy": 5.817458915710449,
      "epoch": 0.43586719883889696,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004986547092305216,
      "loss": 5.7795,
      "mean_token_accuracy": 0.14750458002090455,
      "num_tokens": 9986094.0,
      "step": 4805
    },
    {
      "entropy": 5.865862941741943,
      "epoch": 0.4363207547169811,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004986511739300984,
      "loss": 5.8163,
      "mean_token_accuracy": 0.14368128031492233,
      "num_tokens": 9997241.0,
      "step": 4810
    },
    {
      "entropy": 5.855591487884522,
      "epoch": 0.43677431059506533,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004986476340045119,
      "loss": 5.7254,
      "mean_token_accuracy": 0.1495523065328598,
      "num_tokens": 10008091.0,
      "step": 4815
    },
    {
      "entropy": 5.795777940750122,
      "epoch": 0.4372278664731495,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004986440894538354,
      "loss": 5.7912,
      "mean_token_accuracy": 0.14798561334609986,
      "num_tokens": 10018794.0,
      "step": 4820
    },
    {
      "entropy": 5.776800107955933,
      "epoch": 0.4376814223512337,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004986405402781421,
      "loss": 5.6904,
      "mean_token_accuracy": 0.14998709559440612,
      "num_tokens": 10028414.0,
      "step": 4825
    },
    {
      "entropy": 5.859041595458985,
      "epoch": 0.43813497822931785,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004986369864775054,
      "loss": 5.7709,
      "mean_token_accuracy": 0.14849811941385269,
      "num_tokens": 10038749.0,
      "step": 4830
    },
    {
      "entropy": 5.881982755661011,
      "epoch": 0.438588534107402,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004986334280519989,
      "loss": 5.7227,
      "mean_token_accuracy": 0.14857089072465895,
      "num_tokens": 10049866.0,
      "step": 4835
    },
    {
      "entropy": 5.816167640686035,
      "epoch": 0.4390420899854862,
      "grad_norm": 2.53125,
      "learning_rate": 0.000498629865001696,
      "loss": 5.7514,
      "mean_token_accuracy": 0.1548391565680504,
      "num_tokens": 10060680.0,
      "step": 4840
    },
    {
      "entropy": 5.862650632858276,
      "epoch": 0.43949564586357037,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004986262973266706,
      "loss": 5.8481,
      "mean_token_accuracy": 0.14525855779647828,
      "num_tokens": 10070931.0,
      "step": 4845
    },
    {
      "entropy": 5.7950972557067875,
      "epoch": 0.4399492017416546,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004986227250269964,
      "loss": 5.7682,
      "mean_token_accuracy": 0.1551634520292282,
      "num_tokens": 10080189.0,
      "step": 4850
    },
    {
      "entropy": 5.814848613739014,
      "epoch": 0.44040275761973874,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004986191481027472,
      "loss": 5.6731,
      "mean_token_accuracy": 0.15222601294517518,
      "num_tokens": 10088813.0,
      "step": 4855
    },
    {
      "entropy": 5.785250854492188,
      "epoch": 0.44085631349782295,
      "grad_norm": 3.125,
      "learning_rate": 0.0004986155665539971,
      "loss": 5.7113,
      "mean_token_accuracy": 0.15193801522254943,
      "num_tokens": 10098125.0,
      "step": 4860
    },
    {
      "entropy": 5.8375824928283695,
      "epoch": 0.4413098693759071,
      "grad_norm": 1.71875,
      "learning_rate": 0.00049861198038082,
      "loss": 5.7335,
      "mean_token_accuracy": 0.1524299070239067,
      "num_tokens": 10109125.0,
      "step": 4865
    },
    {
      "entropy": 5.789612865447998,
      "epoch": 0.4417634252539913,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004986083895832903,
      "loss": 5.7042,
      "mean_token_accuracy": 0.15111606419086457,
      "num_tokens": 10120182.0,
      "step": 4870
    },
    {
      "entropy": 5.803175401687622,
      "epoch": 0.44221698113207547,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004986047941614821,
      "loss": 5.7837,
      "mean_token_accuracy": 0.14436635076999665,
      "num_tokens": 10130941.0,
      "step": 4875
    },
    {
      "entropy": 5.944940805435181,
      "epoch": 0.4426705370101596,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004986011941154696,
      "loss": 5.7612,
      "mean_token_accuracy": 0.14681328535079957,
      "num_tokens": 10141708.0,
      "step": 4880
    },
    {
      "entropy": 5.836460781097412,
      "epoch": 0.44312409288824384,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004985975894453275,
      "loss": 5.6717,
      "mean_token_accuracy": 0.1519172064960003,
      "num_tokens": 10151816.0,
      "step": 4885
    },
    {
      "entropy": 5.8214634418487545,
      "epoch": 0.443577648766328,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004985939801511304,
      "loss": 5.7987,
      "mean_token_accuracy": 0.15131675601005554,
      "num_tokens": 10162470.0,
      "step": 4890
    },
    {
      "entropy": 5.886533737182617,
      "epoch": 0.4440312046444122,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004985903662329527,
      "loss": 5.6955,
      "mean_token_accuracy": 0.1570177286863327,
      "num_tokens": 10173171.0,
      "step": 4895
    },
    {
      "entropy": 5.8201783180236815,
      "epoch": 0.44448476052249636,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004985867476908693,
      "loss": 5.7265,
      "mean_token_accuracy": 0.1524544969201088,
      "num_tokens": 10184185.0,
      "step": 4900
    },
    {
      "entropy": 5.7874305725097654,
      "epoch": 0.44493831640058057,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004985831245249549,
      "loss": 5.7848,
      "mean_token_accuracy": 0.14795219525694847,
      "num_tokens": 10195516.0,
      "step": 4905
    },
    {
      "entropy": 5.796139764785766,
      "epoch": 0.4453918722786647,
      "grad_norm": 3.34375,
      "learning_rate": 0.0004985794967352846,
      "loss": 5.7202,
      "mean_token_accuracy": 0.15106817930936814,
      "num_tokens": 10204936.0,
      "step": 4910
    },
    {
      "entropy": 5.855431652069091,
      "epoch": 0.44584542815674894,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004985758643219334,
      "loss": 5.7589,
      "mean_token_accuracy": 0.14915177449584008,
      "num_tokens": 10216218.0,
      "step": 4915
    },
    {
      "entropy": 5.896589231491089,
      "epoch": 0.4462989840348331,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004985722272849762,
      "loss": 5.7595,
      "mean_token_accuracy": 0.1486516922712326,
      "num_tokens": 10226370.0,
      "step": 4920
    },
    {
      "entropy": 5.761085844039917,
      "epoch": 0.44675253991291725,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004985685856244884,
      "loss": 5.6748,
      "mean_token_accuracy": 0.15937331169843674,
      "num_tokens": 10236170.0,
      "step": 4925
    },
    {
      "entropy": 5.8097490787506105,
      "epoch": 0.44720609579100146,
      "grad_norm": 3.734375,
      "learning_rate": 0.0004985649393405453,
      "loss": 5.7641,
      "mean_token_accuracy": 0.1467866986989975,
      "num_tokens": 10246119.0,
      "step": 4930
    },
    {
      "entropy": 5.838061428070068,
      "epoch": 0.4476596516690856,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004985612884332223,
      "loss": 5.7597,
      "mean_token_accuracy": 0.14624503403902053,
      "num_tokens": 10256762.0,
      "step": 4935
    },
    {
      "entropy": 5.759787511825562,
      "epoch": 0.4481132075471698,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004985576329025949,
      "loss": 5.7058,
      "mean_token_accuracy": 0.1520256593823433,
      "num_tokens": 10267317.0,
      "step": 4940
    },
    {
      "entropy": 5.869536733627319,
      "epoch": 0.448566763425254,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004985539727487385,
      "loss": 5.8374,
      "mean_token_accuracy": 0.13999374434351922,
      "num_tokens": 10279111.0,
      "step": 4945
    },
    {
      "entropy": 5.854530906677246,
      "epoch": 0.4490203193033382,
      "grad_norm": 3.28125,
      "learning_rate": 0.0004985503079717292,
      "loss": 5.718,
      "mean_token_accuracy": 0.15307011604309081,
      "num_tokens": 10289383.0,
      "step": 4950
    },
    {
      "entropy": 5.875187730789184,
      "epoch": 0.44947387518142234,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004985466385716425,
      "loss": 5.7275,
      "mean_token_accuracy": 0.1502274066209793,
      "num_tokens": 10299806.0,
      "step": 4955
    },
    {
      "entropy": 5.806089878082275,
      "epoch": 0.44992743105950656,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004985429645485543,
      "loss": 5.7382,
      "mean_token_accuracy": 0.1461896352469921,
      "num_tokens": 10310348.0,
      "step": 4960
    },
    {
      "entropy": 5.827132749557495,
      "epoch": 0.4503809869375907,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004985392859025407,
      "loss": 5.7644,
      "mean_token_accuracy": 0.1516891375184059,
      "num_tokens": 10320354.0,
      "step": 4965
    },
    {
      "entropy": 5.757310485839843,
      "epoch": 0.45083454281567487,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004985356026336776,
      "loss": 5.6703,
      "mean_token_accuracy": 0.1507007971405983,
      "num_tokens": 10330292.0,
      "step": 4970
    },
    {
      "entropy": 5.858229494094848,
      "epoch": 0.4512880986937591,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004985319147420414,
      "loss": 5.6474,
      "mean_token_accuracy": 0.15889344215393067,
      "num_tokens": 10340449.0,
      "step": 4975
    },
    {
      "entropy": 5.822214269638062,
      "epoch": 0.45174165457184323,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004985282222277082,
      "loss": 5.6964,
      "mean_token_accuracy": 0.14781344383955003,
      "num_tokens": 10350637.0,
      "step": 4980
    },
    {
      "entropy": 5.680160665512085,
      "epoch": 0.45219521044992744,
      "grad_norm": 2.0,
      "learning_rate": 0.0004985245250907544,
      "loss": 5.6621,
      "mean_token_accuracy": 0.15216653496026994,
      "num_tokens": 10361548.0,
      "step": 4985
    },
    {
      "entropy": 5.7882819175720215,
      "epoch": 0.4526487663280116,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004985208233312564,
      "loss": 5.6475,
      "mean_token_accuracy": 0.14174009189009668,
      "num_tokens": 10371765.0,
      "step": 4990
    },
    {
      "entropy": 5.8632818222045895,
      "epoch": 0.4531023222060958,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004985171169492908,
      "loss": 5.7738,
      "mean_token_accuracy": 0.15304608941078185,
      "num_tokens": 10382108.0,
      "step": 4995
    },
    {
      "entropy": 5.799547719955444,
      "epoch": 0.45355587808417996,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004985134059449344,
      "loss": 5.7167,
      "mean_token_accuracy": 0.15100618749856948,
      "num_tokens": 10392390.0,
      "step": 5000
    },
    {
      "entropy": 5.829653596878051,
      "epoch": 0.4540094339622642,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004985096903182637,
      "loss": 5.7569,
      "mean_token_accuracy": 0.15035886466503143,
      "num_tokens": 10402094.0,
      "step": 5005
    },
    {
      "entropy": 5.823336267471314,
      "epoch": 0.45446298984034833,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004985059700693557,
      "loss": 5.7002,
      "mean_token_accuracy": 0.15384241491556166,
      "num_tokens": 10413350.0,
      "step": 5010
    },
    {
      "entropy": 5.836109685897827,
      "epoch": 0.4549165457184325,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004985022451982872,
      "loss": 5.8458,
      "mean_token_accuracy": 0.14465131014585494,
      "num_tokens": 10424292.0,
      "step": 5015
    },
    {
      "entropy": 5.888298034667969,
      "epoch": 0.4553701015965167,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004984985157051354,
      "loss": 5.6868,
      "mean_token_accuracy": 0.15214828550815582,
      "num_tokens": 10433724.0,
      "step": 5020
    },
    {
      "entropy": 5.819750642776489,
      "epoch": 0.45582365747460085,
      "grad_norm": 2.953125,
      "learning_rate": 0.0004984947815899774,
      "loss": 5.779,
      "mean_token_accuracy": 0.14743213057518006,
      "num_tokens": 10444724.0,
      "step": 5025
    },
    {
      "entropy": 5.838175058364868,
      "epoch": 0.45627721335268506,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004984910428528902,
      "loss": 5.7053,
      "mean_token_accuracy": 0.15772704035043716,
      "num_tokens": 10455086.0,
      "step": 5030
    },
    {
      "entropy": 5.7590735912322994,
      "epoch": 0.4567307692307692,
      "grad_norm": 2.375,
      "learning_rate": 0.0004984872994939514,
      "loss": 5.6661,
      "mean_token_accuracy": 0.1486493766307831,
      "num_tokens": 10465076.0,
      "step": 5035
    },
    {
      "entropy": 5.837244844436645,
      "epoch": 0.45718432510885343,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004984835515132382,
      "loss": 5.7028,
      "mean_token_accuracy": 0.1532436266541481,
      "num_tokens": 10474817.0,
      "step": 5040
    },
    {
      "entropy": 5.835756492614746,
      "epoch": 0.4576378809869376,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004984797989108282,
      "loss": 5.7118,
      "mean_token_accuracy": 0.14745451956987382,
      "num_tokens": 10485735.0,
      "step": 5045
    },
    {
      "entropy": 5.790848636627198,
      "epoch": 0.4580914368650218,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004984760416867991,
      "loss": 5.7919,
      "mean_token_accuracy": 0.14835846349596976,
      "num_tokens": 10496731.0,
      "step": 5050
    },
    {
      "entropy": 5.760481691360473,
      "epoch": 0.45854499274310595,
      "grad_norm": 3.8125,
      "learning_rate": 0.0004984722798412286,
      "loss": 5.6739,
      "mean_token_accuracy": 0.14979240894317628,
      "num_tokens": 10506787.0,
      "step": 5055
    },
    {
      "entropy": 5.826909589767456,
      "epoch": 0.4589985486211901,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004984685133741941,
      "loss": 5.7387,
      "mean_token_accuracy": 0.1452489860355854,
      "num_tokens": 10517170.0,
      "step": 5060
    },
    {
      "entropy": 5.772784662246704,
      "epoch": 0.4594521044992743,
      "grad_norm": 1.875,
      "learning_rate": 0.000498464742285774,
      "loss": 5.7163,
      "mean_token_accuracy": 0.15115534663200378,
      "num_tokens": 10527342.0,
      "step": 5065
    },
    {
      "entropy": 5.81494026184082,
      "epoch": 0.45990566037735847,
      "grad_norm": 1.5,
      "learning_rate": 0.000498460966576046,
      "loss": 5.7022,
      "mean_token_accuracy": 0.15441648811101913,
      "num_tokens": 10537253.0,
      "step": 5070
    },
    {
      "entropy": 5.858026695251465,
      "epoch": 0.4603592162554427,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004984571862450884,
      "loss": 5.7309,
      "mean_token_accuracy": 0.15261692702770233,
      "num_tokens": 10547249.0,
      "step": 5075
    },
    {
      "entropy": 5.862608909606934,
      "epoch": 0.46081277213352684,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004984534012929791,
      "loss": 5.7598,
      "mean_token_accuracy": 0.14884943068027495,
      "num_tokens": 10558174.0,
      "step": 5080
    },
    {
      "entropy": 5.871862173080444,
      "epoch": 0.46126632801161105,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004984496117197967,
      "loss": 5.721,
      "mean_token_accuracy": 0.15395016670227052,
      "num_tokens": 10567610.0,
      "step": 5085
    },
    {
      "entropy": 5.7651379108428955,
      "epoch": 0.4617198838896952,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004984458175256193,
      "loss": 5.7356,
      "mean_token_accuracy": 0.14666065573692322,
      "num_tokens": 10578494.0,
      "step": 5090
    },
    {
      "entropy": 5.876489353179932,
      "epoch": 0.4621734397677794,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004984420187105253,
      "loss": 5.8277,
      "mean_token_accuracy": 0.14741355702280998,
      "num_tokens": 10589482.0,
      "step": 5095
    },
    {
      "entropy": 5.92623553276062,
      "epoch": 0.46262699564586357,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004984382152745936,
      "loss": 5.7812,
      "mean_token_accuracy": 0.14485349804162978,
      "num_tokens": 10599965.0,
      "step": 5100
    },
    {
      "entropy": 5.797193765640259,
      "epoch": 0.4630805515239477,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004984344072179026,
      "loss": 5.8158,
      "mean_token_accuracy": 0.15118358284235,
      "num_tokens": 10610500.0,
      "step": 5105
    },
    {
      "entropy": 5.832291603088379,
      "epoch": 0.46353410740203194,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004984305945405312,
      "loss": 5.7608,
      "mean_token_accuracy": 0.15085830986499787,
      "num_tokens": 10619843.0,
      "step": 5110
    },
    {
      "entropy": 5.8759160995483395,
      "epoch": 0.4639876632801161,
      "grad_norm": 2.546875,
      "learning_rate": 0.000498426777242558,
      "loss": 5.6799,
      "mean_token_accuracy": 0.1561095081269741,
      "num_tokens": 10631026.0,
      "step": 5115
    },
    {
      "entropy": 5.648857736587525,
      "epoch": 0.4644412191582003,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004984229553240623,
      "loss": 5.6994,
      "mean_token_accuracy": 0.15481915920972825,
      "num_tokens": 10641130.0,
      "step": 5120
    },
    {
      "entropy": 5.7813159942626955,
      "epoch": 0.46489477503628446,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004984191287851228,
      "loss": 5.7541,
      "mean_token_accuracy": 0.1515883758664131,
      "num_tokens": 10652305.0,
      "step": 5125
    },
    {
      "entropy": 5.945868587493896,
      "epoch": 0.46534833091436867,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004984152976258188,
      "loss": 5.7772,
      "mean_token_accuracy": 0.15462889671325683,
      "num_tokens": 10663083.0,
      "step": 5130
    },
    {
      "entropy": 5.808717489242554,
      "epoch": 0.4658018867924528,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004984114618462296,
      "loss": 5.8091,
      "mean_token_accuracy": 0.14427706003189086,
      "num_tokens": 10673602.0,
      "step": 5135
    },
    {
      "entropy": 5.731534767150879,
      "epoch": 0.46625544267053703,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004984076214464343,
      "loss": 5.6296,
      "mean_token_accuracy": 0.15683090388774873,
      "num_tokens": 10684622.0,
      "step": 5140
    },
    {
      "entropy": 5.81270661354065,
      "epoch": 0.4667089985486212,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004984037764265126,
      "loss": 5.7507,
      "mean_token_accuracy": 0.14948814660310744,
      "num_tokens": 10695129.0,
      "step": 5145
    },
    {
      "entropy": 5.793086814880371,
      "epoch": 0.46716255442670535,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004983999267865438,
      "loss": 5.6087,
      "mean_token_accuracy": 0.15965820848941803,
      "num_tokens": 10705236.0,
      "step": 5150
    },
    {
      "entropy": 5.784671878814697,
      "epoch": 0.46761611030478956,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004983960725266078,
      "loss": 5.7217,
      "mean_token_accuracy": 0.14745316356420518,
      "num_tokens": 10715748.0,
      "step": 5155
    },
    {
      "entropy": 5.831699419021606,
      "epoch": 0.4680696661828737,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004983922136467838,
      "loss": 5.6555,
      "mean_token_accuracy": 0.15694286674261093,
      "num_tokens": 10725799.0,
      "step": 5160
    },
    {
      "entropy": 5.813610506057739,
      "epoch": 0.4685232220609579,
      "grad_norm": 1.8046875,
      "learning_rate": 0.000498388350147152,
      "loss": 5.751,
      "mean_token_accuracy": 0.14641758650541306,
      "num_tokens": 10735668.0,
      "step": 5165
    },
    {
      "entropy": 5.861870384216308,
      "epoch": 0.4689767779390421,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004983844820277922,
      "loss": 5.87,
      "mean_token_accuracy": 0.1426786296069622,
      "num_tokens": 10746466.0,
      "step": 5170
    },
    {
      "entropy": 5.888263511657715,
      "epoch": 0.4694303338171263,
      "grad_norm": 3.140625,
      "learning_rate": 0.0004983806092887843,
      "loss": 5.7561,
      "mean_token_accuracy": 0.15089037865400315,
      "num_tokens": 10756724.0,
      "step": 5175
    },
    {
      "entropy": 5.855864477157593,
      "epoch": 0.46988388969521044,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004983767319302086,
      "loss": 5.8112,
      "mean_token_accuracy": 0.1499728061258793,
      "num_tokens": 10766919.0,
      "step": 5180
    },
    {
      "entropy": 5.766339826583862,
      "epoch": 0.47033744557329465,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004983728499521451,
      "loss": 5.6087,
      "mean_token_accuracy": 0.1593857705593109,
      "num_tokens": 10777550.0,
      "step": 5185
    },
    {
      "entropy": 5.767188262939453,
      "epoch": 0.4707910014513788,
      "grad_norm": 1.625,
      "learning_rate": 0.0004983689633546741,
      "loss": 5.6586,
      "mean_token_accuracy": 0.14953138381242753,
      "num_tokens": 10788456.0,
      "step": 5190
    },
    {
      "entropy": 5.820877647399902,
      "epoch": 0.47124455732946297,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004983650721378761,
      "loss": 5.7126,
      "mean_token_accuracy": 0.1494821533560753,
      "num_tokens": 10800049.0,
      "step": 5195
    },
    {
      "entropy": 5.8708991527557375,
      "epoch": 0.4716981132075472,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004983611763018314,
      "loss": 5.7514,
      "mean_token_accuracy": 0.1483040362596512,
      "num_tokens": 10810743.0,
      "step": 5200
    },
    {
      "entropy": 5.935766887664795,
      "epoch": 0.47215166908563133,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004983572758466208,
      "loss": 5.9259,
      "mean_token_accuracy": 0.14382343962788582,
      "num_tokens": 10821897.0,
      "step": 5205
    },
    {
      "entropy": 5.911200714111328,
      "epoch": 0.47260522496371554,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004983533707723247,
      "loss": 5.78,
      "mean_token_accuracy": 0.15174445733428002,
      "num_tokens": 10831589.0,
      "step": 5210
    },
    {
      "entropy": 5.881593132019043,
      "epoch": 0.4730587808417997,
      "grad_norm": 2.125,
      "learning_rate": 0.0004983494610790241,
      "loss": 5.6938,
      "mean_token_accuracy": 0.14457672983407974,
      "num_tokens": 10841404.0,
      "step": 5215
    },
    {
      "entropy": 5.787980270385742,
      "epoch": 0.4735123367198839,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004983455467667996,
      "loss": 5.7393,
      "mean_token_accuracy": 0.1467074014246464,
      "num_tokens": 10851577.0,
      "step": 5220
    },
    {
      "entropy": 5.788097476959228,
      "epoch": 0.47396589259796806,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004983416278357322,
      "loss": 5.6485,
      "mean_token_accuracy": 0.1567993551492691,
      "num_tokens": 10861403.0,
      "step": 5225
    },
    {
      "entropy": 5.844185018539429,
      "epoch": 0.4744194484760523,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004983377042859032,
      "loss": 5.702,
      "mean_token_accuracy": 0.14790162742137908,
      "num_tokens": 10871933.0,
      "step": 5230
    },
    {
      "entropy": 5.761803388595581,
      "epoch": 0.47487300435413643,
      "grad_norm": 2.984375,
      "learning_rate": 0.0004983337761173936,
      "loss": 5.773,
      "mean_token_accuracy": 0.14696283340454103,
      "num_tokens": 10881800.0,
      "step": 5235
    },
    {
      "entropy": 5.915017938613891,
      "epoch": 0.4753265602322206,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004983298433302843,
      "loss": 5.74,
      "mean_token_accuracy": 0.14846391826868058,
      "num_tokens": 10891830.0,
      "step": 5240
    },
    {
      "entropy": 5.812942409515381,
      "epoch": 0.4757801161103048,
      "grad_norm": 2.0,
      "learning_rate": 0.0004983259059246572,
      "loss": 5.655,
      "mean_token_accuracy": 0.15466391891241074,
      "num_tokens": 10901557.0,
      "step": 5245
    },
    {
      "entropy": 5.7293657779693605,
      "epoch": 0.47623367198838895,
      "grad_norm": 4.0625,
      "learning_rate": 0.0004983219639005934,
      "loss": 5.764,
      "mean_token_accuracy": 0.14264445900917053,
      "num_tokens": 10912155.0,
      "step": 5250
    },
    {
      "entropy": 5.743430709838867,
      "epoch": 0.47668722786647316,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004983180172581745,
      "loss": 5.6787,
      "mean_token_accuracy": 0.15394465178251265,
      "num_tokens": 10922291.0,
      "step": 5255
    },
    {
      "entropy": 5.898777866363526,
      "epoch": 0.4771407837445573,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004983140659974819,
      "loss": 5.7493,
      "mean_token_accuracy": 0.1463671863079071,
      "num_tokens": 10933176.0,
      "step": 5260
    },
    {
      "entropy": 5.883290433883667,
      "epoch": 0.47759433962264153,
      "grad_norm": 2.96875,
      "learning_rate": 0.0004983101101185977,
      "loss": 5.8327,
      "mean_token_accuracy": 0.145734940469265,
      "num_tokens": 10944049.0,
      "step": 5265
    },
    {
      "entropy": 5.773307657241821,
      "epoch": 0.4780478955007257,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004983061496216035,
      "loss": 5.6794,
      "mean_token_accuracy": 0.1519426852464676,
      "num_tokens": 10954347.0,
      "step": 5270
    },
    {
      "entropy": 5.770019006729126,
      "epoch": 0.4785014513788099,
      "grad_norm": 2.375,
      "learning_rate": 0.0004983021845065812,
      "loss": 5.7304,
      "mean_token_accuracy": 0.1466204196214676,
      "num_tokens": 10965123.0,
      "step": 5275
    },
    {
      "entropy": 5.798799276351929,
      "epoch": 0.47895500725689405,
      "grad_norm": 3.578125,
      "learning_rate": 0.0004982982147736128,
      "loss": 5.7335,
      "mean_token_accuracy": 0.1530810609459877,
      "num_tokens": 10976395.0,
      "step": 5280
    },
    {
      "entropy": 5.84113039970398,
      "epoch": 0.4794085631349782,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004982942404227805,
      "loss": 5.6978,
      "mean_token_accuracy": 0.1490973189473152,
      "num_tokens": 10986797.0,
      "step": 5285
    },
    {
      "entropy": 5.87009449005127,
      "epoch": 0.4798621190130624,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004982902614541662,
      "loss": 5.7569,
      "mean_token_accuracy": 0.14732319861650467,
      "num_tokens": 10997229.0,
      "step": 5290
    },
    {
      "entropy": 5.840237808227539,
      "epoch": 0.48031567489114657,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004982862778678526,
      "loss": 5.6982,
      "mean_token_accuracy": 0.15742301493883132,
      "num_tokens": 11008180.0,
      "step": 5295
    },
    {
      "entropy": 5.926225042343139,
      "epoch": 0.4807692307692308,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004982822896639218,
      "loss": 5.7236,
      "mean_token_accuracy": 0.1533257931470871,
      "num_tokens": 11019248.0,
      "step": 5300
    },
    {
      "entropy": 5.791766881942749,
      "epoch": 0.48122278664731494,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004982782968424563,
      "loss": 5.649,
      "mean_token_accuracy": 0.15525579005479812,
      "num_tokens": 11029760.0,
      "step": 5305
    },
    {
      "entropy": 5.7674986839294435,
      "epoch": 0.48167634252539915,
      "grad_norm": 3.171875,
      "learning_rate": 0.0004982742994035388,
      "loss": 5.7206,
      "mean_token_accuracy": 0.14568620324134826,
      "num_tokens": 11039211.0,
      "step": 5310
    },
    {
      "entropy": 5.839351844787598,
      "epoch": 0.4821298984034833,
      "grad_norm": 2.0,
      "learning_rate": 0.0004982702973472518,
      "loss": 5.6961,
      "mean_token_accuracy": 0.1473838321864605,
      "num_tokens": 11049892.0,
      "step": 5315
    },
    {
      "entropy": 5.943693685531616,
      "epoch": 0.4825834542815675,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004982662906736783,
      "loss": 5.7621,
      "mean_token_accuracy": 0.14789720624685287,
      "num_tokens": 11060823.0,
      "step": 5320
    },
    {
      "entropy": 5.757358884811401,
      "epoch": 0.48303701015965167,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004982622793829009,
      "loss": 5.7463,
      "mean_token_accuracy": 0.15420066863298415,
      "num_tokens": 11072353.0,
      "step": 5325
    },
    {
      "entropy": 5.798610353469849,
      "epoch": 0.4834905660377358,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004982582634750027,
      "loss": 5.7571,
      "mean_token_accuracy": 0.14944387078285218,
      "num_tokens": 11082655.0,
      "step": 5330
    },
    {
      "entropy": 5.828657102584839,
      "epoch": 0.48394412191582004,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004982542429500667,
      "loss": 5.7616,
      "mean_token_accuracy": 0.15139087289571762,
      "num_tokens": 11093315.0,
      "step": 5335
    },
    {
      "entropy": 5.714878511428833,
      "epoch": 0.4843976777939042,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004982502178081761,
      "loss": 5.6677,
      "mean_token_accuracy": 0.15089037865400315,
      "num_tokens": 11103074.0,
      "step": 5340
    },
    {
      "entropy": 5.84651894569397,
      "epoch": 0.4848512336719884,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004982461880494141,
      "loss": 5.7078,
      "mean_token_accuracy": 0.15072353780269623,
      "num_tokens": 11113636.0,
      "step": 5345
    },
    {
      "entropy": 5.837501668930054,
      "epoch": 0.48530478955007256,
      "grad_norm": 1.9765625,
      "learning_rate": 0.000498242153673864,
      "loss": 5.6617,
      "mean_token_accuracy": 0.15472064316272735,
      "num_tokens": 11123509.0,
      "step": 5350
    },
    {
      "entropy": 5.811651086807251,
      "epoch": 0.48575834542815677,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004982381146816094,
      "loss": 5.732,
      "mean_token_accuracy": 0.1466941475868225,
      "num_tokens": 11133405.0,
      "step": 5355
    },
    {
      "entropy": 5.874093151092529,
      "epoch": 0.4862119013062409,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004982340710727336,
      "loss": 5.784,
      "mean_token_accuracy": 0.1411792032420635,
      "num_tokens": 11144725.0,
      "step": 5360
    },
    {
      "entropy": 5.835286045074463,
      "epoch": 0.48666545718432513,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004982300228473203,
      "loss": 5.6809,
      "mean_token_accuracy": 0.14511124938726425,
      "num_tokens": 11155446.0,
      "step": 5365
    },
    {
      "entropy": 5.843464946746826,
      "epoch": 0.4871190130624093,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004982259700054533,
      "loss": 5.7373,
      "mean_token_accuracy": 0.15083874613046647,
      "num_tokens": 11165380.0,
      "step": 5370
    },
    {
      "entropy": 5.833011054992676,
      "epoch": 0.48757256894049344,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004982219125472163,
      "loss": 5.7805,
      "mean_token_accuracy": 0.15250826701521875,
      "num_tokens": 11175641.0,
      "step": 5375
    },
    {
      "entropy": 5.952740001678467,
      "epoch": 0.48802612481857766,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004982178504726933,
      "loss": 5.9333,
      "mean_token_accuracy": 0.1492827832698822,
      "num_tokens": 11186257.0,
      "step": 5380
    },
    {
      "entropy": 5.917461729049682,
      "epoch": 0.4884796806966618,
      "grad_norm": 2.125,
      "learning_rate": 0.0004982137837819682,
      "loss": 5.7581,
      "mean_token_accuracy": 0.14257122650742532,
      "num_tokens": 11197284.0,
      "step": 5385
    },
    {
      "entropy": 5.890596342086792,
      "epoch": 0.488933236574746,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004982097124751252,
      "loss": 5.7259,
      "mean_token_accuracy": 0.15704453438520433,
      "num_tokens": 11208413.0,
      "step": 5390
    },
    {
      "entropy": 5.756617689132691,
      "epoch": 0.4893867924528302,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004982056365522486,
      "loss": 5.7385,
      "mean_token_accuracy": 0.15123408585786818,
      "num_tokens": 11219016.0,
      "step": 5395
    },
    {
      "entropy": 5.743880319595337,
      "epoch": 0.4898403483309144,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004982015560134225,
      "loss": 5.7147,
      "mean_token_accuracy": 0.1525794416666031,
      "num_tokens": 11228934.0,
      "step": 5400
    },
    {
      "entropy": 5.801280117034912,
      "epoch": 0.49029390420899854,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004981974708587312,
      "loss": 5.6416,
      "mean_token_accuracy": 0.15183479487895965,
      "num_tokens": 11239596.0,
      "step": 5405
    },
    {
      "entropy": 5.771134996414185,
      "epoch": 0.49074746008708275,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004981933810882594,
      "loss": 5.6322,
      "mean_token_accuracy": 0.14867930263280868,
      "num_tokens": 11249560.0,
      "step": 5410
    },
    {
      "entropy": 5.860190916061401,
      "epoch": 0.4912010159651669,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004981892867020917,
      "loss": 5.8181,
      "mean_token_accuracy": 0.14608463272452354,
      "num_tokens": 11259787.0,
      "step": 5415
    },
    {
      "entropy": 5.8484776496887205,
      "epoch": 0.49165457184325106,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004981851877003125,
      "loss": 5.7599,
      "mean_token_accuracy": 0.14876429736614227,
      "num_tokens": 11269717.0,
      "step": 5420
    },
    {
      "entropy": 5.834319734573365,
      "epoch": 0.4921081277213353,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004981810840830068,
      "loss": 5.7344,
      "mean_token_accuracy": 0.15793785825371742,
      "num_tokens": 11279863.0,
      "step": 5425
    },
    {
      "entropy": 5.850916337966919,
      "epoch": 0.49256168359941943,
      "grad_norm": 2.90625,
      "learning_rate": 0.0004981769758502596,
      "loss": 5.8704,
      "mean_token_accuracy": 0.1406077429652214,
      "num_tokens": 11291161.0,
      "step": 5430
    },
    {
      "entropy": 5.7810022830963135,
      "epoch": 0.49301523947750364,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004981728630021555,
      "loss": 5.6669,
      "mean_token_accuracy": 0.15320780575275422,
      "num_tokens": 11301151.0,
      "step": 5435
    },
    {
      "entropy": 5.785669231414795,
      "epoch": 0.4934687953555878,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004981687455387797,
      "loss": 5.6748,
      "mean_token_accuracy": 0.15113574340939523,
      "num_tokens": 11311466.0,
      "step": 5440
    },
    {
      "entropy": 5.812684631347656,
      "epoch": 0.493922351233672,
      "grad_norm": 3.765625,
      "learning_rate": 0.0004981646234602173,
      "loss": 5.6467,
      "mean_token_accuracy": 0.15649000257253648,
      "num_tokens": 11321522.0,
      "step": 5445
    },
    {
      "entropy": 5.847482442855835,
      "epoch": 0.49437590711175616,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004981604967665539,
      "loss": 5.761,
      "mean_token_accuracy": 0.14855778515338897,
      "num_tokens": 11332317.0,
      "step": 5450
    },
    {
      "entropy": 5.819078731536865,
      "epoch": 0.4948294629898404,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004981563654578743,
      "loss": 5.6897,
      "mean_token_accuracy": 0.14819339513778687,
      "num_tokens": 11343024.0,
      "step": 5455
    },
    {
      "entropy": 5.848114490509033,
      "epoch": 0.49528301886792453,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004981522295342643,
      "loss": 5.6589,
      "mean_token_accuracy": 0.15384979099035262,
      "num_tokens": 11353114.0,
      "step": 5460
    },
    {
      "entropy": 5.850389432907105,
      "epoch": 0.4957365747460087,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004981480889958092,
      "loss": 5.7088,
      "mean_token_accuracy": 0.15356869697570802,
      "num_tokens": 11363446.0,
      "step": 5465
    },
    {
      "entropy": 5.791082286834717,
      "epoch": 0.4961901306240929,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004981439438425949,
      "loss": 5.7092,
      "mean_token_accuracy": 0.15395713597536087,
      "num_tokens": 11373413.0,
      "step": 5470
    },
    {
      "entropy": 5.702733945846558,
      "epoch": 0.49664368650217705,
      "grad_norm": 3.265625,
      "learning_rate": 0.000498139794074707,
      "loss": 5.5953,
      "mean_token_accuracy": 0.15706127732992173,
      "num_tokens": 11383142.0,
      "step": 5475
    },
    {
      "entropy": 5.815199375152588,
      "epoch": 0.49709724238026126,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004981356396922313,
      "loss": 5.7352,
      "mean_token_accuracy": 0.14936144053936004,
      "num_tokens": 11394635.0,
      "step": 5480
    },
    {
      "entropy": 5.790682744979859,
      "epoch": 0.4975507982583454,
      "grad_norm": 3.125,
      "learning_rate": 0.0004981314806952535,
      "loss": 5.641,
      "mean_token_accuracy": 0.15651369094848633,
      "num_tokens": 11404442.0,
      "step": 5485
    },
    {
      "entropy": 5.753213262557983,
      "epoch": 0.49800435413642963,
      "grad_norm": 2.703125,
      "learning_rate": 0.00049812731708386,
      "loss": 5.6012,
      "mean_token_accuracy": 0.151226407289505,
      "num_tokens": 11414217.0,
      "step": 5490
    },
    {
      "entropy": 5.8672788619995115,
      "epoch": 0.4984579100145138,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004981231488581367,
      "loss": 5.7436,
      "mean_token_accuracy": 0.1472836948931217,
      "num_tokens": 11423834.0,
      "step": 5495
    },
    {
      "entropy": 5.754918384552002,
      "epoch": 0.498911465892598,
      "grad_norm": 4.875,
      "learning_rate": 0.0004981189760181699,
      "loss": 5.6162,
      "mean_token_accuracy": 0.1506452038884163,
      "num_tokens": 11433613.0,
      "step": 5500
    },
    {
      "entropy": 5.744574022293091,
      "epoch": 0.49936502177068215,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004981147985640456,
      "loss": 5.6953,
      "mean_token_accuracy": 0.15378040224313735,
      "num_tokens": 11443915.0,
      "step": 5505
    },
    {
      "entropy": 5.902297592163086,
      "epoch": 0.4998185776487663,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004981106164958506,
      "loss": 5.7512,
      "mean_token_accuracy": 0.14607876390218735,
      "num_tokens": 11454588.0,
      "step": 5510
    },
    {
      "entropy": 5.8737144947052,
      "epoch": 0.5002721335268505,
      "grad_norm": 2.9375,
      "learning_rate": 0.0004981064298136712,
      "loss": 5.7896,
      "mean_token_accuracy": 0.14923810511827468,
      "num_tokens": 11465533.0,
      "step": 5515
    },
    {
      "entropy": 5.816037034988403,
      "epoch": 0.5007256894049347,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004981022385175939,
      "loss": 5.7349,
      "mean_token_accuracy": 0.14456416741013528,
      "num_tokens": 11475776.0,
      "step": 5520
    },
    {
      "entropy": 5.810181140899658,
      "epoch": 0.5011792452830188,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004980980426077054,
      "loss": 5.7047,
      "mean_token_accuracy": 0.15837073922157288,
      "num_tokens": 11486200.0,
      "step": 5525
    },
    {
      "entropy": 5.8160165309906,
      "epoch": 0.5016328011611031,
      "grad_norm": 3.59375,
      "learning_rate": 0.0004980938420840926,
      "loss": 5.739,
      "mean_token_accuracy": 0.14561933577060698,
      "num_tokens": 11496768.0,
      "step": 5530
    },
    {
      "entropy": 5.837372159957885,
      "epoch": 0.5020863570391872,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004980896369468423,
      "loss": 5.7015,
      "mean_token_accuracy": 0.15523829013109208,
      "num_tokens": 11507096.0,
      "step": 5535
    },
    {
      "entropy": 5.847644376754761,
      "epoch": 0.5025399129172714,
      "grad_norm": 4.46875,
      "learning_rate": 0.0004980854271960415,
      "loss": 5.7618,
      "mean_token_accuracy": 0.15393387004733086,
      "num_tokens": 11517301.0,
      "step": 5540
    },
    {
      "entropy": 5.862361669540405,
      "epoch": 0.5029934687953556,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004980812128317772,
      "loss": 5.8145,
      "mean_token_accuracy": 0.14930694699287414,
      "num_tokens": 11528616.0,
      "step": 5545
    },
    {
      "entropy": 5.91984920501709,
      "epoch": 0.5034470246734397,
      "grad_norm": 3.046875,
      "learning_rate": 0.0004980769938541364,
      "loss": 5.7421,
      "mean_token_accuracy": 0.15533946603536605,
      "num_tokens": 11539344.0,
      "step": 5550
    },
    {
      "entropy": 5.893903303146362,
      "epoch": 0.503900580551524,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004980727702632066,
      "loss": 5.8022,
      "mean_token_accuracy": 0.14859232157468796,
      "num_tokens": 11549511.0,
      "step": 5555
    },
    {
      "entropy": 5.845568323135376,
      "epoch": 0.5043541364296081,
      "grad_norm": 3.671875,
      "learning_rate": 0.0004980685420590752,
      "loss": 5.7162,
      "mean_token_accuracy": 0.14990185871720313,
      "num_tokens": 11559425.0,
      "step": 5560
    },
    {
      "entropy": 5.871610403060913,
      "epoch": 0.5048076923076923,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004980643092418294,
      "loss": 5.8337,
      "mean_token_accuracy": 0.1418799228966236,
      "num_tokens": 11570406.0,
      "step": 5565
    },
    {
      "entropy": 5.810862731933594,
      "epoch": 0.5052612481857764,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004980600718115568,
      "loss": 5.7154,
      "mean_token_accuracy": 0.15226266756653786,
      "num_tokens": 11580509.0,
      "step": 5570
    },
    {
      "entropy": 5.8717328071594235,
      "epoch": 0.5057148040638607,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004980558297683452,
      "loss": 5.7229,
      "mean_token_accuracy": 0.15168080031871795,
      "num_tokens": 11590860.0,
      "step": 5575
    },
    {
      "entropy": 5.922792911529541,
      "epoch": 0.5061683599419449,
      "grad_norm": 3.828125,
      "learning_rate": 0.0004980515831122822,
      "loss": 5.7336,
      "mean_token_accuracy": 0.15418553948402405,
      "num_tokens": 11601954.0,
      "step": 5580
    },
    {
      "entropy": 5.866706800460816,
      "epoch": 0.506621915820029,
      "grad_norm": 3.34375,
      "learning_rate": 0.0004980473318434555,
      "loss": 5.7499,
      "mean_token_accuracy": 0.1491977170109749,
      "num_tokens": 11612581.0,
      "step": 5585
    },
    {
      "entropy": 5.79789137840271,
      "epoch": 0.5070754716981132,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004980430759619532,
      "loss": 5.6749,
      "mean_token_accuracy": 0.15472817420959473,
      "num_tokens": 11622599.0,
      "step": 5590
    },
    {
      "entropy": 5.773746681213379,
      "epoch": 0.5075290275761973,
      "grad_norm": 5.09375,
      "learning_rate": 0.0004980388154678633,
      "loss": 5.682,
      "mean_token_accuracy": 0.15070171654224396,
      "num_tokens": 11633611.0,
      "step": 5595
    },
    {
      "entropy": 5.802550792694092,
      "epoch": 0.5079825834542816,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004980345503612738,
      "loss": 5.6761,
      "mean_token_accuracy": 0.14901446551084518,
      "num_tokens": 11643159.0,
      "step": 5600
    },
    {
      "entropy": 5.781796932220459,
      "epoch": 0.5084361393323658,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004980302806422729,
      "loss": 5.6478,
      "mean_token_accuracy": 0.15606949031352996,
      "num_tokens": 11653174.0,
      "step": 5605
    },
    {
      "entropy": 5.798026847839355,
      "epoch": 0.5088896952104499,
      "grad_norm": 3.375,
      "learning_rate": 0.0004980260063109491,
      "loss": 5.694,
      "mean_token_accuracy": 0.15325014442205429,
      "num_tokens": 11662621.0,
      "step": 5610
    },
    {
      "entropy": 5.7645632266998295,
      "epoch": 0.5093432510885341,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004980217273673906,
      "loss": 5.6488,
      "mean_token_accuracy": 0.1541965365409851,
      "num_tokens": 11673538.0,
      "step": 5615
    },
    {
      "entropy": 5.807270860671997,
      "epoch": 0.5097968069666183,
      "grad_norm": 2.5,
      "learning_rate": 0.0004980174438116859,
      "loss": 5.6645,
      "mean_token_accuracy": 0.15759781897068023,
      "num_tokens": 11683203.0,
      "step": 5620
    },
    {
      "entropy": 5.808385896682739,
      "epoch": 0.5102503628447025,
      "grad_norm": 5.0,
      "learning_rate": 0.0004980131556439236,
      "loss": 5.6727,
      "mean_token_accuracy": 0.15713814795017242,
      "num_tokens": 11693313.0,
      "step": 5625
    },
    {
      "entropy": 5.834480953216553,
      "epoch": 0.5107039187227866,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004980088628641925,
      "loss": 5.7533,
      "mean_token_accuracy": 0.15072424262762069,
      "num_tokens": 11704194.0,
      "step": 5630
    },
    {
      "entropy": 5.80663833618164,
      "epoch": 0.5111574746008708,
      "grad_norm": 3.515625,
      "learning_rate": 0.0004980045654725812,
      "loss": 5.6112,
      "mean_token_accuracy": 0.15680344700813292,
      "num_tokens": 11715109.0,
      "step": 5635
    },
    {
      "entropy": 5.749876594543457,
      "epoch": 0.511611030478955,
      "grad_norm": 3.84375,
      "learning_rate": 0.0004980002634691787,
      "loss": 5.6527,
      "mean_token_accuracy": 0.15471058934926987,
      "num_tokens": 11724961.0,
      "step": 5640
    },
    {
      "entropy": 5.743565797805786,
      "epoch": 0.5120645863570392,
      "grad_norm": 2.515625,
      "learning_rate": 0.000497995956854074,
      "loss": 5.6811,
      "mean_token_accuracy": 0.15321707502007484,
      "num_tokens": 11735334.0,
      "step": 5645
    },
    {
      "entropy": 5.80997257232666,
      "epoch": 0.5125181422351234,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004979916456273561,
      "loss": 5.6801,
      "mean_token_accuracy": 0.1457176223397255,
      "num_tokens": 11746858.0,
      "step": 5650
    },
    {
      "entropy": 5.898235130310058,
      "epoch": 0.5129716981132075,
      "grad_norm": 3.765625,
      "learning_rate": 0.0004979873297891141,
      "loss": 5.7874,
      "mean_token_accuracy": 0.14425734728574752,
      "num_tokens": 11757819.0,
      "step": 5655
    },
    {
      "entropy": 5.879952812194825,
      "epoch": 0.5134252539912917,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004979830093394372,
      "loss": 5.7022,
      "mean_token_accuracy": 0.1498760312795639,
      "num_tokens": 11767470.0,
      "step": 5660
    },
    {
      "entropy": 5.842562484741211,
      "epoch": 0.513878809869376,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004979786842784149,
      "loss": 5.6223,
      "mean_token_accuracy": 0.15927857905626297,
      "num_tokens": 11777060.0,
      "step": 5665
    },
    {
      "entropy": 5.728598117828369,
      "epoch": 0.5143323657474601,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004979743546061366,
      "loss": 5.6863,
      "mean_token_accuracy": 0.15647987723350526,
      "num_tokens": 11787176.0,
      "step": 5670
    },
    {
      "entropy": 5.639022636413574,
      "epoch": 0.5147859216255443,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004979700203226918,
      "loss": 5.5698,
      "mean_token_accuracy": 0.15824757367372513,
      "num_tokens": 11796419.0,
      "step": 5675
    },
    {
      "entropy": 5.869472932815552,
      "epoch": 0.5152394775036284,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004979656814281702,
      "loss": 5.8015,
      "mean_token_accuracy": 0.14808933883905412,
      "num_tokens": 11807279.0,
      "step": 5680
    },
    {
      "entropy": 5.864350700378418,
      "epoch": 0.5156930333817126,
      "grad_norm": 4.15625,
      "learning_rate": 0.0004979613379226615,
      "loss": 5.6757,
      "mean_token_accuracy": 0.14973624050617218,
      "num_tokens": 11817669.0,
      "step": 5685
    },
    {
      "entropy": 5.7543477535247805,
      "epoch": 0.5161465892597968,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004979569898062554,
      "loss": 5.6793,
      "mean_token_accuracy": 0.15661610066890716,
      "num_tokens": 11827247.0,
      "step": 5690
    },
    {
      "entropy": 5.805063152313233,
      "epoch": 0.516600145137881,
      "grad_norm": 2.0,
      "learning_rate": 0.000497952637079042,
      "loss": 5.7034,
      "mean_token_accuracy": 0.147939595580101,
      "num_tokens": 11838201.0,
      "step": 5695
    },
    {
      "entropy": 5.80039496421814,
      "epoch": 0.5170537010159652,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004979482797411114,
      "loss": 5.6264,
      "mean_token_accuracy": 0.15654842257499696,
      "num_tokens": 11847561.0,
      "step": 5700
    },
    {
      "entropy": 5.81939845085144,
      "epoch": 0.5175072568940493,
      "grad_norm": 2.25,
      "learning_rate": 0.0004979439177925534,
      "loss": 5.7384,
      "mean_token_accuracy": 0.1583888441324234,
      "num_tokens": 11858272.0,
      "step": 5705
    },
    {
      "entropy": 5.923577070236206,
      "epoch": 0.5179608127721336,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004979395512334584,
      "loss": 5.7881,
      "mean_token_accuracy": 0.148820598423481,
      "num_tokens": 11869660.0,
      "step": 5710
    },
    {
      "entropy": 5.893476390838623,
      "epoch": 0.5184143686502177,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004979351800639166,
      "loss": 5.8122,
      "mean_token_accuracy": 0.14301129132509233,
      "num_tokens": 11880239.0,
      "step": 5715
    },
    {
      "entropy": 5.912051820755005,
      "epoch": 0.5188679245283019,
      "grad_norm": 2.125,
      "learning_rate": 0.0004979308042840186,
      "loss": 5.7862,
      "mean_token_accuracy": 0.14405866116285324,
      "num_tokens": 11890630.0,
      "step": 5720
    },
    {
      "entropy": 5.940375518798828,
      "epoch": 0.519321480406386,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004979264238938547,
      "loss": 5.8105,
      "mean_token_accuracy": 0.14483141005039216,
      "num_tokens": 11900948.0,
      "step": 5725
    },
    {
      "entropy": 5.788791179656982,
      "epoch": 0.5197750362844702,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004979220388935155,
      "loss": 5.6941,
      "mean_token_accuracy": 0.15050781145691872,
      "num_tokens": 11912528.0,
      "step": 5730
    },
    {
      "entropy": 5.782585954666137,
      "epoch": 0.5202285921625545,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004979176492830918,
      "loss": 5.6521,
      "mean_token_accuracy": 0.15226173400878906,
      "num_tokens": 11922211.0,
      "step": 5735
    },
    {
      "entropy": 5.8104184627532955,
      "epoch": 0.5206821480406386,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004979132550626742,
      "loss": 5.6858,
      "mean_token_accuracy": 0.14871056675910949,
      "num_tokens": 11932520.0,
      "step": 5740
    },
    {
      "entropy": 5.8572922706604,
      "epoch": 0.5211357039187228,
      "grad_norm": 2.25,
      "learning_rate": 0.0004979088562323538,
      "loss": 5.6209,
      "mean_token_accuracy": 0.15842307060956956,
      "num_tokens": 11943312.0,
      "step": 5745
    },
    {
      "entropy": 5.783070468902588,
      "epoch": 0.5215892597968069,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004979044527922214,
      "loss": 5.7697,
      "mean_token_accuracy": 0.1494937114417553,
      "num_tokens": 11954455.0,
      "step": 5750
    },
    {
      "entropy": 5.808009147644043,
      "epoch": 0.5220428156748912,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004979000447423681,
      "loss": 5.6679,
      "mean_token_accuracy": 0.1576451078057289,
      "num_tokens": 11964650.0,
      "step": 5755
    },
    {
      "entropy": 5.751487350463867,
      "epoch": 0.5224963715529753,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004978956320828851,
      "loss": 5.7023,
      "mean_token_accuracy": 0.15352913439273835,
      "num_tokens": 11975134.0,
      "step": 5760
    },
    {
      "entropy": 5.815647315979004,
      "epoch": 0.5229499274310595,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004978912148138637,
      "loss": 5.7442,
      "mean_token_accuracy": 0.14895366728305817,
      "num_tokens": 11986486.0,
      "step": 5765
    },
    {
      "entropy": 5.89086766242981,
      "epoch": 0.5234034833091437,
      "grad_norm": 2.484375,
      "learning_rate": 0.000497886792935395,
      "loss": 5.7652,
      "mean_token_accuracy": 0.14870162308216095,
      "num_tokens": 11998045.0,
      "step": 5770
    },
    {
      "entropy": 5.9239579200744625,
      "epoch": 0.5238570391872278,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004978823664475707,
      "loss": 5.7574,
      "mean_token_accuracy": 0.15128326863050462,
      "num_tokens": 12008130.0,
      "step": 5775
    },
    {
      "entropy": 5.880002403259278,
      "epoch": 0.5243105950653121,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004978779353504824,
      "loss": 5.8101,
      "mean_token_accuracy": 0.14375324547290802,
      "num_tokens": 12018753.0,
      "step": 5780
    },
    {
      "entropy": 5.836540651321411,
      "epoch": 0.5247641509433962,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004978734996442214,
      "loss": 5.6768,
      "mean_token_accuracy": 0.1561191499233246,
      "num_tokens": 12028202.0,
      "step": 5785
    },
    {
      "entropy": 5.83511757850647,
      "epoch": 0.5252177068214804,
      "grad_norm": 3.0,
      "learning_rate": 0.0004978690593288798,
      "loss": 5.7631,
      "mean_token_accuracy": 0.15131065398454666,
      "num_tokens": 12039652.0,
      "step": 5790
    },
    {
      "entropy": 5.919553232192993,
      "epoch": 0.5256712626995645,
      "grad_norm": 2.984375,
      "learning_rate": 0.0004978646144045491,
      "loss": 5.7,
      "mean_token_accuracy": 0.14130512177944182,
      "num_tokens": 12051269.0,
      "step": 5795
    },
    {
      "entropy": 5.771428203582763,
      "epoch": 0.5261248185776488,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004978601648713215,
      "loss": 5.5623,
      "mean_token_accuracy": 0.15090945810079576,
      "num_tokens": 12060647.0,
      "step": 5800
    },
    {
      "entropy": 5.792234802246094,
      "epoch": 0.526578374455733,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004978557107292889,
      "loss": 5.7274,
      "mean_token_accuracy": 0.1574611634016037,
      "num_tokens": 12071669.0,
      "step": 5805
    },
    {
      "entropy": 5.736810398101807,
      "epoch": 0.5270319303338171,
      "grad_norm": 2.25,
      "learning_rate": 0.0004978512519785433,
      "loss": 5.6046,
      "mean_token_accuracy": 0.15474170595407485,
      "num_tokens": 12081971.0,
      "step": 5810
    },
    {
      "entropy": 5.7692629337310795,
      "epoch": 0.5274854862119013,
      "grad_norm": 2.203125,
      "learning_rate": 0.000497846788619177,
      "loss": 5.7381,
      "mean_token_accuracy": 0.15169834345579147,
      "num_tokens": 12092465.0,
      "step": 5815
    },
    {
      "entropy": 5.852247858047486,
      "epoch": 0.5279390420899854,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004978423206512825,
      "loss": 5.6706,
      "mean_token_accuracy": 0.15314571261405946,
      "num_tokens": 12101752.0,
      "step": 5820
    },
    {
      "entropy": 5.919044303894043,
      "epoch": 0.5283925979680697,
      "grad_norm": 2.296875,
      "learning_rate": 0.000497837848074952,
      "loss": 5.7533,
      "mean_token_accuracy": 0.1472005233168602,
      "num_tokens": 12111805.0,
      "step": 5825
    },
    {
      "entropy": 5.848569869995117,
      "epoch": 0.5288461538461539,
      "grad_norm": 2.171875,
      "learning_rate": 0.000497833370890278,
      "loss": 5.7834,
      "mean_token_accuracy": 0.14824461191892624,
      "num_tokens": 12122511.0,
      "step": 5830
    },
    {
      "entropy": 5.827112245559692,
      "epoch": 0.529299709724238,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004978288890973529,
      "loss": 5.6648,
      "mean_token_accuracy": 0.15470681414008142,
      "num_tokens": 12132510.0,
      "step": 5835
    },
    {
      "entropy": 5.7759528160095215,
      "epoch": 0.5297532656023222,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004978244026962698,
      "loss": 5.6782,
      "mean_token_accuracy": 0.15179212763905525,
      "num_tokens": 12142419.0,
      "step": 5840
    },
    {
      "entropy": 5.80072717666626,
      "epoch": 0.5302068214804064,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004978199116871212,
      "loss": 5.7047,
      "mean_token_accuracy": 0.14955106675624846,
      "num_tokens": 12152759.0,
      "step": 5845
    },
    {
      "entropy": 5.8397301197052,
      "epoch": 0.5306603773584906,
      "grad_norm": 1.8125,
      "learning_rate": 0.00049781541607,
      "loss": 5.6693,
      "mean_token_accuracy": 0.15128355473279953,
      "num_tokens": 12162758.0,
      "step": 5850
    },
    {
      "entropy": 5.860444831848144,
      "epoch": 0.5311139332365747,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004978109158449993,
      "loss": 5.7164,
      "mean_token_accuracy": 0.1495297521352768,
      "num_tokens": 12172664.0,
      "step": 5855
    },
    {
      "entropy": 5.820270347595215,
      "epoch": 0.5315674891146589,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004978064110122121,
      "loss": 5.6847,
      "mean_token_accuracy": 0.15420603305101394,
      "num_tokens": 12182028.0,
      "step": 5860
    },
    {
      "entropy": 5.807129764556885,
      "epoch": 0.532021044992743,
      "grad_norm": 3.328125,
      "learning_rate": 0.0004978019015717314,
      "loss": 5.7133,
      "mean_token_accuracy": 0.15545213222503662,
      "num_tokens": 12192168.0,
      "step": 5865
    },
    {
      "entropy": 5.763472938537598,
      "epoch": 0.5324746008708273,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004977973875236508,
      "loss": 5.6219,
      "mean_token_accuracy": 0.15059156119823455,
      "num_tokens": 12202889.0,
      "step": 5870
    },
    {
      "entropy": 5.8198224067687985,
      "epoch": 0.5329281567489115,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004977928688680634,
      "loss": 5.7018,
      "mean_token_accuracy": 0.15930801928043364,
      "num_tokens": 12212514.0,
      "step": 5875
    },
    {
      "entropy": 5.7587038516998295,
      "epoch": 0.5333817126269956,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004977883456050626,
      "loss": 5.5441,
      "mean_token_accuracy": 0.158404940366745,
      "num_tokens": 12222353.0,
      "step": 5880
    },
    {
      "entropy": 5.8088079452514645,
      "epoch": 0.5338352685050798,
      "grad_norm": 1.8515625,
      "learning_rate": 0.000497783817734742,
      "loss": 5.7913,
      "mean_token_accuracy": 0.15034977793693544,
      "num_tokens": 12232845.0,
      "step": 5885
    },
    {
      "entropy": 5.813466548919678,
      "epoch": 0.534288824383164,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004977792852571954,
      "loss": 5.7517,
      "mean_token_accuracy": 0.1444981053471565,
      "num_tokens": 12242950.0,
      "step": 5890
    },
    {
      "entropy": 5.866303157806397,
      "epoch": 0.5347423802612482,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004977747481725164,
      "loss": 5.7329,
      "mean_token_accuracy": 0.14827523976564408,
      "num_tokens": 12253132.0,
      "step": 5895
    },
    {
      "entropy": 5.847100877761841,
      "epoch": 0.5351959361393324,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004977702064807987,
      "loss": 5.6634,
      "mean_token_accuracy": 0.15220373421907424,
      "num_tokens": 12263214.0,
      "step": 5900
    },
    {
      "entropy": 5.721499347686768,
      "epoch": 0.5356494920174165,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004977656601821366,
      "loss": 5.6111,
      "mean_token_accuracy": 0.16137520372867584,
      "num_tokens": 12273250.0,
      "step": 5905
    },
    {
      "entropy": 5.920711851119995,
      "epoch": 0.5361030478955007,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004977611092766238,
      "loss": 5.7896,
      "mean_token_accuracy": 0.15604286193847655,
      "num_tokens": 12283767.0,
      "step": 5910
    },
    {
      "entropy": 5.821226930618286,
      "epoch": 0.5365566037735849,
      "grad_norm": 4.5625,
      "learning_rate": 0.0004977565537643544,
      "loss": 5.6478,
      "mean_token_accuracy": 0.14899064600467682,
      "num_tokens": 12294612.0,
      "step": 5915
    },
    {
      "entropy": 5.75635838508606,
      "epoch": 0.5370101596516691,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004977519936454227,
      "loss": 5.7324,
      "mean_token_accuracy": 0.15379261076450348,
      "num_tokens": 12305614.0,
      "step": 5920
    },
    {
      "entropy": 5.7148298740386965,
      "epoch": 0.5374637155297532,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004977474289199231,
      "loss": 5.6016,
      "mean_token_accuracy": 0.14994118511676788,
      "num_tokens": 12316152.0,
      "step": 5925
    },
    {
      "entropy": 5.771187782287598,
      "epoch": 0.5379172714078374,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004977428595879499,
      "loss": 5.6922,
      "mean_token_accuracy": 0.156583671271801,
      "num_tokens": 12325940.0,
      "step": 5930
    },
    {
      "entropy": 5.850227928161621,
      "epoch": 0.5383708272859217,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004977382856495975,
      "loss": 5.6883,
      "mean_token_accuracy": 0.15705590099096298,
      "num_tokens": 12335482.0,
      "step": 5935
    },
    {
      "entropy": 5.897109460830689,
      "epoch": 0.5388243831640058,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004977337071049606,
      "loss": 5.7659,
      "mean_token_accuracy": 0.14804613441228867,
      "num_tokens": 12346386.0,
      "step": 5940
    },
    {
      "entropy": 5.881980228424072,
      "epoch": 0.53927793904209,
      "grad_norm": 1.859375,
      "learning_rate": 0.000497729123954134,
      "loss": 5.7127,
      "mean_token_accuracy": 0.15534048080444335,
      "num_tokens": 12356282.0,
      "step": 5945
    },
    {
      "entropy": 5.8060948848724365,
      "epoch": 0.5397314949201741,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004977245361972123,
      "loss": 5.7073,
      "mean_token_accuracy": 0.15651538074016572,
      "num_tokens": 12366994.0,
      "step": 5950
    },
    {
      "entropy": 5.762063932418823,
      "epoch": 0.5401850507982583,
      "grad_norm": 3.984375,
      "learning_rate": 0.0004977199438342904,
      "loss": 5.5904,
      "mean_token_accuracy": 0.15490693598985672,
      "num_tokens": 12377524.0,
      "step": 5955
    },
    {
      "entropy": 5.752496385574341,
      "epoch": 0.5406386066763426,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004977153468654634,
      "loss": 5.7448,
      "mean_token_accuracy": 0.15410600006580352,
      "num_tokens": 12387593.0,
      "step": 5960
    },
    {
      "entropy": 5.861761140823364,
      "epoch": 0.5410921625544267,
      "grad_norm": 2.875,
      "learning_rate": 0.0004977107452908262,
      "loss": 5.7276,
      "mean_token_accuracy": 0.14881233870983124,
      "num_tokens": 12398732.0,
      "step": 5965
    },
    {
      "entropy": 5.880192470550537,
      "epoch": 0.5415457184325109,
      "grad_norm": 2.578125,
      "learning_rate": 0.000497706139110474,
      "loss": 5.7008,
      "mean_token_accuracy": 0.15882584676146508,
      "num_tokens": 12409821.0,
      "step": 5970
    },
    {
      "entropy": 5.842268705368042,
      "epoch": 0.541999274310595,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004977015283245021,
      "loss": 5.7328,
      "mean_token_accuracy": 0.14897113144397736,
      "num_tokens": 12420044.0,
      "step": 5975
    },
    {
      "entropy": 5.760256767272949,
      "epoch": 0.5424528301886793,
      "grad_norm": 3.03125,
      "learning_rate": 0.0004976969129330059,
      "loss": 5.6512,
      "mean_token_accuracy": 0.15109718292951585,
      "num_tokens": 12430980.0,
      "step": 5980
    },
    {
      "entropy": 5.786817407608032,
      "epoch": 0.5429063860667634,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004976922929360808,
      "loss": 5.7144,
      "mean_token_accuracy": 0.15620118379592896,
      "num_tokens": 12441999.0,
      "step": 5985
    },
    {
      "entropy": 5.855512714385986,
      "epoch": 0.5433599419448476,
      "grad_norm": 3.65625,
      "learning_rate": 0.0004976876683338222,
      "loss": 5.7487,
      "mean_token_accuracy": 0.14880464673042298,
      "num_tokens": 12452837.0,
      "step": 5990
    },
    {
      "entropy": 5.810596656799317,
      "epoch": 0.5438134978229318,
      "grad_norm": 2.421875,
      "learning_rate": 0.000497683039126326,
      "loss": 5.666,
      "mean_token_accuracy": 0.1583888977766037,
      "num_tokens": 12462142.0,
      "step": 5995
    },
    {
      "entropy": 5.80487813949585,
      "epoch": 0.5442670537010159,
      "grad_norm": 2.375,
      "learning_rate": 0.0004976784053136878,
      "loss": 5.706,
      "mean_token_accuracy": 0.15387148559093475,
      "num_tokens": 12472756.0,
      "step": 6000
    },
    {
      "epoch": 0.5442670537010159,
      "eval_entropy": 5.610305411999042,
      "eval_loss": 5.737067699432373,
      "eval_mean_token_accuracy": 0.15633446278547808,
      "eval_num_tokens": 12472756.0,
      "eval_runtime": 26.751,
      "eval_samples_per_second": 1321.818,
      "eval_steps_per_second": 165.227,
      "step": 6000
    }
  ],
  "logging_steps": 5,
  "max_steps": 110230,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 3000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.8256294637568e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}